Re: Has z/Journal gone to the dark side
There is such a site with information on moving to z/OS, Linux on System z, and other mainframe operating systems: Destination z. You can start here: http://www.ibm.com/systems/z/destinationz - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: C++ Workable Mainframe Debuggers (and a Public z/OS HTTP Server)
Miklos Szigetvari writes: We are struggling with the debugger for several years The main problems: - very very slow - unable to handle properly multithread application - abends in various places etc etc By the way we had a common project ( a PMR) a few years ago with the Toronto Lab, we sent and received the source code, and in the source code received from them, we seen a number of printf's , debug macro's etc etc We are in z/OS 1.8 and we had WD4Z (Web Sphere) That doesn't suprise me at all for a few reasons. The first reason is the few years ago part. I think it was only late 2006 (Version 7) when there were some feature improvements for C++ added that a lot of developers demanded. The second reason is that IBM cannot anticipate exactly which debugger you have (if any), so if we're working collaboratively we may have to fall back on printfs. I know when I was in support I'd often start with the most basic troubleshooting inspection process, often thinking at the same time that it would be so darn useful if my customer had tool X or utility Y installed (and if I had remote access). Sometimes I would remember to ask, but there's an overwhelming tendency for support-type people to walk you through the universal approach by default, which in this case is printf. The third reason is that you can still use printfs and the like, so even programmers armed with debuggers they love still use both techniques. The fourth reason is there are many production scenarios where recording log files is essential, so you might be putting that stuff in your code anyway. Especially in the mainframe world where more people worry about deploying highly supportable code. We got now a new z9, we will try again, but till now , we were unable to find any real error with the help of the debugger The hello word program, and its variants are working with the debugger, but a real complex application with DLL's , multithread not. OK, please get whatever you find back to IBM. If the problems you describe are still going on, it sounds like IBM has a lot of gap to close to meet your needs. Make sure you challenge my colleages and give them a shot to do better. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: C++ Workable Mainframe Debuggers
Dave Cole writes: We are on the verge of publishing C Source code support (c/XDC) to beta testers. That's excellent. Thanks, Dave, for getting me more up-to-date. Side note: Another dividing line for the debuggers seems to be whether they support 64-bit code or not. z/XDC and IBM Debug Tool do, to pick two examples. So I would advise the original poster to check that aspect also. It's quite likely the C++ programmers will want 64-bit support -- if not now, then soon. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Specialty Engine Prices (Was: Another RNL question)
John McKown writes: Hum, then I wonder why when we went from a z890 to a z9BC, I lost my IFL? No, I'm not going to complain. I don't much care anymore. Did someone tell IBM to turn it off? That is a choice. The only reason I can think of off the top of my head that someone might do that is that the IFL has a (very small) hardware maintenance charge associated with it after the warranty period. If it's totally idle then someone might object to the maintenance charge, although that probably doesn't explain why it went missing during the warranty period. But you might still have rights to an IFL, John, so I'd dig around a bit to figure out what happened to it and maybe have a chat with someone on why they're turning off hardware you already own and paying for, say, another 100 distributed servers that you don't. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Faxing from the mainframe
Shmuel Metz writes: No; OV/VM may be PROFS, but OV/MVS is quite different. Perhaps it's more precise to say that OV/MVS is, by far, the closest current match to PROFS. Is that fair? - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Faxing from the mainframe
It is possible to get reliable e-mail-to-fax service. For example, you may wish to set up a private IPSec SMTP tunnel between your mainframe and your e-mail-to-fax service provider. Then you are connecting directly and over a private link, and you can have reasonable confidence in the transmission. Your e-mail-to-fax service provider should also have various means of generating transmission reports to indicate who got what (and when). Or, more precisely, that some fax machine answered and the paper is perhaps curled up under the filing cabinet. :-) - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
IBM PR: Rational Business Developer Trial Download Available
To follow up on the test drive information, if you're particularly interested in Rational Enterprise Generation Language (EGL) and would like to try the developer workbench, you can obtain a no charge trial download of Rational Business Developer here: http://www.ibm.com/developerworks/downloads/r/rbex/?S_TACT=105AGX15S_CMP=LP You can run RBD on either your Linux or your Windows PC, although the VisualAge Generator migration tool requires a Windows-based PC. It's a 60-day trial, but you do not have to reinstall if you end up buying a license: just add the license key to your trial installation. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Faxing from the mainframe
PROFS is still around, although the new name (and new version) is now OfficeVision. You can get more details on OfficeVision/MVS here: http://www.ibm.com/software/applications/office/officevision/index.html Yes, it's available for ordering. The IBM product number is 5685-106. Re: Faxing, I think I'd echo what everyone else says about e-mailing. For the occasional fax you can purchase e-mail-to-faxing service (over the Internet) from any of several companies, then just use the same e-mail path from your mainframe. PDF would be a good format to e-mail. However, I do see something in the IBM catalog for a z/OS fax gateway. It's listed as IBM product number 5620-FIE, and it's called MessagePlus/Open for z/OS from INTERSCOPE. There's some more information about that product here: http://www.ibm.com/software/data/cm/solutions_mpo.html - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Archiving open systems data on WORM via MF
likely to lose more data in the event there is a distributed server failure, because there will be a greater time gap between your last copy (copied to WORM) and the data residing on your distributed server's own hard disk compared to the live access approaches in 1(a) and 1(b). 2(b). Similar to 2(a), you can transfer files to the mainframe using scheduled FTP. (z/OS includes an FTP server.) SFTP and SSH are other possibilities. The drawbacks are similar. 3(a). I assume you are referring to TSM for z/OS and expressing some reluctance about that particular product. And I'm guessing that it's a reflexive price reaction to that product, as in, your boss told you to solve this problem without spending any money. (I want to date a supermodel, but I don't want to pay for dinner. :-)) Let's explore that for a moment. If you plan to use TSM for z/OS exclusively for purposes of backing up distributed server files or databases to WORM tapes, you may wish to review the IBM zNALC announcement, complete a zNALC questionnaire, and send that questionnaire to IBM. I can make no promises at all here, but you may be able to create a zNALC z/OS LPAR -- probably a very small one, as little as 3 MSUs -- exclusively for TSM for z/OS and then run your backups through that LPAR from TSM clients installed on your distributed servers. Ask your favorite IBM representative to see if that path works for you and if IBM will approve it. (As a reminder, I do not speak for IBM, at least not today.) 3(b). A slight modification to 3(a) is to choose TSM for Linux on z and implement Linux on z. This is probably getting less simple if you don't already have Linux on z (and do have z/OS) since it would be the first time you would implement Linux on z, but that's another possible option. I believe you have to dedicate at least one particular physical tape drive to TSM for Linux on z's exclusive use, so sharing the tape drives between Linux and z/OS from hour to hour becomes a little more interesting (and less simple). But TSM EE for Linux on z does support most mainframe-attached tape drives. There are probably other solution options, but these are the 6 major options that come to mind immediately, if my assumptions are correct about your requirements. I hope this information is helpful. Please keep us all updated on your progress. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
IBM PR: Enterprise Modernization Sandbox Now Available
For those of you who would like to take a test drive of IBM's enterprise modernization tools including Rational Developer for System z (RDz), Rational Transformation Workbench (RTW), and Rational Business Developer (RBD), please point your Web browser here to get more information: http://www.ibm.com/developerworks/downloads/emsandbox/systemz.html There's a lot of good information here which I also recommend exploring, but please click on the Skills link (third bullet) for access to the RDz/RTW/RBD test drive. This test drive provides remote access to a live z/OS system and a pre-configured desktop environment for writing, compiling, and debugging code using the modern RDz graphical workbench. You can try C, C++, Assembler, COBOL, PL/I, Java, REXX, JCL, etc. You can use RTW to gain insight into code and subsystems in order to locate service entry points and modernize the code in various ways, such as changing data elements to Unicode, pruning redundant code, etc. RBD lets you write, compile, and debug Rational Enterprise Generation Language (EGL) and deploy EGL to the mainframe. This 4th generation programming language is easy to learn and highly productive. It's also a great target language for migration from older 4GLs, such as IBM's CSP and VisualAge Generator, and from non-IBM 4GLs. There's a 3 (or maybe 4) hour time limit, but you can take the test drive as often as you want (within reason). Please save your work if you'd like to pick up where you left off as you start a second test drive, per the instructions. There is no formal support for this test drive, and it is subject to change at any time. I hope everyone finds it educational and useful. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
IBM PR: WebSphere MQ V7 for z/OS Announced
IBM just announced WebSphere MQ V7 for z/OS. There are a lot of enhancements, but some of the major ones include new integrated publish/subscribe support, improved performance (especially for JMS), new MQI calls and behaviors for easier programming, MQ client performance and availability enhancements, and Web 2.0 support. The announcement letter is available here: http://www.ibm.com/common/ssi/rep_ca/7/897/ENUS208-067/ENUS208067.PDF WebSphere MQ is the ubiquitous standard for reliable messaging and an extremely important foundation for service-oriented architectures (SOA). WMQ helps developers avoid costly and time-consuming communications-related programming to assure one and only once delivery of messages between applications, application services, and platforms. WMQ is available on more platforms than perhaps any other middleware. On z/OS there are unique capabilities, such as exploitation of the Coupling Facility for shared queue support to provide the highest levels of availability. (Note: While you need a Coupling Facility and DB2 to use shared queues, you do not need either to use other features in WMQ.) WebSphere MQ V7 for z/OS should be generally available at the end of June, 2008. The WebSphere MQ homepage has been updated with a lot more information about Version 7, such as details about other MQ platforms including Linux on System z. Check here: http://www.ibm.com/software/integration/wmq Over the next few days there will be some more WMQ V7 information appearing, so please check back often. ABOUT LICENSING: WebSphere MQ for z/OS is a Monthly License Charge (MLC) software product for z/OS. WMQ for z/OS is available for sub-capacity Variable Workload License Charge (VWLC) licensing, and it produces its own SMF records for individual charging. That means, with VWLC, you only pay for MQ according to the MQ MSUs you use, and only for the LPARs where MQ is installed. You can license as few as 3 MSUs. If you wish to attach MQ clients to a WMQ for z/OS queue manager, you should also license the Client Attachment Feature (CAF). However, if you use either the MQ Explorer (to manage z/OS parts of your MQ network) or the WebSphere Message Broker Configuration Manger (to manage and configure WebSphere Message Broker for z/OS), you can use up to five of those client connections (using specific channel names) without buying the CAF. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
IBM PR: WebSphere MQ V7 for z/OS Open Beta Program
...And I did say check back often for more information on WebSphere MQ V7 for z/OS, so here's more information. IBM has announced an open beta program. That means z/OS customers may sign up for and download pre-release WebSphere MQ V7 for z/OS software, to test in their own environments. You do not need to be a current WebSphere MQ for z/OS customer to participate. You will have to discontinue use of the pre-release version soon after WMQ V7 for z/OS becomes generally available. For more information on the open beta program, and a link to download the pre-release software which should be active on April 4, 2008, please visit here: https://www14.software.ibm.com/iwm/web/cc/earlyprograms/websphere/wmq7open You will need to have an LPAR running z/OS 1.8 or higher to install and use WebSphere MQ V7 for z/OS. Documentation is available now if you'd like to start planning your installation: http://publib.boulder.ibm.com/infocenter/wmqv7/v7r0/index.jsp - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: CICS Installation
Mark, You may want to join the CICS-L mailing list, too. You may see references to CICSPlex System Manager (CPSM). Please don't get too confused. CPSM still has value even if you only have a single CICS Transaction Server system. CPSM does not require Parallel Sysplex. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: CSP/AD and zOS 1.8
Others have provided some good answers to the short-term unsupported fix. The supported fix is to get thee updated to Rational Enterprise Generation Language (EGL). There are migration guides located here (scroll down to the bottom): http://www.ibm.com/developerworks/rational/products/egl/egldoc.html The last link on that page (Cross System Product v4.1 Migration to EGL) is the most relevant in your case, so please start there. Migration is a two-step procedure, first migrating to VisualAge Generator (last version) and then to Rational EGL. Ultimately on the mainframe you'd likely license and install Rational COBOL Runtime for zSeries (5655-R29) in lieu of CSP. (You can also generate Java from EGL and deploy to other runtimes such as WebSphere Application Server for z/OS.) On the development client side you'd use Rational Business Developer (5724-S50). Hope that helps. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: IBM C/C++ compiler cost?
I might have said something incorrect on this topic earlier, although I don't think it applies to John's situation. Parallel Sysplex pricing is better than I implied previously. If you've got Parallel Sysplex, and you only license the C/C++ compiler to one physical machine, then you'll only be charged on that machine (based on total regular z/OS MSUs or total zNALC z/OS MSUs, depending on which side of that fence you've installed the compiler). For some reason I was thinking of DFSMStvs (another z/OS element) which, presumably, you'd be installing on both machines in a Parallel Sysplex for functional reasons. I think I implied that both machines would get charged if you only have the compiler on one, and that's not correct. Sorry for the confusion. As a tip, if you don't put the C/C++ compiler on both machines you'll likely want to put the compiler on the smaller machine in a Parallel Sysplex. Smaller is defined as the machine which typically clocks lower total z/OS MSUs (regular or zNALC, as appropriate) each month. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z9 over-heating prevention
One of the more interesting attributes is that the machine will do everything possible to keep your work moving as best it can even if there's a double MRU failure. It can slow down the processors to reduce heat. WLM (assuming z/OS) is still in control, so heat-related forced slowdowns will tend to disfavor work in lower service classes, starting with purely discretionary workloads. You're probably already monitoring how well WLM is meeting its goals, so that's one (indirect) way to get a warning that something unusual is happening. I think an MRU failure is a Call Home event, by the way. A technician is going to initiate a telephone call to schedule installation of a replacement part, if necessary. If you're visiting IBM (in Poughkeepsie, for example), they may be able to simulate a double MRU failure for you so you can see what happens, at least if you request that demonstration ahead of time. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: IBM C/C++ compiler cost?
