More IBM vs. Sun was The Register article on HP replacing z
http://www.crn.com/hardware/212500803;jsessionid=ULQY55VGCN2DCQSNDLRSKHSCJUNN2JVN (watch the link wrap, possibly.) Seb. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: The Register article on HP replacing z
Dredging up an old topic, but it was too funny to let go unnoticed. From SearchDataCenter today ANALYST GROUP DISSES HP REPORT ABOUT MAINFRAME MIGRATION Mark Fontecchio, News Writer It's not every day that an analyst group refutes vendor news that includes research from said analyst. Especially when it was research sponsored by the vendor -- in this case, Hewlett-Packard Co. It's a rare occurrence. But that's what happened recently in the midst of the ongoing battle between IBM selling its mainframes and everyone else selling mainframe migration. http://itknowledgeexchange.techtarget.com/mainframe-blog/analyst-group- disses-hewlett-packard-report-about-mainframe-migration/?track=NL- 576ad=676023USCAasrc=EM_NLN_5182355uid=279318 Mind the wrap... -- 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 Register article on HP replacing z
*heh-heh* we just started year 20 of our 5-year plan to get off of our mainframe. and I was just informed this week that our 4Q 2009 target for rolling our z890 out the door is now being pushed back to 1Q 2011! --Stephen Ken Gunther wrote: Rex, Sadly our goal of getting off the mainframe still persists. We are in YEAR 15 of an 18 MONTH project to accomplish this goal ;-}. Ken G. -- 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 Register article on HP replacing z
Check out the news item on www.trexxers.com on this. HP is overclaiming - there are a lot of mainframe displacements of HP too -Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Stephen Y Odo Sent: Wednesday, November 26, 2008 9:49 AM To: IBM-MAIN@bama.ua.edu Subject: Re: The Register article on HP replacing z *heh-heh* we just started year 20 of our 5-year plan to get off of our mainframe. and I was just informed this week that our 4Q 2009 target for rolling our z890 out the door is now being pushed back to 1Q 2011! --Stephen Ken Gunther wrote: Rex, Sadly our goal of getting off the mainframe still persists. We are in YEAR 15 of an 18 MONTH project to accomplish this goal ;-}. Ken G. -- 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 -- 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 Register article on HP replacing z
Rex, Sadly our goal of getting off the mainframe still persists. We are in YEAR 15 of an 18 MONTH project to accomplish this goal ;-}. Ken G. Pommier, Rex R. wrote: I concur also. I wonder what year my company was included in the 250 per 2 years statistic. We brought in a brand-spanking-new HP superdome back in 2001 as our mainframe (a 7060H50!!) killer. As per my CTO back then don't do any maintenance to the mainframe because it will be gone in 3 years. You can guess the rest. Our z9-BC is running happily; the superdome is in the process of being replaced by newer, smaller (physically) HP-UX boxes, and the get off the mainframe project has been shelved. Rex -- 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 Register article on HP replacing z
If they start talking layoffs, suggest an 18-month schedule... On Tue, Nov 25, 2008 at 8:26 PM, Ken Gunther [EMAIL PROTECTED] wrote: Rex, Sadly our goal of getting off the mainframe still persists. We are in YEAR 15 of an 18 MONTH project to accomplish this goal ;-}. Ken G. Pommier, Rex R. wrote: I concur also. I wonder what year my company was included in the 250 per 2 years statistic. We brought in a brand-spanking-new HP superdome back in 2001 as our mainframe (a 7060H50!!) killer. As per my CTO back then don't do any maintenance to the mainframe because it will be gone in 3 years. You can guess the rest. Our z9-BC is running happily; the superdome is in the process of being replaced by newer, smaller (physically) HP-UX boxes, and the get off the mainframe project has been shelved. Rex -- 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 -- 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 Register article on HP replacing z
On 19 Nov 2008 09:14:56 -0800, in bit.listserv.ibm-main you wrote: Interesting note in this article, Unfortunately, HP couldnt provide a single customer out of a reported 250 that could speak to me about their migration. That says a lot, you would figure these 250 (or at least some contigent thereof) would be going, hey this is great, and would be willing to talk to anyone and everyone about how wonderful their migration went. I contracted as a COBOL programmer analyst at a shop that as a part of their Y2K effort converted from OS390 to HP-UX using Uni-SPF and Uni-Kix. They converted all of the VSAM to Oracle. I cleaned up small pieces of debris left from that conversion. They also used a shell that could provide GDG function for sequential files. It seemed to work and be cheaper. I forget what the source management system was and exactly how security was done. They had a test server and one or two production servers. They were happy with it and working to get off the long in tooth CICS application. SYNCSORT for HP-UX made the mainframe version look brain dead. Actually the biggest threat to mainframes is the fact that most mainframe applications are in interesting shape and don't provide adequate help and documentation for the users. -- 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 Register article on HP replacing z