John, One major factor that will influence the price is whether the C/C++ compiler is fenced to a zNALC LPAR or not. I'll sketch out for you the two scenarios so you know the right question(s) to ask your IBM dude or dudette. In the usual scenario, the price depends on your total z/OS MSUs on your machine (or in your Sysplex if you have aggregated pricing). For example, let's assume you have no zNALC LPARs (and no NALC and no z/OS.e, the older offerings) and you have five normal z/OS LPARs. You install and run the C/C++ compiler in LPARs 2 and 3. You have a single machine. In that case your C/C++ compiler price will be regular price, and it will be equal to the number of MSUs your sub-capacity report shows for z/OS across all 5 LPARs. (The C/C++ compiler is not separately reported. It is considered an element of the operating system.) I'm assuming sub-capacity VWLC here, and that you're submitting SCRT reports. Thus, if you set an LPAR group capacity limit (now available in the newer z/OSes), your total C/C++ price cannot exceed that group limit. It also cannot exceed the sum total of individual LPAR limits. Now let's suppose you have 5 usual z/OS LPARs and 2 zNALC z/OS LPARs. You install and run the C/C++ compiler in 1 zNALC LPAR but not in any of the usual z/OS LPARs. You still have a single machine. In that case your C/C++ compiler price will be the zNALC price -- it's an operating system element, so it gets that benefit, too. And the price will be equal to the number of MSUs your sub-capacity report shows for the total of your zNALC LPARs. The Sysplex examples aren't really more complicated. Just insert qualifying aggregated Sysplex everywhere it says single machine above. I think the minimum number of MSUs is 3 in any of these cases -- unless it's zero, of course. So, to oversimplify a bit, the zNALC domain is treated much like a separate machine. If the workload qualifies, you can license software separately to that zNALC domain, even on a single machine (or single Sysplex), and it will be treated separately even if it is something like a base operating system element that ordinarily wouldn't be. The same is true in reverse: products licensed to the regular z/OS LPARs aren't counted over on the zNALC side if they're not licensed to the zNALC LPARs. Net net, you'll want to read the zNALC announcement to see if you qualify -- and if your use of the C/C++ compiler would qualify, in particular. For example, if you're porting code currently running on distributed UNIX to your mainframe using that compiler to do it, then you just might. Or if you're considering buying a new vendor application, and that application is capable of running on distributed UNIX or Windows, but you'd much rather run that application on your mainframe on z/OS, and you need a C/C++ compiler to support the application, then you might qualify. If you think you might qualify, send IBM a completed zNALC questionnaire to find out for sure. If you do qualify, you'd then consider whether it makes sense to carve out another LPAR, which has its own pros and cons. It may or may not make sense to do that just for that one compiler, especially if you have a small machine. OK, now, to answer your question directly (since I've done it before publicly, so water under the bridge I guess): U.S. commercial use pricing for 3 MSUs (the minimum) of the IBM C/C++ compiler, with or without Debug, on zNALC LPAR(s) is ... $6 per month. So the marketing people might say, As low as 19.4 cents per day !!! Order NOW, and we'll also include 3 MSUs of Debug Tool, absolutely free(*)! BUT THAT'S NOT ALL Please be aware that that number could be a LOT different -- even orders of magnitude different -- depending on your circumstances. Also, that doesn't include the base zNALC z/OS charge, although your mileage varies a lot on how you'd fairly divide that. (*) Actually, free means no additional charge. The lawyers make us say that, because stuff like your time is presumably worth more than zero. When they say free, they really mean free. And they never say free. Even their pro bono work is no additional charge. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
NY Times: Why Old Technologies Are Still Kicking
The New York Times has a short but fairly interesting article about old technologies which uses the mainframe as an illustration. Here's the link: http://www.nytimes.com/2008/03/23/technology/23digi.html - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Changing the MF IP Address
Would you expand a bit on why you think you have to change the IP address? I didn't really understand the firewall comment. I think I agree with others that it's more graceful to add a second IP address first and then monitor any use of the old IP address. If anyone uses the old address, you can trace it back and notify that user that a loss of service is imminent without an address change. Run double for as long as you need, then retire the old address when you feel comfortable. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z/OS and Linux on same z/VM Image
John McKown writes: We have, if I remember, one vendor product which is based on the MSU rating of the system. Yes, that would be the exception. In fairness there are a few IBM products charged according to full capacity, but all the ones I've run into are products that predated (by a lot) the introduction of VWLC. And in many of those cases they're withdrawn products. If you do have software charged by full capacity MSUs, then that might be a factor, but of course you'd have to weigh that together with all the other factors. YMMV. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Assembler to C or C++ Conversion
Perhaps a dumb question, but what kind of Assembler? (Can we assume 390?) And what does it do? Does it run with any middleware involved? - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z/OS and Linux on same z/VM Image
John McKown writes: ...the CPs running z/VM and z/Linux still cause the z/OS software prices to go up (in many cases). Assuming VWLC, I'm trying to figure out why that would be true. I could see how running some Linux workload on CPs might cause z/OS software charges to *decrease*, though. Mark Post writes: My understanding was that within a particular z/VM guest, you would still have the restriction of not mixing IFLs and other processor types. I don't think IBM's Statement of Direction gets into that level of detail, so we'll have to see how things evolve. But, even if so, that wouldn't be a particularly onerous restriction IMHO. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z/OS 1.4 Running on 9672 CPU
Jason, For reference, your 9672-Z87 has a full capacity software rating of 217 MSUs. That's a relatively large machine, and I would strongly recommend running the financial numbers immediately on getting to a 64-bit machine. Every month your employer is paying full capacity (217 MSUs) for z/OS and all other software. This is especially problematic if you have more than one LPAR and could be running middleware selectively by LPAR. For example, if you only need a little bit of IMS, you're currently paying full capacity pricing for IMS instead of paying a price for just a smaller LPAR's worth. You're also missing out on as many as 3 technology dividends, which are MSU reductions for equivalent capacities. The good news is the migration shouldn't be particularly difficult and doesn't involve a lot of steps. You have some not-so-difficult routes available which I will sketch out for you. Please talk to somebody (like IBM) about it ASAP. ** Path #1 ** 1. You could move to a 2094 model (System z9 EC) -- possibly a used machine if it makes sense and if it is available -- and keep your operating system and middleware the same initially, running z/OS 1.4 in Bimodal Accommodation mode (ESA 31-bit mode) for up to 12 months if necessary. Since there are no software changes (except maybe a z/OS PTF or two), you can accomplish this hardware-only migration very quickly indeed. You'll still be charged full capacity as long as you run in 31-bit mode, but you could get, for example, a 2094-408 model which should be about the same performance as your current machine yet only require 185 MSUs of software. I think the 408 would also have the same number of engines as your current machine, so if you're happy with that then there's not too much thinking involved. 2. Start sending in SCRT reports. Get your z/OS 1.4 IPLed into 64-bit mode. When you do that for every LPAR, and once IBM has a couple months of SCRT reports from you, you can start enjoying sub-capacity VWLC pricing for your software. You will also enjoy storage relief: the 2094 comes standard with at least 16 GB, and you can assign all of it (except HSA) to z/OS with 64-bit memory addressing. 3. Migrate to z/OS 1.7 (and the latest middleware product releases), which you can do with coexistence and fallback on the same machine in new LPARs. 4. Consider adding zIIPs or zAAPs at this point if your workloads could benefit. (IFLs may also be useful.) 5. Once you've migrated to z/OS 1.7, consider upgrading the machine to a 2097 (System z10 EC), likely a 408 model again. The 408 model will have more capacity than your current 9672 machine, but it will drop the full capacity MSUs again down to 180. You can perform this upgrade and keep the same serial number as the System z9 EC if you wish. 6. Keep staying current, with your next step up to z/OS 1.9 (now available). ** Path #2 ** 1. If you're not growing very fast, you could move directly to a 2096 (System z9 BC) machine. Your current machine is up near the top of the range of capacities for the z9 BC, but it does fit. For example, a Z03 model will probably give you just a little more capacity but have a full capacity software rating of only 189 MSUs. I believe you can run z/OS 1.4 Bimodal Accommodation mode on this machine. This machine will have fewer engines, however, so you'll want to make sure it makes sense for your workloads. 2. Same as step 2 above, except that the 2096 arrives with a minimum of 8 GB of memory. 3. Same as step 3 above. 4. Same as step 4 above. 5. Same as step 6 above (skip step 5 above). In short, in addition to the out-of-support concerns, the financials probably don't make any sense for you to keep the 9672 machine. Let somebody run the numbers (including an ELA I'd recommend) and please get moving, now. Good luck. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: DB2 for z/OS V7.1 on z/OS V1.9
Yes, you can run DB2 V7 on z/OS 1.9. However, support will end on June 30, 2008, for DB2 V7 for z/OS. End of service for z/OS 1.7 is September 30, 2008. I don't know your exact situation, but just going by those dates you might want to tackle migration to DB2 V8 first, then jump from z/OS 1.7 to 1.9, then come back and get to DB2 9. (I'm assuming you're moving from z/OS 1.7 to 1.9.) I hestitate to make a blanket recommendation on that -- your mileage (and likely schedule) may vary -- but as a first order guess that's what I'd think about. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z/OS 1.4 Running on 9672 CPU
I assumed the original poster's employer had done the right thing and ordered z/OS 1.7 at some point in time. Perhaps that's a bad assumption. Regardless, I would strongly recommend he ring IBM as soon as possible. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z/OS and Linux on same z/VM Image
John Giltner wrote: A single LPAR can't have a IFL and a CP. On the System z10, this restriction is removed by using something new called a z/VM-mode LPAR (a.k.a. z/VM Image Mode LPAR). Such an LPAR can contain almost any mix of CPs, IFLs, zIIPs, zAAPs, and/or ICFs and must run z/VM as the base operating system. You'll need HMC and SE Version 2.10.0 or higher to configure this mode. This hardware support was announced with the System z10 EC on February 26, 2008, and on the same day IBM published a Statement of Direction concerning a future release of z/VM which (IBM expects) will support this new LPAR mode. (Statements of Direction are subject to change.) I know a lot of people will appreciate having this new capability. It should greatly simplify configurations in many cases, and it also helps take a lot of pressure off pre-planning. It also should allow you to do some interesting new things, like use spare CP capacity more easily to supplement IFL capacity on demand. I guess you could say that the IFLs become a lot more like zAAPs and zIIPs, since you could view them as supplementing the CP capacity pool for Linux workloads rather than as a completely separate pool. And that further increases the cost- and performance-efficiency of the mainframe. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Tapeless backup
As another difference, there are WORM tape cartridges available which are quite unlike disk storage. Some industries require WORM capabilities. Also, TS1120 drives can encrypt at the controller (in the drive), with keys managed by z/OS's ICSF. That functionality is becoming increasingly important, and it's quite effective to encrypt at the drive. I agree with the other commenter that nothing really beats tape for off-site long-term vaulting on-the-cheap. It depends what your RPO requirement is, but unless you invest in a long distance mirroring solution -- GDPS for example -- a single site catastrophe would also wipe out all your data. So you have to look at the whole picture, and in many situations you're comparing tape (times 1 with cartridge shipment) to long-distance mirroring (disk times 2 plus the interconnects). - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Cobol Java Ldap and between
Rob Schramm asks: I haven't looked this up... but is the license for the IBM Metal C more attractive to the customers that may have balked at the IBM C/C++ compiler? Metal C is both a new compiler option and a runtime library found in the base z/OS. You still have to license the IBM C/C++ compiler to use it. However, the C/C++ compiler is surprisingly affordable, especially if you are eligible to license it for a zNALC LPAR. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: IBM Tivoli Monitoring (ITM for short)
For a 100% mainframe-hosted solution, you can run TEMS on z/OS and TEPS on Linux on z. That arrangement supports z/OS agents (the OMEGAMON XE elements), Linux on z agents (OMEGAMON XE on z/VM and ITM elements), and/or distributed agents (ITM elements). There is no requirement for yet another box, and there are no extra license charges. If you'd like TEPS also on z/OS, please ask your friendly IBM Software representative. Please do remember that TEPS derives from the CandleNet Portal Server which (I believe) wasn't offered on z/OS, but IBM may have a different viewpoint now that Candle is part of IBM. In fact, I don't think CandleNet Portal ran on Linux on z either, so that happened after the acquisition. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Cobol Java Ldap and between
And to answer the other half of Magen Nargalit's question: If a c/c++ connector s required and I don't have c/c++ compiler for z/os can I compile on another platform and port the object / Module to z/os environment? Yes, of course. You do not need the IBM C/C++ compiler installed in your production z/OS LPAR. As long as you have a valid compiler license somewhere -- on another machine, say -- and compile there, you're fine. I'm referring to the IBM compiler here, but Dignus and SAS also have compilers. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: IBM Tivoli Monitoring (ITM for short)
Bruce: Somebody will jump in and correct me if I'm incorrect, I'm sure, but were you aware that the Tivoli Enterprise Portal (TEP) license (and code) is already included with any of the Tivoli OMEGAMON XE products, including the no charge OMEGAMON XE Management Console for z/OS? I wasn't sure since you were talking about paying license fees. Also, you may install TEP on Linux on System z if you'd like a 100% mainframe-hosted solution. (TEMS can run on z/OS.) Would that be an option for you? You'll still need a PC for the graphical client, but IBM has made that compatible with both Linux and Windows now. So you don't even need a Microsoft Windows license if that's an obstacle. Hope that helps. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Tool that will help to access the Teradata tables from the mainframe job
Rajeev, I think we'll need a little more information to answer your question well. What sort of mainframe job is it? Let's start with programming language if you know. (Do you have the source code, or is it code somebody else supplied?) Also, where's it running? Is it a batch job, or a CICS program, or an IMS program, or a DB2 stored procedure, or something else? What operating system? (I guess a lot of people assume z/OS, but it could be any of at least four others.) And how much data do you think you'll access? How often? Will you only be reading it, or will you also be updating? And do you need live data access or would a copy suffice? Any particular security requirements, or is this an open database and it doesn't matter? Do you have any particular preferences we should know about? I'll assume lowest (full) cost is one. :-) There are a lot of options, but knowing those parameters should help narrow down the range of choices. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
The System z10s Are Shipping
In an earlier post I said something like the System z10s are apparently shipping. I've just found out I can strike the word apparently. They're shipping. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: IBM announced z10 ..why so fast...any problem on z 9
I agree, the fixed, reserved HSA is great. Much simpler, much easier to plan -- even if you can't plan. Tommy, your System z9 will keep operating just fine. There are many reasons why z9 (and previous) model mainframe owners upgrade, though, and it really doesn't matter how recently they upgraded to the z9. One big reason not mentioned yet is software: the new machine gets more software price-performance, more software value. It's worth looking at the total financial picture and deciding whether to order the z10 upgrade. For some customers it's an immediate order because every month without a z10 is less financially attractive. But situations vary. You can upgrade a System z9 EC to a System z10 EC and keep your machine's serial number. You can order those upgrades now, and they are apparently shipping now. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: System z10 announcement (in English)
If it's Microsoft .NET stuff it should be possible to run it on System z either using Mono (http://www.mono-project.com) (on Linux on z, for example SLES 9) or using software from Mainsoft (http://www.mainsoft.com/solutions/serverconsolidation.aspx) (on WebSphere Application Server for z/OS or for Linux on z). Ideally you would also begin a transition away from .NET development at the same time, but you don't have to throw it away cold turkey. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
PR: IBM Announces the System z10 for the Next Generation Data Center
Good (late) evening to everyone in the U.S. It's early afternoon on Tuesday here in Japan. If you read Japanese (or use an online translator), then you can get a headstart on the big news here: http://www.ibm.com/jp/press/2008/02/2601.html For those of you who prefer English, please watch here: http://www.ibm.com/press and you should see a press release at a more reasonable hour on Tuesday, New York time. Much more news and many more details to follow, I'm sure. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Is IBM/Tivoli turning into CA?
Doug Fuerst writes: All I can say is I have 4.1 or 550 for z/OS and there is not one shred of USS support in the product that I can find. I was told that if I wanted that support I needed Omegamon for USS. So if this is the case, even the support people and the marketing people are unaware of it. I actually use Sysview from CA for my USS monitoring. Omegamon has no idea if I have an HFS or where it is. So if its there, I'd love to know where. I'm sorry you got that information. IBM tried to provide a lot of notice in the announcement letter, New in this release info, product documentation, etc. Here's the Planning and Configuration Guide (watch the wrap): http://publib.boulder.ibm.com/infocenter/tivihelp/v15r1/topic/com.ibm.omegamon_xezos.doc/km5cfmst.pdf and here's the User's Guide: http://publib.boulder.ibm.com/infocenter/tivihelp/v15r1/topic/com.ibm.omegamon_xezos.doc/km5usmst.pdf Try searching in those documents for UNIX System Services to see if that's what you need. Please advise if it's not. Rex Pommier writes: Usually I am in agreement with most of what you say, as I am partly with what you say below, in that IBM has done some good things with Omegamon since acquiring it. But, that being said, I must strongly disagree with your assertion that we should be putting in Share requirements or so on to get some of these things fixed. Thanks for that, that's good to hear. Yes, I agree, IBM has helped move the OMEGAMON products forward in a positive direction, putting more resources into development. (More to come.) And all the heritage Candle people I've met are terrific. Here's what I actually wrote (*emphasis mine*): As for the other comments in this thread, I would suggest getting the requirements into SHARE *and through other avenues*. Other avenues can mean a lot of things, e.g. PMRs. Yes, you can open PMRs against documentation. That was one of the avenues I had in mind when I wrote that. Sorry if I was a little obtuse. Fortunately, as Joe Winterton writes, the documentation will get updated. (Thanks, Joe.) - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Is IBM/Tivoli turning into CA?
Doug Fuerst writes: And every piece is another charge. SMS? Gimme some money. USS? Gimme some money. DASD, VTAM, whatever? gimme money. I cannot comment on everything mentioned in this thread, but I can comment on this area. Candle used to have every monitor priced separately, yes. IBM changed at least some of that. As of V4.1 (December, 2006), the single Tivoli OMEGAMON XE for z/OS product now includes UNIX System Services monitoring, for example. Cryptographic monitoring used to be separate, and now that's in the same product, too. As another example, Tivoli OMEGAMON XE for Mainframe Networks combines TCP/IP and VTAM functions. IBM also has a no-charge Tivoli OMEGAMON XE for z/OS Management Console product which you may download. You may be remembering MAINVIEW, in fact, and have this reversed in your recollection. BMC lists MAINVIEW for z/OS, MAINVIEW for VTAM, MAINVIEW for IP, and MAINVIEW for UNIX System Services separately, among other monitors. I believe ASG's TMON is very similar in its splits, to pick another example. Or it's possible you missed the December, 2006, announcements where this changed after IBM acquired Candle. That's OK -- sometimes it's hard to keep up. Mainframes have a lot of velocity now. In fairness I really don't think this packaging factor is particularly important. Each vendor is trying to establish the right granularity for their monitoring products because each customer is different. That way you can pay for as much or as little function as you need. Lately IBM has been consolidating more functions into the single products (e.g. OMEGAMON XE for z/OS and for Mainframe Networks) since, at least in IBM's experience, most customers now need more base functions -- simplicity over too much granularity, basically. At the same time IBM may introduce more new monitors. Tivoli OMEGAMON XE for CICS Transaction Gateway is the only product that monitors CICS TG, for example, and is a very recent addition to the family. As for the other comments in this thread, I would suggest getting the requirements into SHARE and through other avenues. I know there's a lot of IBM effort to work through requirements, and a lot has been done but more to come. One thing that I care about personally is that V4.1 OMEGAMON XE products now have complete Japanese language support, so that's quite helpful here. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Price of CPU seconds
I think if you're simply trying to educate someone familiar with distributed servers, I would ask something like this: Are servers free? And (hopefully) they'll answer, No. Next question: So how would you measure their costs? If it's a multi-user server, as most servers are, how would you allocate their costs among users? And they might give a long answer which may or may not be insightful. Then I would simply say (assuming they demonstrated some insight): There's really nothing different here. One of the ways you can measure and allocate costs is by calculating the number of CPU-seconds that a user requires and calculating a cost per CPU-second. That cost might be a very small number, but it's never zero. CPU-seconds certainly aren't the only way to base costs or even the best way. But it's one way you can do it. The same thing is true for electricity. Your electricity consumption is measured in Kilowatt-hours. One Kilowatt-hour might be very cheap but, added over many thousands of users and over many years, that small number adds up to a very big number, enough to pay for huge electric power plants. The same is true of servers of any type: there are underlying costs and, for every platform, one user might consume a very small amount, but the cost is never zero. Try something like that and see where it goes. Everything I said earlier still applies, though. Continuing: Now, if CPU-seconds always cost some money, doesn't it make sense to try to save CPU-seconds if possible so you can get more users for the same cost? You can keep the lights in your house turned on all day while you are away, using electricity, but it will cost more and provide no value. Doesn't it make sense to get energy-efficient lightbulbs to save money and to turn them off when you are not using them? That's what performance engineering is about, and good performance is important on every system. On systems which serve the largest number of users, it's can be even more important. A little bit better performance can mean some large cost savings. Users like faster systems, too, because they can get more work done -- it saves them money because they aren't wasting time. And at that point, hopefully, the lightbulb will go off. No pun intended. :-) Hope that helps. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Replacement for CA - TPX - Session Manager?
IBM has three session managers, yes. IBM prefers you look at Session Manager for z/OS first since it's the strategic one: http://www.ibm.com/software/htp/cics/smanag/ Version 2.1 was introduced this past September, so it's worth a fresh look if your knowledge is pre-2.1. (One major plus for me is that Session Manager is available in the Japanese language.) If you would like more information on migration assistance for Session Manager please contact Twila Brent ([EMAIL PROTECTED]) and she can point you in the right direction. For the record there are two others. AFAIK there are no plans to discontinue the other two; they're still supported. One is CL/SUPERSESSION: http://www.ibm.com/software/htp/cics/clss/ which IBM acquired via Candle and added to the IBM catalog in 2005 (along with CL/CONFERENCE). The other is Tivoli NetView Access Services: http://www.ibm.com/software/tivoli/products/netview_as/ which enjoyed its most recent 2.1.1 update (up from 2.1.0) in December, 1997, and which is also still supported. There are many other products in this category, so please check the archives for a list. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: IBM contact for documentation servers?
John, Are you talking about stuff on the Internet? At the bottom of every ibm.com page there should be a Contact link. Click on that, then click on the E-mail link. Fill out the form as best you can and please include the Web address you're complaining about. I've had very good results getting stuff corrected or updated this way. You'll get a level 1 sort of semi-automatic response initially, but each page has an owner who will eventually receive your complaint, and usually you'll get a personal response and be able to follow-up with that person. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Price of CPU seconds
not. Of course the outsourcers charge a profit for their services, and they perform to written agreements. If it's not written down, the outsourcer doesn't have to do it. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Linux zSeries questions
Yes, the various C/C++ compiler enhancements (Metal C, graphical workbench support in Rational Developer for System z, improved debugging, etc.) that recently shipped aren't only for external use. Although please enjoy them, too. Other bits I'd take an educated guess (without inside knowledge) that are largely or entirely C and/or C++: - parts of the XML Toolkit for z/OS - at least parts of UNIX System Services (one would think most of the command binaries have got to be, for example) - the IBM HTTP Server for z/OS (including the Apache-derived one shipping with WebSphere Application Server V6.1 for z/OS) - some security bits (like the GSKKYMAN stuff) And there's probably a lot more. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Linux zSeries questions
FWIW, there's a fair (and increasing) amount of z/OS written in C and/or C++. The SDK for Java immediately leaps to mind as an example. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Linux zSeries questions
Alan Altmark writes: However, you will pay z/VM and Linux middleware license fees based on the total number of CPs, even if you have just one LPAR with one logical CPU. Those with larger CP counts will likely find adding a couple of IFLs to be a better strategy. True. Just to expand on what Alan is saying, for z/VM and much (but not all) Linux middleware you have to license all CPs in the machine (if you run on at least one CP), all IFLs (if you run on at least one IFL), or both (if you run on both). So if you have a large number of CPs that's something to consider. That said, licenses are portable. For example, if today you have 3 CPs in your machine, you can start running Linux on, say, a fraction of one CP. You'll then have to buy three z/VM processor licenses (assuming you're running under z/VM) and three middleware licenses (with the exception of some sub-capacity products, e.g. WebSphere Application Server for Linux on z). But if you then decide you like what you see, you can buy 3 IFLs, move z/VM and Linux onto the IFLs (and off the CPs), and you already have all the software licenses you need. Also, at least in IBM's case, middleware licenses for Linux are portable. For example, you can retire a Windows X86 processor and move that platform's IBM middleware licenses to Linux on z without even phoning IBM. I got the sense the original poster just wanted to boot up Linux on a CP and have fun with it, probably in an LPAR or two without z/VM and without any particular licensed middleware products. (Or that he already had z/VM.) Hence my original answer responded to that situation. But hopefully Alan and I are helping explain the bigger picture now. In short, there's certainly no technical impediment to running Linux on CPs, and, at least for kicking the tires, that often makes financial sense, too. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Person-Months for Version-to-Version Migrations?