On Thu, 2008-11-20 at 11:05 -0400, Clark Morris wrote: SYNCSORT for HP-UX made the mainframe version look brain dead. How so? -- David Andrews A. Duda and Sons, Inc. [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 Register article on HP replacing z
Interesting note in this article, Unfortunately, HP couldnt provide a single customer out of a reported 250 that could speak to me about their migration. That says a lot, you would figure these 250 (or at least some contigent thereof) would be going, hey this is great, and would be willing to talk to anyone and everyone about how wonderful their migration went. -- 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 Register article on HP replacing z
Timothy Sipples wrote: Bill and I have fun (and ultimately academic) debates on this point. :-) My view: if it's easier to explain to the boss that a specialty engine is an accelerator, go right ahead. It's a reasonable way to think about its impact. If you Google zAAP (or zIIP) and accelerator you'll get plenty of hits. It is a method to cheat your boss. Usually I have to un-cheat management: He: Radoslaw, why don't you buy zIIP or zAAP? It will speed-up our production! Me: No sir, it won't. He: How is it possible? Do we have DB2? I KNOW WE DO! So, why don't you use DB2 accelerator? Me: No sir, we have DB2, but zIIP won't help us. He: What??? Do you want to say they lied? Do IBMers lie? Isn't zIIP for DB2? Me: Well, they didn't lie, BUT... ... It's a little bit easier with JAVA, because it's simpler to explain (or understand) that our JAVA apps run on other platforms. As a sum up specialty is a kind of euphemism. The word doesn't lie, but it suggests untrue features. -- Radoslaw Skorupka Lodz, Poland -- BRE Bank SA ul. Senatorska 18 00-950 Warszawa www.brebank.pl Sd Rejonowy dla m. st. Warszawy XII Wydzia Gospodarczy Krajowego Rejestru Sdowego, nr rejestru przedsibiorców KRS 025237 NIP: 526-021-50-88 Wedug stanu na dzie 01.01.2008 r. kapita zakadowy BRE Banku SA wynosi 118.642.672 zote i zosta w caoci wpacony. -- 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 Register article on HP replacing z
On Tue, 18 Nov 2008 15:17:59 +0900, Timothy Sipples [EMAIL PROTECTED] wrote: Bill and I have fun (and ultimately academic) debates on this point. :-) My view: if it's easier to explain to the boss that a specialty engine is an accelerator, go right ahead. It's a reasonable way to think about its impact. If you Google zAAP (or zIIP) and accelerator you'll get plenty of hits. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Based in Tokyo, Serving IBM Japan / Asia-Pacific E-Mail: [EMAIL PROTECTED] Jeez, Tim, I don't think it's much harder to just say Boss, let's use a zIIP/zAAP. It's cheaper. :-) Bill Seubert System z I/T Architect IBM Corp [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 Register article on HP replacing z
HP has been caught misquoting and misrepresenting the industry analyst they cited. HP's misrepresentation resulted in what rarely happens but just did happen: a very public press release and very noisy rebuke from the analyst: http://www.marketwire.com/press-release/Robert-Frances-Group-921499.html Doesn't HP realize that their credibility is at stake? What else did they misrepresent? - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Based in Tokyo, Serving IBM Japan / 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 Register article on HP replacing z
On Wed, 12 Nov 2008 12:42:31 -0500, Tony Harminc [EMAIL PROTECTED] wrote: OK, OK - you (and IBM) win! ... Tony, to the best of my knowledge, no one in IBM System z marketing, or for that matter, anyone with significant knowledge of the technical aspects of System z, has made official, public statements about specialty processors being features to boost performance. There may be a well-meaning sales rep or specialist or press person out there who does not have a full understanding of the hardware who have made that claim, but it should not have been an official IBM claim. As has already been stated, there is one notable exception to the Specialty engines are not performance enhancers rule - machines that run at subcapacity. If you have a box that doesn't run at the fully-rated uni speed, a specialty engine will provide better performance. There's one other performance benefit, but it is a roundabout way of claiming that the specialty engine provides improved performance - if one were to install a zIIP or zAAP and relieve the general purpose CP pool of a CPU bottleneck, then that would indirectly result in a performance benefit by offloading Java and/or other MIPS and relieving the constraint on the GPs. Thus you got cheaper MIPS with the zAAP/zIIP and fixed a performance bottleneck. But that's obviously stretching things... FWIW. Bill Seubert System z I/T Architect IBM Corp [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 Register article on HP replacing z