The discussion here about how many z/OS system programmers? is really interesting to me, and I'd like to ask a related question to expand the discussion a bit more. What are your current experiences in terms of the number of person-months it's taking for version-to-version (v2v) upgrades? Here's a sample template for how you might frame your answer since there are lots of factors that could influence the person-months: 1. Start date of v2v project? 2. First in-production date? 3. Final in-production date? 4. Estimated total person-months? a. Internally supplied: b. From outside service providers: 5. From what to what? a. Operating systems: b. Middleware (CICS, DB2, IMS TM, etc.): c. Tools/utilities: d. Hardware: 6. How many LPARs? (Any logical consolidation?) 7. How many machines? (Any physical consolidation?) 8. How many (logical) Parallel Sysplexes (e.g. two 3-way Sysplexes)? 9. Any other quantifications (e.g. lines of code, function points, number of programs, number of CICS regions, database sizes/number, etc.)? 10. How many (planned) outage(s)? How long? 11. What other impact(s), if any, were there to business users during the migration (e.g. performance issue, unplanned outage, function loss, new function, data loss, etc.)? 12. Testing method(s)/person-months? 13. Other comments? (Examples: What were the most difficult issues? What took the most time? Any unusual factors, such as relocating the data center? Did you use an Early Support Program?) You may not know the answers to all that, so your best guess is fine and much appreciated. Private e-mail is OK if you prefer. I'm asking because my guess is that the Japanese experience in these areas is quite different from most of the rest of the world, probably because of different IT practices here. But that's just a guess, so it would be interesting to hear what IBM-MAINer experiences are around the world in the 21st century. The from what to what? question might be very different also. In Japan the big hop/all-at-once method seems popular, while I tend to think in the rest of the world the ongoing rolling wave technique is at least more popular. So your answer might be z/OS 1.6 to 1.8 for one migration, then followed by CICS TS 2.3 to 3.2 for another migration, then z990/z890 to z9, etc., each as separate v2v projects. Thanks in advance for your feedback. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Linux zSeries questions
1. In addition to the general Linux information sources, all of which still apply -- Linux is Linux -- you may wish to try the LINUX-390 LISTSERV, hosted by Marist University. You can sign up here: http://www2.marist.edu/htbin/wlvindex?linux-390 Please scroll down toward the bottom of that page, and there's a sign-up form. 2. I'm not sure about the wait state codes, but they may not be too useful to you anyway. You may want to ask in LINUX-390. 3. Yes, you can certainly start and run Linux under z/VM on CPs (general purpose processors). General purpose means you can run anything on them -- CPs are the universal processors. IFLs are the processors dedicated to Linux, but CPs work just fine, too. If you want to run a trivial amount of Linux -- booting it, having fun with it, experimenting with it, doing a little real work with it, etc. -- then existing CPs you already own are probably the most economical. (You already own them, and spare capacity is basically free.) Once you get serious about running Linux beyond some trivial amount you'll probably want to invest in at least one IFL. Enjoy. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Linux zSeries questions
Ed Gould asks: Can I ask an innocent(?) question here.. LINUX is it documented like MVS or something like early 360 (YES 360) manuals ? Linux is maintained and enhanced in a community-driven manner, with no single entity (e.g. IBM) controlling its entire development. As a consequence the documentation won't be IBM-style or for that matter HP-style. While Linux developers in the community do produce a lot of documentation, and there's a lot more beyond that in books and on the Web, one of the major benefits of a Linux distributor, such as Novell and Red Hat, is in their packaging and documentation services, especially concerning installation and configuration procedures. But, for example, while you'll find wonderful diagnostic code numbers on every line as you observe a z/OS IPL, the Linux kernel doesn't provide such troubleshooting niceties. You'll have to figure out what a sometimes cryptic string means without a code number, much less a book with a list of code numbers and troubleshooting guidance. It's a bit more Wild West. Which is not to say it's bad, it just has different design origins and different community values. Linux does tend to inherit UNIX-style in-built documentation features. The man pages (man command) are notable examples. You can download Linux from lots of places and take a look at the documentation it includes (or doesn't include), depending on the distribution. To pick one example at random, you can download Linux on System z from Novell by going here: http://www.novell.com/mainframe and clicking on one of the download links at right. Novell recently introduced a starter system package for System z, a pre-built image which is more convenient to set up than the traditional approach. Wikipedia lists many other Linux distributions for System z. To cite another example, Red Hat posts their documentation here: http://www.redhat.com/docs/manuals/enterprise/ - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
IBM PR: The Future Runs on ?????? ? Events in North America
IBM has posted a large number of city events billed as The Future Runs on ?? ? -- yes, that's really what it says -- now open for registration at: http://www.ibm.com/events/systemz The Web page currently lists cities in North America. If I spot other Web sites with comparable listings outside the U.S. and Canada I'll follow up with a post. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
IBM PR: Almost Introducing the Extraordinary New XXXXX
Copy of an invitation which I think this audience may be interested in. The X represents redacted text which appears in the invitation as black bars. == Sign up now for live Web seminar on February 26 at http://www.on24.com/clients/ibm/102818 ALMOST INTRODUCING THE EXTRAORDINARY NEW X We can't reveal much about the amazing new X just yet. Through virtualization and consolidation it can help lower your power and cooling costs by up to a remarkable X. It can help cut floor space needs by up to X. It is designed to deliver capacity improvements of up to X. Yes, amazing. Simply, your data center has never seen the likes of the new X before. Just when you thought the X couldn't get any more powerful, well, it has. Learn from those in the know. Speakers for the live Web seminar have the inside scoop -- and they'll share it with you. They are: Jim Stallings, GM, Enterprise Systems Division, IBM Systems Technology Group and Bob Hoey, VP, Enterprise Systems Division, IBM Systems Technology Group. The Web link will take you to IBM's partner, On24, who will help complete your Web seminar registration. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Curious(?) way to ZIP a mainframe file.
John, when you zip a file using the jar command, do you see any zAAP eligible workload? Just curious. Although the jar command is part of the Java runtime, I'm not sure whether jar sends anything to the zAAP. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Arial Font printing on Mainframe
In addition to the suggestions you've received, a few more comments: 1. I cannot find that printer model number you mention, so could you double check that? 2. Windows is capable of printing all fonts using bitmaps. (There's a checkbox in the printer driver settings that enables that.) The size of the print job will grow, so that's one disadvantage. If you can transfer and print bitmaps in some form, this is one option. 3. According to Wikipedia, Arial is just the new Microsoft-created name for Sonoran Sans Serif. Arial/Sonoran Sans Serif was designed by Monotype. It's very close in appearance to Linotype's Helvetica, and it is a variation of Monotype's Grotesque font. Triumvirate, Helios, Megaron, and Newton fonts are other substitutes for Arial and Helvetica. 4. Both the InfoPrint Fonts (5648-E76) and the AFP Font Collection ( 5648-B33) include Helvetica. The former includes only outline Helvetica while the latter includes both outline and raster (bitmap) Helvetica (240 and 300 dpi). You can still order Sonoran Sans Serif (5771-ABB) -- just checked -- but I think that's 240 dpi raster format only (e.g. 3820 laser printing). That may be sufficient, depending on your output device and appearance goals, but I think I'd look at outline Helvetica first. Also, PSF for z/OS supports OpenType/TrueType font formats directly. See this guide (IBM Publication No. G544-5876) for more details: http://publibfp.boulder.ibm.com/epubs/pdf/g3a00mst.pdf Assuming you license Monotype's Arial -- it is available for purchase -- presumably you could use it in PSF for z/OS. I don't have too much information on Exstream Dialogue, but presumbly you could do something similar. Hope all that helps. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: New Opcodes
As far as I know there are only two things IBM has said publicly about a future mainframe processor, and I guess they could be the same or different. Here's the first, from August: http://www2.hursley.ibm.com/decimal/IBM-z6-mainframe-microprocessor-Webb.pdf which says 50+ instructions added to improve compiled code efficiency. The second is what the IBM CFO said in his prepared remarks concerning IBM's 4th quarter, 2007 earnings. Beyond that, I haven't seen any more information. So let's wait to see if/when IBM says more, OK? - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: AIX CICS TRANSACTION GATEWAY EPI QUESTION
YY, HATS is another option, yes. If you absolutely must drive screens, you may also want to look at the CICS Service Flow Feature. There's a very good redbook here: http://www.redbooks.ibm.com/abstracts/sg245466.html which discusses most of the ways you can integrate WebSphere Application Server and CICS Transaction Server, if you're looking for some good background reading. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Problems with SNA Consoles in Z/OS V1R8
Mark may have more current information than I have on this issue. Maybe z/OS 1.7 changed something again. I still think this is important to fix on the z/OS side, because increasing numbers of clients will negotiate contention resolution by default. That's true for multiple IBM client products like Host On-Demand and PComm, other products like HATS, and (from what I've been reading) many non-IBM products. Lots of products have already implemented that RFC, and more will over time. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: DB2 queries without using MF.
Rob Wunderlich writes: Ideally, I would use Connect Unlimited. Problem for me is the price tag. It's only available as OTC, and a fairly hefty one at that. If it could be folded into the monthly cost of DB2 it would be a no brainer. That particular OTC happens to exceed a magic number for this organization that makes approval difficult (long story). Ah yes, the bureaucratic expense number limit policy. Can't buy a building, so the company pays more buying brick by brick. I think everyone working for a business with at least a few employees runs into that problem. I don't know your exact situation and wouldn't be the expert even if I did, but if you don't yet have an Enterprise License Agreement that might be one way to solve this particular problem. I wrote a layperson's explanation of the ELA a few weeks ago that should be in the IBM-MAIN archives. You could also just ask IBM straight up if they can price DB2 Connect Unlimited as MLC, like MQ's Client Attach Feature. Or IBM can finance software purchases to adjust the cash flows, but often that's not the issue when it comes to the bureaucratic approvals. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: DB2 queries without using MF.
An IP address can be easily spoofed, so I'm not sure why people think that restricting to particular inbound IP addresses offers much protection. But even if it did, that's a single level: all or nothing. I doubt that most businesses would find all-or-nothing access to their entire collection of core enterprise data sufficient. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: DB2 queries without using MF.
Rob Wunderlich writes: JDBC/ODBC access to z/os DB2 works well, but it's expensive (relative term). I think you used the word perceived elsewhere, and there are those perceptions, yes. It's a multi-party effort to make sure the truth is understood. The cost for drivers, whether DB2 Connect or another vendor, is enormous. I have web servers, data warehousing, windows server apps and desktop clients all accessing DB2 data. If I want to add a CPU to webapp server, the driver upgrage fee is more than the cost of the entire server. You would seem to be in the perfect situation, at least outwardly, for DB2 Connect Unlimited Edition. It's very much like the MQ Client Access Feature: you pay a fixed rate based on your MSUs, and you're done. You don't even need to contact IBM when you add server #685 or user #3163. (Put Connect on Linux on z and add a zIIP for best results.) Other vendors may offer similar terms. I'm sure all those Web servers, data warehousing servers, and server apps are free to acquire and maintain, but that's a topic for another day. :-) On the general topic, there are about a million ways already to copy DB2 z/OS data somewhere and do something with it. They all share some common disadvantages, many already mentioned. The winds seem to be blowing against doing that sort of thing nowadays. A lot of businesses are terribly worried about failing to protect sensitive data, and the word copy is inherently antithetical to data protection (except in the narrow and tightly controlled DR sense). I call the trend data recentralization. There's also an increasing appreciation for the high costs of too highly distributed data models, and data warehouses are becoming much more mission-critical (and numerous) Interesting times we live in. My free advice, for what it's worth, is to figure out better ways for customers to take advantage of DB2 (and other data) right on the mainframe, to answer critical business questions on a need-to-know basis, with up-to-the-second consistency. There's a tremendous market for that. IMHO, creating the million-and-first way to copy data somewhere else to then operate on it won't be as interesting. By the way, I expect that going behind DB2's back and accessing underlying files will become increasingly less and less fruitful. DB2 function has been galloping ahead rapidly, and it's going to get more and more difficult to make any sense of what's underneath. There's also no guarantee whatsoever that what's underneath will stay the same from version to version. And more and more of it is going to be encrypted anyway as, for example, customers use SQL ENCRYPT vocabularies, so you'll need the key(s) to decode it. Those who have the key probably won't (shouldn't) give it. There are also complications like stored procedures, which are getting more numerous and complex, expanding data types, XML, rapid changes to indices, conversion to Unicode This stuff is in very rapid motion, and you'll need to keep up. Basically what you're talking about is reverse engineering a good chunk of DB2, and to keep reverse engineering it as DB2 evolves. My hunch is that's a big, never-ending project. :-) - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: AIX CICS TRANSACTION GATEWAY EPI QUESTION
YY, If you're programming using EPI in WebSphere Application Server with CICS Transaction Gateway, you may want to refer to the EPI sample code included with CTG. I'm copying this section of the CICS Transaction Gateway V7.1 InfoCenter for reference: - - - - - The EPI Sample consists of a stateful session bean, a client application, a custom record which demonstrates the use of the Screenable interface, and a custom LogonLogoff class.The following files are part of the EPI Sample: EPIPlayScript.java Enterprise bean remote interface EPIPlayScriptHome.java Enterprise bean home interface EPIPlayScriptBean.java Enterprise bean implementation EPIPlayScriptClient.java Enterprise bean client program CICSCESNLogon.java A LogonLogoff class Ejb-jar-epi-1.1.xml Example of a deployment descriptor - - - - - These samples will be either in the samples directory (wherever you install CTG, including z/OS zFS or HFS) or in the deployable directory (for WebSphere Application Server EAR-deployable samples). Note that there are multiple ways to connect WebSphere Application Server to CICS Transaction Server, and CICS Transaction Gateway is but one of them. Also please bear in mind that EPI connections are problematic in certain deployment situations. In general, if you are using CTG, if you can move to IPIC or at least ECI you're probably going to end up with a more robust solution. (IPIC supports several new things that ECI doesn't, such as channels and containers for accessing 32K of CICS application data at a time.) You might also want to ask your questions on the CICS-L mailing list. Hope this helps. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Problems with SNA Consoles in Z/OS V1R8
Miguel, Disabling contention resolution on the client might be a temporary workaround, but I wouldn't recommend that as a permanent workaround. Contention resolution is a very good thing to have enabled -- it can dramatically speed up things like macros -- and you can just about guarantee you'll have an endless number of phone calls from clients otherwise if you don't fix it before too long. I thought way before z/OS 1.8 the contention resolution problems were fixed and everything runs smoothly. Let me go dig up my old notes OK, here's the history. Contention resolution debuted in z/OS 1.2. It takes two to tango, though (server and client), and for 1.2, 1.3, and 1.4 there were discrepancies in how z/OS implemented contention resolution and (later) how the first clients did. Before z/OS 1.5's GA everything settled down, and there were PTFs released for prior z/OSes. The contention resolution specification is defined in IETF RFC2355, but sometimes it's tough to implement the specs exactly the same way. (As to why IBM released a client that didn't work with its own server contemporaneously... can't explain that one.) There's an informational APAR called Common Telnet Problems Under z/OS which you should check to see if there are PTFs that apply to your particular z/OS release. The APAR is II13135. To save you the reading, Miguel, at this instant that APAR advises these fixes for z/OS 1.8: UK15163, UK15227, UK16016, UK19642, UK16746, UK19164, UK19835, UK21579, UK25194, UK26064, UK32220, UK25807 and further advises searching on R170 TSASO if you're running TELNET in its own ASID. In addition, these non-TELNET fixes may be relevant because they might impact TELNET: OA11652, OA11841, OA15828, OA16468, and OA17750. (Some/all of those may not be relevant to 1.8.) My hunch is that Mark Zelden correctly remembers when contention resolution indeed was a problem, but I thought IBM had permanently buried those specific problems as of z/OS 1.5. Maybe they've come back, or maybe it's something else. Applying the fixes in II13135 should squash anything IBM knows about on the z/OS side. Do note there's a Personal Communications 5.9.2 now, so I would also try that to see if the behavior changes. (There's also a 5.8.3.) Don't automatically assume the mainframe is at fault, which reminds me I remember arguing with the PComm team a few years ago that they had a ridiculously short timeout value as the default. I think the timeout affected TN3270E SSL handshaking across slow network connections, including modem dial-up, the average corporate VPN, and other high latency/high congestion situations, making it impossible to connect. They argued that lengthening the default might break something else, so they didn't want to change it. I argued that the certainty of having something broken (SSL handshakes) ought to trump the remote possibility that a longer timeout might break one user's macro (or something -- I don't think anybody could think of a plausible breakage scenario there). My argument won, and the default timeout is now longer. All of which is very interesting but may have nothing to do with your problem. :-) Hope all that helps. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Co:Z Co-Processing Toolkit for z/OS is now FREE
Kirk, It's worth mentioning, from the narrow view of network encryption, that you could traverse a HiperSocket to Linux on z using the unencrypted socket connection and not have to worry about anyone intercepting the data. There are many other security issues to consider, but that's one way to avoid the cost of encryption yet still protect the connections. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z/OS 1.8 or 1.9 ?
Re: Coexistence, yes, it's extremely important to stay current. However, if you're sitting back at z/OS 1.(low number) or OS/390 or (gasp!) MVS, the best (and maybe only) strategy is to make one big, giant jump to currency, then stay current. In that first jump you won't be able to share datasets and be supported -- it's a cold turkey upgrade. Basically you'll quiesce the old system, unload the data from the old OS, load the data into the new OS, and go into production. You'll need to do a lot of testing before you do this, you'll have to take a scheduled outage, and the risk is higher than N+2 or N+1 steps -- but still the lowest risk of any other option. (Although I do point out that it's what every Windows and UNIX administrator does by and large. Boy, are we spoiled having fully supported N+2 coexistence. :-)) You may get lucky and be able to do this in chunks, for example moving a single less critical application first. I know there are many of you who will need to do this cold turkey upgrade, just please don't let it happen ever again. It's a lot less stressful having that safety net. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: DB2 queries without using MF.
Shadow is quite different, as are DB2 tools. They're all running on the mainframe itself. In addition to the problems mentioned, you may have all kinds of data integrity and consistency problems if you try to write something. Even reads aren't necessarily going to have full consistency unless you quiesce DB2. One can think of a lot of drawbacks, but what are the benefits versus, say, JDBC or ODBC with zIIP? - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: New Mainframes coming in February
I have found that my company (before downsizing me) was totally confused with the EC/BC determinations. I wish IBM would try to simplify their offerings. I've dealt with their marketting cr*p for 27 years, and I find their distintions only make sense to IBM sales. Ted, what are the confusing parts about EC and BC? It should be very simple: there are two pieces of hardware. The BC starts at 26 MIPS of CP capacity and goes up to almost 1,800 per frame. The EC starts at about 200 and goes up to nearly 18,000. (There's plenty of overlap between the two so you have room to grow.) If the BC provides enough capacity, that's what you buy, otherwise the EC is available. You can upgrade a BC to an EC. Both run the same software portfolio. Both deliver mainframe qualities of service -- both are 100% genuine mainframes. (The EC always has at least two spare processors, while the BC lets you choose whether to have spares or not. The EC does have an optional feature available to let you replace a whole processor book while the machine continues to operate, but this is perhaps academic if you have Parallel Sysplex.) You can stuff more memory inside an EC, but the BC still offers lots. No surprise that the EC is physically about twice the size of the BC. I have heard about some confusion out there about certain things, but EC v. BC is a new one, so I'm curious to hear more about that. I think ever since the z800 came out 5 or 6 years ago this two machine strategy has been in place, and it hasn't changed fundamentally. The EC and BC are round #3 of that strategy, after round #2 (z990/z890). In all three rounds the bigger one debuted first and the smaller one second. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z890 2086-160 w/ 2 IFLs on eBay
My question is: If you have two z890s, A with 1 CP and 1 IFL and B with 2 IFL, then in a disaster where A is under a pile of rubble, can you get IBM to turn on a CP and an IFL in B and still pay the same price for z/OS z/VM? Yes, although it's a good idea to contract for that in advance using CBU (Capacity Backup) in order to be able to turn on that disaster recovery capacity promptly. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z890 2086-160 w/ 2 IFLs on eBay
If I recall from the marketing, you only 'buy' an IFL once and when you upgrade your IFL's are already paid for. Now if I can pick up this 2 IFL box for $10-20K can I just have IBM turn on 2 IFL's on my z990 and just trash the z890? No, I don't think so. You will then own two z890 IFLs. BUT, whenever you're ready to upgrade/consolidate to z9 or beyond, things may get more interesting. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Really stupid question about z/OS HTTP server
John, Looks like you've received some good replies. One other thought is to ask the question whether you actually need authentication for the content that your z/OS HTTP server is providing. Also, assuming you're using basic authentication, there are some cheats. Your users can add a bookmark/favorite to log in automatically, like this: http://userid:[EMAIL PROTECTED]/mypage.html Better yet, if your Web site supports SSL: https://userid:[EMAIL PROTECTED]/mypage.html Or you could use a Web form-based authentication instead of the basic authentication pop-up box. With the form-based authentication Microsoft Internet Explorer can remember user IDs and passwords if you wish, to automatically log the user on. (Your company may have disabled that in the IE settings, but it is a standard feature of most browsers, including IE.) - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: IBM complementary tools
The products mentioned: CICS Batch Application Control http://www-306.ibm.com/software/htp/cics/bac/ Session Manager for z/OS http://www-306.ibm.com/software/htp/cics/smanag/ Workload Simulator for z/OS http://www-306.ibm.com/software/awdtools/workloadsimulator/ CICS VSAM Transparency http://www-306.ibm.com/software/htp/cics/vt/ are all licensed products with a one-time charge and (optional) annual subscription and support. They are not free. For the record, Session Manager for z/OS supports TN3270E. It's the much older Tivoli Netview Access Services (NVAS) product that doesn't. However, there is a free trial download for CICS Batch Application Control and some other CICS tools here: http://www-306.ibm.com/software/os/systemz/trials/cicstools/ Workload Simulator and some other tools are available for a test drive here: http://zserveros.dfw.ibm.com/ - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
PR: IBM Beacon Award (for Business Partners) Nominations Now Open
IBM is now accepting nominations for the IBM Beacon Awards. Details here: http://www.ibm.com/partnerworld/pwhome.nsf/weblook/2008_awards.html Nominations are due by January 25, 2008. This award is given to outstanding business partners in a variety of categories. The most relevant category for IBM-MAIN readers is likely Excellence in the Use of System z for New Workloads or Applications, but there are many other mainframe-relevant categories. Last year, Cornerstone Systems won in the category Best IBM System z9 Solution for their excellent work to help CompuCredit deploy their new Installment Loan Application. Some of the other mainframe-relevant winners include NIWS Co., Ltd. (Japan) and Sogeti Group (France) for Overall Technical Excellence. If you are an Advanced or Premier Business Partner, you can nominate yourself. Your solution should be new or substantially improved in 2007. Judging criteria and other details are available on the Web site. Good luck, everyone. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
PR: New Videos on IBM TV about IT Costs
There is a new series of videos posted to IBM TV that I found interesting. First, start here: http://www.ibm.com/software/info/television/index.jsp Click on the Select a topic button, then choose either Systems Storage (or Software) and navigate to the System z - All media types section. You'll then get a list of videos. Several are interesting, but look for the Scorpion series and start with part 1 if you want to view the cost-related ones. One thing I actually disagree with slightly in Part 1 is the speaker's statement, Everybody knows the cost of their mainframe software, as if it's a fixed given. That may be a U.S.-oriented perspective perhaps, or perhaps it was an oversimplification for a short video. But I've met a lot of customers that have many misperceptions in this area, and many do not manage their software portfolios optimally. That's regardless of platform. Software mismanagement includes not optimizing what you've got for cost, and not keeping the portfolio in line with current needs. In simple terms you buy software to avoid labor. Theoretically you could buy a machine and hire your own army to write an operating system, middleware, tools and utilities, etc. When computing first started, that's what you had to do. It's expensive, so almost no one writes all their own software. Even big software companies like IBM, Microsoft, and Oracle with seemingly endless programmer resources still buy lots of software. But where you draw that line (between buy and build) varies and will likely change over time as labor costs change. Typically that line should move up over time, because typically labor costs are increasing, and the ability of software vendors to spread those costs is increasing, especially due to international software market expansion. That's a universal pair of trends, not platform-specific. Balanced against that is the fact that your own software code is your business because it's just for you, so you don't have to modify your business to match more general-purpose software. Which is yet another reason why Service-Oriented Architecture (SOA), to meld your own software bits seamlessly with the commercial stuff, is so important, but that's a topic for another day. The correct answer may be, given these trends, Buy more software, but optimize well. On the buy more side, I generally favor looking first at application developer productivity, because that's where you'll find significant IT cost. That's also where business benefits result, if developers can deliver quicker and with higher quality. I get very concerned with organizations where this software line in the developer productivity area hasn't moved in 20+ years: that's often a big warning sign that there's mismanagment. Anyway, I found the videos interesting and pass them along. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Capacity on Demand?
and raising your softcap, you can in effect transfer the OTC license for that day. It's total enterprise Value Units at any/every four hour rolling average that counts. In this example there's only one machine, but hypothetically if you could reduce the CICS LPAR by 2 MSUs let's say, you'd end up with this: 33 MSUs for Tivoli zSecure Admin (+3) 33 MSUs for DB2 Cloning Tool for z/OS (+3) 13 MSUs for CICS Interdependency Analyzer (no change; you already have 15 permanently licensed) 13 MSUs for WebSphere Application Server for z/OS (no change; you already have 15 permanently licensed) So you can have various permutations like that, and you might be able to slide just under another rounded up Value Unit. You can also buy a mixture: some more perpetual licenses, and some MSU-days. Whatever makes sense in your particular situation. [And this also points to an interesting effect of the technology dividend: when you buy IBM zOTC for a specific machine, when you upgrade to the next model you'll find that your existing stash of Value Units should cover more system capacity (MIPS), maybe 10% more. You can either take the extra capacity immediately and pay nothing more for IBM zOTC (and MLC), or you can wait to grow back up to your Value Units before buying more, or some combination. Are you reflecting that benefit in your chargebacks? :-)] A couple other points are worth mentioning: 1. I don't think OOCoD affects your qualification for Parallel Sysplex aggregation, as long as the deviation is genuinely temporary. So if CoD throws you out of the 50% rule for a day, for example, don't worry about it, unless you think that signals a future, more permanent condition. 2. Previously I talked about IBM Enterprise License Agreements (ELAs). You'll want to make sure you understand the effect that increasing MLC would have on your OTC. It's quite possible that it would make sense, earlier, to opt for increasing your MLC forecast in order to get some OTC benefits. Take a look at my previous ELA write-up to get some background on that. All that said, John, there are some very clear cases where CoD makes enormous business sense, and they typically involve businesses either with short duration annual spikes or a truly once-in-a-lifetime spike. An example of the latter that I can think of is when the government changes a law or regulation, causing a once-only rush of some kind, perhaps to take advantage of an expiring tax provision. There are also cases where you're genuinely not sure whether you need the extra capacity or not, or how much you need, so you can pay much less for a day or two and see how it goes. And there are things like mergers, when you're buying another company and you need some extra capacity to load all their data onto your mainframe overnight, within a certain time window. CoD, for hardware and software, are attempts to price fairly for the way businesses work in the real world, to offer flexibility. Everything I described is published, announced, and standard. You don't have to go begging to a vendor (IBM anyway) for a special favor. The z world changed in 2000, as I'm fond of saying. There is a flavor of CoD called Customer Initiated (CI). I recommend that almost everyone get that if possible. That gives you (and your management) full control should the unexpected happen -- or if you just forecasted badly. :-) It's a prudent contingency feature, pre-negotiated, and I recommend it. But don't let CoD substitute for what should be a permanent capacity increase, and, as a general rule, try to avoid buying anything remotely resembling permanent capacity on N-1 (or older) models. CoD is a great business tool, but please use it wisely. Still confused? Fire away, and I'll try to clarify. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: About 1 in 5 IBM employees now in India - so what ?