I usually stress that adding zIIPs and zAAPs provides a performance benefit for both of the reasons that Bill mentions below. The most significant for everyone is the second benefit (reducing the strain on the CPs). But you need to realize that the 'notable exception' is getting to be the rule. For example, the z10-BC zIIP/zAAP is the speed of a z01 (about 700 MIPS) for all 130 models, including the A01 (30 MIPS for a UP). If you were running DB2 on an A01, wouldn't you prefer to run DB2 on a 700 MIPS zIIP than a 30 MIPS CP? I think it's a no-brainer, especially since the cost of the specialty processors is lower than the regular CPs. Be sure to run WSC's zPCR to determine what you can expect to see in your installation. Cheryl On Nov 17, 2008, at 9:52 AM, Bill Seubert wrote: On Wed, 12 Nov 2008 12:42:31 -0500, Tony Harminc [EMAIL PROTECTED] wrote: OK, OK - you (and IBM) win! ... Tony, to the best of my knowledge, no one in IBM System z marketing, or for that matter, anyone with significant knowledge of the technical aspects of System z, has made official, public statements about specialty processors being features to boost performance. There may be a well-meaning sales rep or specialist or press person out there who does not have a full understanding of the hardware who have made that claim, but it should not have been an official IBM claim. As has already been stated, there is one notable exception to the Specialty engines are not performance enhancers rule - machines that run at subcapacity. If you have a box that doesn't run at the fully-rated uni speed, a specialty engine will provide better performance. There's one other performance benefit, but it is a roundabout way of claiming that the specialty engine provides improved performance - if one were to install a zIIP or zAAP and relieve the general purpose CP pool of a CPU bottleneck, then that would indirectly result in a performance benefit by offloading Java and/or other MIPS and relieving the constraint on the GPs. Thus you got cheaper MIPS with the zAAP/zIIP and fixed a performance bottleneck. But that's obviously stretching things... FWIW. Bill Seubert System z I/T Architect IBM Corp [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 -- 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 Register article on HP replacing z
I usually point out to my customers that adding zIIPs and zAAPs provides a performance benefit for both of the reasons that Bill mentions below. The most significant for everyone is the second benefit (reducing the strain on the CPs). But you need to realize that the 'notable exception' is getting to be the rule. For example, the z10-BC zIIP/zAAP is the speed of a z01 (about 700 MIPS) for all 130 models, including the A01 (30 MIPS for a UP). If you were running DB2 on an A01, wouldn't you prefer to run DB2 on a 700 MIPS zIIP than a 30 MIPS CP? I think it's a no-brainer, especially since the cost of the specialty processors is lower than the regular CPs. Be sure to run WSC's zPCR to determine what you can expect to see in your installation. Cheryl On Nov 17, 2008, at 9:52 AM, Bill Seubert wrote: Tony, to the best of my knowledge, no one in IBM System z marketing, or for that matter, anyone with significant knowledge of the technical aspects of System z, has made official, public statements about specialty processors being features to boost performance. There may be a well-meaning sales rep or specialist or press person out there who does not have a full understanding of the hardware who have made that claim, but it should not have been an official IBM claim. As has already been stated, there is one notable exception to the Specialty engines are not performance enhancers rule - machines that run at subcapacity. If you have a box that doesn't run at the fully-rated uni speed, a specialty engine will provide better performance. There's one other performance benefit, but it is a roundabout way of claiming that the specialty engine provides improved performance - if one were to install a zIIP or zAAP and relieve the general purpose CP pool of a CPU bottleneck, then that would indirectly result in a performance benefit by offloading Java and/or other MIPS and relieving the constraint on the GPs. Thus you got cheaper MIPS with the zAAP/zIIP and fixed a performance bottleneck. But that's obviously stretching things... FWIW. Bill Seubert System z I/T Architect IBM Corp [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 -- 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 Register article on HP replacing z
Bill and I have fun (and ultimately academic) debates on this point. :-) My view: if it's easier to explain to the boss that a specialty engine is an accelerator, go right ahead. It's a reasonable way to think about its impact. If you Google zAAP (or zIIP) and accelerator you'll get plenty of hits. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Based in Tokyo, Serving IBM Japan / 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 Register article on HP replacing z
I heard someone make the comment that even the z10 (and possibly earlier) were RISC with CISC in the milli-micro-etc. code? 668 of the 894 instructions on the z10 EC (about 75%) are implemented entirely in hardware. I don't know about anyone else, but I would not exactly call 668 hardware instructions RISC. I might well display my ignorance here but why the heck should I care? What I'd care for is functionality, performance, reliability, and price. I don't care whether this is reached with RISC, CISC, microcode, millicode or any level of combination thereof. -- Peter Hunkeler -- 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 Register article on HP replacing z