Mohammad Khan writes: Even that's not so clear cut anymore. Indian company Hindustan Aeronautics Limited(HAL) has been supplying parts to Boeing for some time. Yes, indeed. Boeing sources parts from around the world, including India. Japan is a major supplier of aircraft parts and subassemblies to Boeing, for example. (And Boeing thoroughly dominates the Japanese airliner market.) However, again, the growth figure I cited was in net U.S. dollar terms, not Indian Rupees. HAL's aircraft parts get counted as exports from India to the U.S. (imports to the U.S.), and then Boeing's whole assembled airplane gets counted as a much bigger export from the U.S. to India. The 25+ growth (2006 v. 2005) measures exports. I can safely assume that U.S. labor productivity did not grow at 25 percent, even including extra overtime, so by definition that export growth created jobs in the United States. There's no question there are both winners and losers (from the U.S. employment perspective) in U.S.-India bilateral trade, and there's also no question that changes in trade flows are disruptive. But there are many, many winners, and that's my point. I leave it to others to compare, contrast, and quantify the winners v. losers, but hopefully not in IBM-MAIN. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Java Problem Analysis
Kirk, I tried to phrase my (limited) answer succinctly yet accurately, but it seems I didn't entirely succeed. The case you're describing is where the original author of the code (Sun in this case) (also) releases their code under non-GPL terms. In that case, it's not GPL code, is it? :-) (That specific code isn't.) Thus my explanation doesn't apply -- I wasn't explaining anything about how code under different licenses works. I was explaining how GPL code works, and that specific code isn't GPL. If IBM or anyone else bases their work off the GPL source, then my explanation does apply. It really is that simple. Yes, IBM has done some GPL-related work on z/OS. Here's some: http://www-03.ibm.com/servers/eserver/zseries/zos/unix/bpxa1ty1.html I believe you'll find several on the list with GPL licenses, or at least licenses that are comparable to the GPL for purposes of this discussion (i.e. which assure access to source code), e.g. emacs. But you did expand our knowledge. So you say that Sun's Java(TM) is available under non-GPL terms. Thus we can conclude there is no assurance that derivatives (from IBM, Sun, or anyone else) of the non-GPL code will be available in source code form. That directly addresses the I don't know part of my answer -- the first sentence :-) -- and I'm grateful for that. Thanks. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: financial functions for Enterprise COBOL
Donald Imbriale writes: I'm looking for a package of financial functions callable from a COBOL program. A search within the Enterprise COBOL and LE bookshelves has turned up nothing and a Google search for Enterprise COBOL financial didn't turn up any useful leads. I'm looking for functions like Excel's XIRR. We should probably explain what XIRR is first. It's the name of an Excel function used to calculate the internal rate of return (IRR). Businesses use the IRR as a basis for choosing between, say, two different investment flows that result in two different cash flows. (I can't imagine how your company could possibly use that. :-)) Mathematically, NPV(C,t,IRR)=0 That is, the Internal Rate of Return (IRR) is the discount rate which sets the Net Present Value (NPV) for a series of Cash flows (C) at certain times (t) to zero. (Thanks to www.invest-faq.com for that explanation.) Cash flows include some negative values (payments or investments) and positive values (dividends or returns). If you have multiple investments, there's no straight up mathematical solution to this formula -- you have to determine IRR iteratively. You try an IRR value (say, 10%) and see how close you are, then iterate until you get close enough. It would seem that COBOL's PRESENT-VALUE function would be a big piece of the puzzle if you wanted to write your own code. You'd iterate until either the PRESENT-VALUE result is close enough to zero -- I think Excel says that +/-0.01% is close enough -- or until you reach, say, 100 iterations, whichever comes first. (Excel's default is to give up at 100.) You need to have at least one negative and one positive value in the cash flows, as mentioned. Excel defaults to 0.1 (i.e. 10%) as the initial IRR guess. Bonus points if you allow the caller to specify the initial guess and desired number of iterations along with the cash flow sequence. Note that PRESENT-VALUE expects end-of-period payments/returns in its sequence and at fixed time intervals (daily, weekly, monthly, etc.) If you use PRESENT-VALUE you'll have to choose the finest time interval that makes sense -- but no finer -- and then pad with zeros for any time periods when there's no cash flow. From the looks of it you'll also have to adjust the IRR value accordingly, with any compounding adjustments applied when you convert to different period rates. (The PRESENT-VALUE function assumes the time basis for the IRR value and cash flow sequence match. So if each payment or return is daily, then the IRR will be a daily rate, too.) Microsoft's relevant documentation on XIRR seems to be here: http://office.microsoft.com/en-us/excel/HP052093411033.aspx SAS immediately came to mind as one package of financial functions that might already be available to you and other z/OSers; SAS (base) has its IRR and INTRR functions. (One returns the value in decimal format, and the other in percentage format, but they're otherwise identical. I think INTRR is the decimal one.) There's a SAS-L list where you could probably find out more about that option, or you can check here around about page 639: http://support.sas.com/documentation/onlinedoc/91pdf/sasdoc_913/base_lrdictionary_10307.pdf There are all kinds of other options, probably including a few from my employer that I haven't been imaginative enough to find yet. If you go to Google and type internal rate of return and source code as your search terms you can get lots of examples in many different programming languages. COBOL can call just about any programming language you wish, so feel free to pick your favorite, which might also be COBOL. Here's one in Perl, to pick a semi-wild example: http://sourceforge.net/projects/xirr4perl/ Perl for z/OS is available as a no charge download. Even if your favorite language isn't available it'll probably be rather easy to reimplement from one language into another, so just about any sample should do. By now somebody might have sent you a sample piece of code. But if not, another place you might want to ask is in the comp.lang.cobol newsgroup. This would be a great candidate to craft as a reusable business service, publish in a service directory, and let lots of developers enjoy at least within your organization. It'd be fun to turn this into a Web service, in particular. You might not be the first person in your organization to solve this problem, but you can be the last. :-) - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: About 1 in 5 IBM employees now in India - so what ?
Edward Jaffe writes: Econ-101. Increased U.S. exports in the wake of declining dollar is expected and should be no surprise to anyone. And, with the exception of Poughkeepsie-made mainframes, the exports you listed are not IBM's product. The year 2007 hasn't ended yet, so it's only possible to compare 2006 versus 2005 (and prior) for whole year comparisons. The U.S. export growth I cited predated recent declines in the U.S. dollar. You are correct in the sense that the declining dollar should boost exports, but it should boost them above the already impressive growth baseline, ceteris paribus. I think it should have been obvious that my parentheticals were examples and not exhaustive lists. And I think I only listed hard goods. Software is another category I could have listed. If IBM sells DB2 for z/OS, for example, the primary beneficiaries are developers living in the vicinity of Santa Teresa, California, although there are certainly other beneficiaries. I do actually try to think about life and economic impacts beyond IBM and beyond the industry I belong to professionally. If the great and growing Indian middle class is demanding more Boeing airplanes, Caterpillar construction equipment (to build infrastructure they want), and CNH plows to grow their food, and (hypothetically) 100 jobs get generated in those industries for every 50 that IBM transfers, that's a good deal, right? Regardless -- and brace yourselves everybody :-) -- I think it's a mistake to view trade with India strictly in terms of a single industry or single company, even if it is IBM. They want you to believe the shift is due to new hiring in hot markets. But, in addition to a 20% workforce shift to India, IBM has exported many jobs to Brazil (the Poughkeepsie operations staff, their sysprogs, their software ordering center, etc. all gone) and Communist China (info development -- even z/OS doc -- is now there, among others). As I mentioned before, I grew up in Connecticut, and my grandparents used to work in the button factory. That button factory is long gone: the work transferred to the southern United States. (It's now in Asia probably.) I expect they or their friends had some choice words for Carolinians or Georgians. :-) But now Connecticut makes helicopters, and insurance, and jet engines, and defense-related products, and aviation lighting systems, and cigar wrapper tobacco leaves, and deep sea exploration and recovery, and professional broadcasters, and Yale degrees, and new Broadway-bound musicals, and lots of other products and services. Nowadays Connecticut typically ranks #1 in per capita income among the 50 U.S. states. Good? Bad? Disruptive, for sure, but probably good. (Connecticut also has more forest than it did 100 years ago, because most agriculture moved elsewhere, mostly to the Midwest.) A large number of the employees IBM is hiring in places like India and China are selling and servicing products IBM exports into those markets. Not all, but many. IBM sells products and services that are most analogous to construction equipment, aircraft, and power transmission systems. IBM supplies capital goods to build core corporate and national infrastructure. India and China are buying lots of that sort of stuff as they grow, to keep growing. IBM is actively engaged in a policy of firing whole divisions of extremely capable workers from the U.S. and Europe -- the very countries to whom they owe their great success -- and replacing them with entry level trainees from the so-called BRIC countries. I have no idea if what you say is true. I suspect it's true to a lesser degree. The unfortunate fact is that capable doesn't mean valuable, but that's not a new phenomenon. Labor markets value workers in very strange ways. I happen to think they grossly undervalue school teachers, for example. I really wonder what readers from India would think about what you're saying and whether they would have different views. I suspect they would. Same with Boeing workers busy filling orders for Indian airlines. Also, somebody added so what? to the subject line. Is this topic devolving into a discussion that has nothing to do with what people might think of as IBM-MAIN topics? - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Java Problem Analysis
John McKown asks: Since SUN is now open sourcing Java, I wonder if IBM will start supplying the source to their port. I don't know the answer directly. However, note that GPL (GNU Public License) code provides access to source code. So if IBM or anyone else modifies someone else's GPL code, you can ask for and must receive the modified source code. And if it isn't modified, you already can get it. I don't know what Sun's license is. But at least for GPL code, you already have an answer. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
PR: IBM and ACI Forge Global Strategic Alliance (BASE24-eps)
There's an interesting System z-related announcement today in which IBM and ACI Worldwide broaden and deepen their relationship: http://www-03.ibm.com/press/us/en/pressrelease/23002.wss ACI Worldwide is best known for their popular BASE24-eps electronic payments/ATM solution. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: About 1 in 5 IBM employees now in India - so what ?
U.S. exports to India are increasing at about 25+ percent per year, or over 2.5 times faster than India's GDP growth, last I checked. Major export categories include engineering goods and machinery (e.g. construction equipment, agricultural equipment), electrical and electronic machinery (e.g. power generation and transmission equipment, Poughkeepsie-made mainframes), precious stones and metals, optical and medical instruments, and aviation-related products, among others. There are a lot of jobs getting created in those industries, I would imagine. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: PR: Bunch of IBM Software Announcements This Week
I wrote: So it is! I spoke too soon then. Congratulations, Apple and Mac OS X: you beat z/OS by 7 days it appears. :-) and it appears Apple didn't. The update is called Java for Mac OS X 10.4, Release 6. It's available here: http://www.apple.com/downloads/macosx/apple/macosx_updates/javaformacosx104release6.html But this isn't Java 6 (or Java 1.6 if you prefer). Release 6 in this case is merely updating J2SE 5.0 to version 1.5.0_13 and Java 1.4 to version 1.4.2_16. It's what IBM might call a PTF package or service update to the Java 5 and Java 4 versions. (Another clue is that the release notes say that Leopard, a.k.a. Mac OS X 10.5, already has these updates. Mac OS X 10.5 did not ship with Java 6.) Apple still has a few days to get Java 6 out the door before IBM does. Stay tuned - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: The future of PDSs
Peter Hunkeler writes: Isn't it about time for one more name change since z/OS is approaching release 10? Any nominations? :-) - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: PR: Bunch of IBM Software Announcements This Week
Paul Gilmartin writes: One interesting bit of trivia: Java 6 is not yet available for the Apple Macintosh. Available in today's Software Update for OS X 10.4/10.5. So it is! I spoke too soon then. Congratulations, Apple and Mac OS X: you beat z/OS by 7 days it appears. :-) - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: NYSE Undertakes IBM Mainframe Migration to Unix and Linux
Dave Kopischke writes: 600 servers to replace 1 1600 MIPS mainframe ??? It reads like there's more to it than this, but it's still incredible. The annual power bill alone could probably buy a new z9. I can only go by what's in the article -- I have no information on this situation. The article does not say that 600 servers replaced one 1,600 MIPS mainframe. It does not say the mainframe is actually replaced. In fact, the article suggests the 600 servers weren't the only ones intended to replace the mainframe, and it suggests there are actually more than 600 servers involved, perhaps many more. The article also suggests that the New York Stock Exchange has run out of data center space, power, and/or cooling, and they are looking for yet another data center location. Last I checked adding an extra data center isn't free, particularly for a stock exchange. The article is very thin on even asserted business benefits. There are few if any assertions about service qualities. There's nothing mentioned about how the NYSE would recover their thousands of servers in a disaster, and what business interruption or data loss they would suffer (and for how long) in the event of a disaster. There is nothing said about energy consumption (except oblique suggestions that it's a major problem). And I found this IT-related assertion incredible: We have only a handful of underutilized servers But that's my read just looking at the words on the screen, so you all might have seen something different in the article and/or have better information than I do. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: how much difference is between deploying a j2ee application in Z/Os and in windows?
Rob Wunderlich writes: z/OS DB2 tends to lag behind Windows in function. Check which version you are running on z/OS compared to the Windows version. Just to expand on what Rob is saying, if you're running an older version of DB2 for z/OS, then Rob's concerns likely apply. However, a lot has changed concerning DB2 for z/OS in the past couple versions. If you draw a Venn diagram, there's mostly overlap, but the area outside DB2 z/OS (and inside DB2 Linux/UNIX/Windows) has gotten smaller (and continues in that direction), and the area outside DB2 LUW (and inside DB2 z/OS) has been getting a bit bigger lately. There are a lot of reasons for these trends, but that's a topic for another thread. There's a great bit of documentation, SQL Reference for Cross-Platform Development, which I would recommend for developer audiences. You can find it here: http://www.ibm.com/developerworks/db2/library/techarticle/0206sqlref/0206sqlref.html Hopefully this book will get updated for DB2 Version 9 very soon, but the DB2 V8 guide is quite valuable. I think every developer who writes code that accesses databases should have a copy of this book close by. As a general rule with WebSphere applications, it's not too hard to redirect them from using DB2 V8 LUW (or earlier) to using DB2 V8 for z/OS. Or V9 to V9 (and often even V9 to V8). Yes, it's something you want to think about and plan, but, in general, it's not a particular challenge. Actually, I think in an earlier thread I alluded to the fact that mainframes measure and monitor, and that alone can be a shock for developers with code that hasn't been previously measured and monitored. If the code and/or database-related logic is inefficient, you can still run it, but the mainframe will tell you exactly where it is inefficient. And that alone can be a blow to many developer egos. :-) Just make sure you understand and prepare for the psychology involved there. :-) - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
PR: Bunch of IBM Software Announcements This Week
Here are some items that I found interesting. 1. WebSphere Dashboard Framework, already available for z/OS, now also available for Linux on z: http://www.ibm.com/common/ssi/rep_ca/9/897/ENUS207-329/ENUS207329.PDF Dashboard Framework lets you build attractive and useful Web dashboards, e.g. management presentation views, quickly and easily. There's business blogging support for making annotations, you can integrate dashboard data with spreadsheets (including .ods Open Document Format), publish alerts, etc. There's a recorded demo here: http://demos.dfw.ibm.com/on_demand/Demo/IBM_Demo_WebSphere_Dashboard_Framework-Mar07.html?S=indexS_CMP=rnav Yes, you read that right, Dashboard Framework runs on z/OS and Linux on z. I guess you'd say this is 2008's SDF III. :-) Along similar lines, Lotus ActiveInsight Version 6.0 is now available for Linux on z: http://www.ibm.com/common/ssi/rep_ca/1/897/ENUS207-331/ENUS207331.PDF This software helps businesses focus their employees on managing goals and objectives, by presenting active scorecards in dashboards. You can arrange for a demonstration here: http://docs.dfw.ibm.com/activeinsight/ 2. The IBM SDK for z/OS, Java Technology Edition, V6 -- Java 6 in shorthand -- is now available as a no charge feature. Both 31-bit and 64-bit are available, and JZOS is included. If you'd like to find out what's new in the Java 6 specifications, here's one place to look: http://java.sun.com/javase/6/ Watch here for details on electronic downloads: http://www-03.ibm.com/servers/eserver/zseries/software/java/ Of course you can keep previous JDK versions installed on your system and use them when and where needed. One interesting bit of trivia: Java 6 is not yet available for the Apple Macintosh. 3. A bunch of Information Management products got announced: DB2 Audit Management Expert for z/OS V2.1, WebSphere Replication Server for z/OS V9.1, WebSphere Data Event Publisher for z/OS V9.1 PTF, DB2 SQL Performance Analyzer for z/OS V3.2, IMS Buffer Pool Analyzer for z/OS V1.2, DB2 Log Analysis Tool for z/OS V3.1, IMS Recovery Expert for z/OS V1.1, DB2 Accessories Suite for z/OS V1.2, and IMS High Performance Fast Path Utilities for z/OS V3.2. IMS Recovery Expert is a brand new product and does what the name suggests. If you had WebSphere Data Integration Connector for z/OS you get WebSphere Data Event Publisher for z/OS, if you have active subscription support. (That goes for the other products, too, if you have an older version today.) 4. There are some withdrawal announcements that might be important. NALC pricing for z/OS is being withdrawn effective March 31, 2008, for new customers anyway. zNALC replaces NALC. NALC and now zNALC are specially priced z/OS licenses for new workload. It looks like zNALC, which also replaces z/OS.e, is a much better offering. The details on zNALC are here: http://www-03.ibm.com/servers/eserver/zseries/swprice/znalc.html The Service Flow Feature for CICS Transaction Server V3.1 will be withdrawn on June 30, 2008. This is a no charge feature; get your orders in before the deadline. (Also includes some rights to Rational Developer for System z; read the license terms in the announcement for details.) Better yet, upgrade to CICS Transaction Server V3.2 and its Service Flow Feature. The Service Flow Feature lets you create higher level flow abstractions to execute multiple CICS transactions in sequence (or in parallel), as microflows. You can create these flows without programming. It's quite useful as a building block for service-oriented architectures and for better aligning your CICS programs, on an ongoing basis, with changing business needs. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: how much difference is between deploying a j2ee application in Z/Os and in windows?
I concur with Larry: deploying J2EE applications to WebSphere Application Server for z/OS is the same as, say, WebSphere Application Server for Windows. There are multiple ways to do it, but typically you'd use the Web-based WebSphere Administrative console install wizard just as you would on any other platform. The WebSphere Information Center has details: http://publib.boulder.ibm.com/infocenter/wasinfo/v6r1/index.jsp I'm trying to think of any exceptions, and the only ones I can think of are very minor indeed. For example, you won't ever see a Windows-style server drive letter (e.g. C:\MyServer\AppsOnMyServer\...) in the Web panels. I think the Administrative console says WebSphere Application Server for z/OS for the branding. Here and there you might see an extra field or option, to provide additional controls available for z/OS, but if you don't know what they are just take the defaults. Administrative access control will be validated with whatever WAS z/OS security system you're using, typically the z/OS Security Server (RACF), rather than a Windows sign-on. You can also use the remote deployment feature in, for example, Rational Developer for System z or Rational Application Developer. This is how your J2EE developers would normally do things when they want to deploy an application to a remote server. Your developers will have absolutely no problem with this, other than perhaps the mental adjustment when they realize how easy and identical it is to deploy to the big iron. The WebSphere product developers really did a beautiful job. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z800 to z9
Patty, With whomever is selling you the z9 BC, try it. See if you can reach an agreement to test two or three different configurations (maybe 1, 2, and 3 CP configurations centered around the capacity level you're aiming for). For example, see if you can get a Capacity On Demand contract in place to let you bump up. I think (if I'm remembering correctly) you have a 2066-001 (z800 with one full speed CP). That's 32 MSUs, by the way. The equivalent 1-way z9 BC is around a 2096-R01 (27 MSUs). (Yes, the MSUs are lower for the same machine capacity.) But I'm assuming you're going to increase the capacity a bit since you're pegging the system for several hours, so the next 1-way is an S01 (30 MSUs). The S01 would have about 12 or 13 percent more capacity than your current z800, typically. But you can also try the N02 (2-way), which is also 30 MSUs. On my chart it looks like the S01 is an upgrade from the N02 (I think), so you can buy the N02 plus a priced option to upgrade to an S01 (and beyond, probably). And see how that goes, assuming my guess about where you want the capacity to be is correct. The N02 v. S01 looks like it'd be as perfect a test for you as possible. I suspect most sellers would be happy to arrange something like this under reasonable terms. Well, that's an educated guess anyway. I would also guess that grow is good and shrink is bad, so you'd want to buy the smaller machine first, test it, use On-Off Capacity On Demand to switch to the other bigger configuration, test that, then turn the capacity back off if you prefer the first one. So in my example I think that's N02 first and S01 second, but that could vary depending on what point you choose. (That's also assuming you can morph an N02 to an S01 via COD -- I'm not thoroughly familiar with all those permutations. But I think that works.) If I'm right, though, this is going to be very easy for you to validate in your exact real world conditions. It sounds like the IBM-MAINers think an N-way could be very useful to you, so I'd strongly consider going through this sort of validation exercise. I think the standard OOCOD contract language is all you'd need. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Dumb Debugginging on the Mainframe
This part of the linked article made me laugh out loud: Years ago, with procedural languages, we used things like loops, function calls, arrays, linked lists, queues and stacks. Now, in the 21st century and with the advent of object-oriented programming, we've evolved to using loops, function (sorry, method) calls, arrays, linked lists, queues and stacks. :-) - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Data Center Theft
Ed Gould writes: Please lets us know that the DS6000 is capable of remote mirroring and that it can indeed be done. It is, and it can, yes. It does support Metro Mirror/PPRC and Global Mirror/XRC. You need two (or more) of them for mirroring of course. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Data Center Theft
In fairness, the DS6000 is physically relatively small, although I wouldn't want to carry one by myself on my bicycle. The spindles (individual drives) are even smaller, but you'd need a number of them to have a RAID set and the complete data. Tough but not impossible. I think the IT marketplace is in for a shock when people figure out that losing the keys means losing the data. It isn't like a bank vault where you can hire a locksmith to drill some holes over several days. It's so critical to store and manage the encryption keys in a safe, secure, recoverable repository. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z/800 to z9 upgrade questions
Eric Bielefeld writes: Thanks for explaining what an ELA is. I was wondering what the cool new stuff that IBM offers. I assume its stuff like Websphere? Sure. Any one-time charge (OTC) IBM software product(s). Actually, they don't have to be cool or new, but that's what I'd choose. Some random examples: WebSphere Process Server for z/OS WebSphere Commerce Server for Linux on z WebSphere Message Broker for z/OS WebSphere Portal Enable for z/OS Lotus Domino for z/OS Lotus Domino for Linux on z Tivoli Composite Application Manager for WebSphere on z/OS Tivoli OMEGAMON XE for CICS Transaction Gateway on z/OS Tivoli Access Manager for e-business for Linux on z Tivoli zSecure Admin for z/OS Rational Developer for System z Rational Performance Tester for z/OS Rational ClearCase Enterprise Change Management for Linux on z Application Time Facility for z/OS ISPF Productivity Tool for z/OS DataQuant for z/OS DB2 Connect for Linux on z IMS Parameter Manager for z/OS WebSphere Classic Federation Server for z/OS ...plus many others. There's a reasonably complete list here, assuming you exclude the MLC products: http://www.ibm.com/software/sw-atoz/systemz/ - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Single signon with AD and kdc on z/OS.
Just out of curiosity, what's on your mainframe? Should I assume grades, transcripts, tuition billing, and degree records, among other things? OK, to answer your question directly, and if I'm interpreting it correctly, you're probably asking about what the IT market generally describes as Enterprise Single Sign-On (ESSO). So if you do Google searches and the like on ESSO you'll probably find several vendors out there. The one I work for has a product called Tivoli Access Manager for Enterprise Single Sign-On. From there you might branch off into areas such as directory integration, identity management, and/or Web single sign-on, among others. That'd be Tivoli Directory Integrator for z/OS, Tivoli Identity Manager for z/OS, and Tivoli Access Manger for e-business, respectively, as examples. There are lots of vendors in this market area, including CA's eTrust products and numerous smaller companies. Gartner lists some of them here: http://mediaproducts.gartner.com/reprints/citrix/vol3/article3/article3.html - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z/800 to z9 upgrade questions
their applications more easily to the rest of the world, or to tighten security compliance and protect their company's information better, or to move some functions to save data center space/power/cooling, or whatever. Glad to hear IBM is offering such a contract but I would have to know a lot more (and understand it ) before giving any blessing. As always, read the fine print, but I just gave you the core essence (and mutual rationale) for these sorts of agreements. IMHO they are underutilized, at least by IBM's customers. I should also say that occasionally an individual sales rep might not totally like an ELA. If you rewind to what I said above, you might guess why. That's generally not true, but there are always a few odd people. :-) I do think customers should challenge IBM (and other vendors) to think and act more strategically in these mutual relationships. Also, I should say there are several three letter acronyms in the IBM lexicon here, and I don't begin to understand them all in detail. ELA, SRO, OIO those are all in the same general category. (What are types of IBM contracts, Alex?) Which leads us to the age old complaint about IBM Sales types and their lack of interaction with the CIO's (or appropriate manager types). Amen. I've seen some outstanding sales reps at work and also some dreadful ones. I don't think the dreadful ones last long. Now its too bad that other vendors don't offer similar contracts. I think they do, at least to some degree. IBM is perhaps a bit unusual in having such a large number of part numbers and service offerings in its full catalog, so the contract scope can be quite broad. For example, if you can't find an OTC software product you like among the 13,000+ (a guess), you ain't trying. :-) I know you can't speak for OEM but they are in it to get all the money they can possibly squeeze out of the rest of us. While profit is good costs are one of the major factors that the MF is going belly up. I would name names but everyone knows who the biggest offender is so make up your you own name. There are more alternatives, thank goodness. I know IBM has done a lot of work in this area, starting especially in the year 2000. There is certainly a cost perception problem, but I've said before that we're in a different world now with radically different economics. I don't want to get side tracked by the evil empire but they aren't the only ones out there. Heck there used to be a DB2 vendor that at times made the other vendor look like like a mouse. Evil is a bad business strategy, at least if you want to stay in business very long. Google has that famous corporate motto, Don't Be Evil, and they're absolutely correct. What can IBM do to educate customers and keep them informed of offerings? Not once a year but 3-4 times a year (at least)? In my own little corner of the world, I offer some thoughts on the Mainframe Blog (http://mainframe.typepad.com), so that's one new way. The ranks of architects are growing, so that's another. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Single signon with AD and kdc on z/OS.
Would you expand on your question a bit, Steve. I assume you mean Kerberos KDC. Does AD mean Active Directory (Microsoft) in the context of your question? Basically, if you could draw a mental picture, that would be helpful. To take a guess at your question, I think there are a fair number of places where identity management is used to grant and revoke user credentials within the mainframe security systems. But very typically the mainframe security system remains responsible for authenticating and authorizing user access to various mainframe-hosted resources, and often to various resources outside the mainframe as well. (The boss security hub, basically.) There are myriad reasons for that, but fundamentally the mainframe-hosted resources are typically extremely high business value, and mainframe security systems are exceptionally, even uniquely, strong and well-proven, assuming competent management of course. So in most environments there's a push to use the mainframe as a security hub, extending its reach to encompass other resources (information, applications) in standards-compliant ways. There are various ways to do that, but I'll stop there in case I'm headed in the wrong direction with that background. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z/800 to z9 upgrade questions
Ed Gould writes: Reasonable advice. One thing I would like to toss in before you consider multi year contracts. *KNOW* your past usage and have a reasonable idea of anything coming down the pipe line (new applications) *BEFORE* getting into a multi year contract (with anyone). I have seen at least twice in my life time a company that did the multi year contract and really getting burned money wise because they were stuck in a contract. They did not want to hear about items that either the SEC was mandating or new applications coming down the pipeline that were really cpu intensive applications. The management was not talking to the other side of the house. I would suggest that you be careful about entering into long term contracts (we saw the cost of CPU's) tumbling as well as DASD cost tumbling. Sometime it makes sense to enter into long terms contracts, more often maybe not, IMO. I agree with this general principle, although my specific frustration was related to a specific situation. That is, you're buying (or upgrading to, or leasing) a new machine, ergo you can confidently predict you're going to run it at *some* baseline level for years. (Otherwise, why are you getting the machine?) Just tell IBM *Software* that, and seriously investigate an ELA (or other agreement which incorporates ELA-type terms) so you can quite probably get some cool new OTC for your System z. An ELA certainly doesn't have to coincide with a hardware purchase to make sense, but the situation I describe is *generally* a particularly easy case. (ISV machines like Mikos's are special cases, of course.) - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z/800 to z9 upgrade questions
Miklos Szigetvari writes: On the hardware side , as we are software development company , we need a SYSPLEX with different z/OS releases, and a strong CPU which could compile our sources in a acceptable time frame. We compile now in the z/800 about 18 hours, in the fastest Unix servers the same is about 2 hours. My guess is you've got a uniprocessor-friendly workload, but I'm curious to hear back what you discover. How often do you do recompiles for things like syntax checking? Conceivably you could see some benefit to using Rational Developer for System z to handle the code syntax checking right on the developer PC. Starting in V7.0 it supports C/C++. Also a shout out to the Dignus people here, which is an alternative or additional option. The code is 98% C++, and we would like to test the optimised code for the z9 machines . We deliver currently the code optimised for 9672. Here is als a question: how to deal with the possibilty to generate code for different hardware architecture levels. You can certainly do that, and there's been a lot of discussion on the list previously about the best ways to support different hardware architectures from the same product, from both support and technical points of view. There's also the interesting option of METAL C (in the z/OS 1.9 compiler and above), which might be of particular interest to an ISV like you. As general (and perhaps unsurprising) advice I'd say that you only need to support the newer instructions (even optionally) if they benefit your product. There are various hardware features you can use that do not depend on the presence of the actual hardware. One notable example is the zIIP. If you zIIP-enable your product (or part of your product), it still runs on non-zIIP-equipped hardware. I believe many (but not all) of the ICSF (cryptographic) services are like that, too, to pick another example. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z/800 to z9 upgrade questions
Miklos, In addition to Jim Marshall's excellent comments, I think you may wish to test capacity increases using the z9 BC. One standard way to do that is to get On/Off Capacity on Demand (OOCoD). That should let you test a higher processor capacity for as little as one day and pay only for that processor-day. I think you currently have a 2066-0C1 model. I have slightly different numbers than Jim for what would be a doubling of capacity, which I define as increasing the approximate transactional capacity measurement by at least 100%. I think you'd have these (ultimate) model choices in the z9 BC (2096-xxx): 1 CP: 2096-V01 - 42 MSUs full capacity 2 CPs: 2096-P02 - 41 MSUs full capacity 3 CPs: 2096-N03 - 43 MSUs full capacity 4 CPs: 2096-M04 - 45 MSUs full capacity Your current 2066-0C1 is 25 MSUs full capacity, so your MSUs will not double when you double capacity with a z9 BC. If you want to double your MSUs, the closest matches would be a 2096-Q02 or 2096-W01 (47 MSUs), or a 2096-O03 or 2096-R02 (52 MSUs). However, typically a -Q02 will provide about 134% more transactional capacity than your current system. Your IBM software charges should not double when your double your MSUs, even if you are licensed for full capacity. (Not even close.) There's no wrong answer here: how many CPs you should get depends on your workload(s). Usually a performance expert can look at your current system's workload (and trends) and predict, with a reasonable degree of confidence, which configuration makes the most sense. Or you can test. Or a little of both. Note that on the z9 BC you have 7 engines you can configure, and up to four of those can be CPs, so you'll want to keep the 7 and 4 numbers in mind as you plan for growth. If you're adding a lot of zAAPs, zIIPs, IFLs, or ICFs, then these numbers might be relevant. I totally agree with Jim that you only need to buy what you need. If you really are doubling, go for it. If not, estimate your growth and work with vendors to get sensible, strategic multi-year contracts. If that means someone wants to sell you something you don't need right away, that's fine and *might* even make sense, but just make sure it makes sense within your overall business case. As a slight digression, one of my personal ongoing frustrations is that a lot of customers buying a new mainframe (or upgrading to a new model) don't also pause briefly to (re)consider software contracts. Yes, you can (and do) pay for IBM software month-to-month (z/OS, z/VSE, DB2, etc.) But if you know you're going to pay for at least a certain amount of DB2, IMS, CICS, z/OS, MQ, etc. for two years, or three years, or whatever, then tell IBM (Software) that! If you're buying the machine, you're probably not buying to run it only one month. If you tell IBM about your probable MLC, then IBM can offer you an Enterprise License Agreement (ELA) or similar. And an ELA could get you access to cool new one-time charge (OTC) software for your System z. If anyone needs suggestions about what OTC products to pick in their particular situation, I'm full of suggestions. The rough analogy here is flying a particular airline regularly but refusing to give the airline your frequent flyer number. I don't know why anyone would do fail to do that, yet some people do. Another possible consideration is what to do with the z800. You can upgrade the z800 to the z9 BC and keep the serial number if you like. If it's your only machine -- no Parallel Sysplex -- then you will need to take a scheduled outage to do that, but many, many customers take that path. (Sounds like you might have Parallel Sysplex, though, although maybe logically within a single frame?) You can also get a brand new machine (new serial number) and sell the old one (or return it if the lease has ended). Or send the old one to my home, freight pre-paid, as a donation :-) Yet another possible option that may make sense is to move the z800 to a second location, put it on cold standby (no software charges), and use it as a disaster recovery machine. Then, whenever the next model comes out, get that model (within a reasonable period of time) and move the z9 BC over to replace the z800. From that point on you maintain an N-1 Cold Standby disaster recovery strategy. Whether this strategy makes sense or not depends on a number of factors, financial and otherwise. You may already have a disaster recovery strategy in place -- maybe even a better one. But I thought I'd mention it as it can often be a very good strategy. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http
Re: performance of db2 7 on mainframe
Luca, In addition to John's excellent questions, another question to ask is what WLM (Workload Manager) goal you've got set. If you tell z/OS Workload Manager that the DB2 user (you) should get a response within X.XXX seconds if possible, then WLM is going to try to meet or beat that goal consistently. That is not the same thing as servicing your request first with nothing else running on the system. If you tell WLM that 2 seconds is good enough, when you get 1.98 seconds you're getting what you told the system. The ability to do this, to instruct a server to deliver predictable performance across subsystems according to a service level agreement (SLA) goal, is probably unique to z/OS. It's a very useful business-oriented feature. And now you know what to ask. :-) Note that, if your application were also running on z/OS, it too would be embraced by WLM and managed according to the goal(s) you set. So, for example, if the application is at risk of taking too long to respond to a user, z/OS Workload Manager will try to allocate more system resources (such as CPU) to get that application to finish within the goal, if possible. That includes everything involved in fulfilling the request: z/OS TCP/IP, z/OS disk I/O, z/OS DB2 activity, and so on. Pretty cool, isn't it? I should also advise you that ODBC is the type of processing that can benefit from the zIIP (a specialty mainframe processor). I would recommend that the people you work with seriously investigate upgrading to DB2 8 (or better yet 9) and adding a zIIP, especially if there is current or planned non-trivial ODBC use. You are accessing DB2 via DRDA when you use ODBC, and much of the DRDA work can run on the zIIP instead of on the main processors. The zIIP could speed up your ODBC/JDBC access, or lower your DB2 CPU needs, or both. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Newspaper: IBM Invests $5M in Illinois State Univ.
Michael Stack writes: Mark, the statement I quoted, With the addition, ISU becomes the only public university in the state to offer its technology students both an undergraduate program in enterprise computing and hands-on access to such an IBM mainframe said nothing about z/Architecture. But this statement is 100% correct, and why shouldn't ISU market this exclusive? Good for them. If ISU has an advantage over other educational institutions, of course they should promote it, and their graduates should enjoy their near 100% placement rate. Said another way, if other institutions have less capability to educate their students, that's their problem (and possibly IBM's), certainly not ISU's. The key word is such. You're not going to see z/Architecture appear in a small city newspaper. As factually correct shorthand for the public at large, it's a fine word choice. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: OT: Outsourcing has its disadvantages
For what it's worth, I think this IBM-MAIN subject line choice is a bad one. First, the word disadvantages: sadly, such incidents are not unique to India. Unfortunately in a global comparison of crime rates the U.S., as an example, might not rank well. Second, the word outsourcing: a huge part of IBM's presence in India involves the myriad activities to sell and support its growing product and service businesses in India, including products and services produced outside India. The word exporting would be more accurate. At least the OT (off-topic) part is correct: the chosen subject line has nothing to do with the story. It's a simple story deserving a simple reaction. I hope we can all agree that this threat is reprehensible and move on. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Newspaper: IBM Invests $5M in Illinois State Univ.
IBM invests $5 million in ISU information technology program By Michele Steinbacher November 15, 2007 NORMAL — Illinois State University senior Joe Ksiazek is among those who say IBM’s estimated $5 million investment in the campus’s information technology program could open doors to more 21st century jobs for him and his classmates. “This offers students like me a great opportunity to learn new technologies being used in the business world,” said Ksiazek, a Chicagoan studying computer sciences. Article continues here: http://www.pantagraph.com/articles/2007/11/15/news/doc473be157569a9875622727.txt - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED]