Well Said! Technology is only a benefit when properly applied! snip I might well display my ignorance here but why the heck should I care? What I'd care for is functionality, performance, reliability, and price. I don't care whether this is reached with RISC, CISC, microcode, millicode or any level of combination thereof. -- Peter Hunkeler /snip -- 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 Register article on HP replacing z
Hunkeler Peter (KIUK 3) wrote: 668 of the 894 instructions on the z10 EC (about 75%) are implemented entirely in hardware. I don't know about anyone else, but I would not exactly call 668 hardware instructions RISC. I might well display my ignorance here but why the heck should I care? What I'd care for is functionality, performance, reliability, and price. I don't care whether this is reached with RISC, CISC, microcode, millicode or any level of combination thereof. CISC is a programming language. RISC is a good reason to program in C or any another HLL. Most drivers don't care what's under the hood of their car. But, your mechanic cares... -- Edward E Jaffe Phoenix Software International, Inc 5200 W Century Blvd, Suite 800 Los Angeles, CA 90045 310-338-0400 x318 [EMAIL PROTECTED] http://www.phoenixsoftware.com/ -- 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 Register article on HP replacing z
On Thu, 13 Nov 2008 16:48:05 -0500, John Eells [EMAIL PROTECTED] wrote: ... I have _no_ idea where these rumors come from! 668 of the 894 instructions on the z10 EC (about 75%) are implemented entirely in hardware. I don't know about anyone else, but I would not exactly call 668 hardware instructions RISC. ... Ha! Only 668 of 'em. See? Reduced all the way from 894 to 668. That's a reduction of 226. Why, 266 is bigger that the whole instruction set of some processors. Let's see them make that kind of reduction! :-) Pat O'Keefe -- 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 Register article on HP replacing z
Edward Jaffe wrote: [...] That's why I translate specialty engines as castrated engines. The only specialization is lack of some abilities, just to support one workload, but not other. Everything behind is sales peach . Big technical effort done to keep high prices on monopoly (MVS) market and be make prices lower for competitive market (JAVA, Linux, etc.). Disclaimer: It's not rant on IBM. It just proper understanding of specialty engines and reasons why there are architected. Just observation. -- Radoslaw Skorupka Lodz, Poland -- BRE Bank SA ul. Senatorska 18 00-950 Warszawa www.brebank.pl Sd Rejonowy dla m. st. Warszawy XII Wydzia Gospodarczy Krajowego Rejestru Sdowego, nr rejestru przedsibiorców KRS 025237 NIP: 526-021-50-88 Wedug stanu na dzie 01.01.2008 r. kapita zakadowy BRE Banku SA wynosi 118.642.672 zote i zosta w caoci wpacony. -- 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 Register article on HP replacing z
On Thu, 13 Nov 2008 07:28:59 -0500, P S wrote: ... specialty engines help keep the cost of z/OS and friends high... They do? Not that I can tell. -- Tom Marchant -- 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 Register article on HP replacing z
Well, they do by helping z shops keep z/OS at all, sorta. It's kind of an nth derivative. I know what you're saying and I could argue that too, but look at it this way: without IFLs et al., more shops would say z/OS was too expensive, and IBM would either be losing more share or would be forced to lower the overall price. On Thu, Nov 13, 2008 at 8:12 AM, Tom Marchant [EMAIL PROTECTED] wrote: On Thu, 13 Nov 2008 07:28:59 -0500, P S wrote: ... specialty engines help keep the cost of z/OS and friends high... They do? Not that I can tell. -- 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 Register article on HP replacing z
P S wrote: You say that like it's a negative thing. No, I clearly mentioned, that it's NOT. See my disclaimer. I just provided more accurate description of things. Specialty really means less functional. Basic CP means all specialties in one. In English (same in Polish) specialty is understood as specially tuned for specific application, optimised for. While I understand marketing requirements, I don't see any reason to follow their language which is unfair. My opinion about: it is bad that MVS legacy world is so expensive, it is good that modern JAVA/Linux/whatever world is cheaper. (additional displaimer: my English is rather poor, so it is really hard to me to express all the nuances and differences I would like to express) Seriously, what's your point? Do you really think a Cadillac costs 2x as much to build as a Chevy? US-centric language. For me Chevy is Korean piece of sh*t (YES! Korean, former Daewoo), but still better than Polonez. Let me present more general example: 4-cyl. 2.0 dm3 engine is usually significantly more expensive than almost identical 1.6 dm3. Cost of production is quite similar. Do you really think a large Coke at McDonald's costs them a fraction of what they charge? I don't go to McDonalds (Wendy's, BurgerKing, other junkfoods), even if I'm in the U.S. And avoid Coca Cola at all. Do you really think that the manufacturing cost of the latest x86 chip drops significantly just because the next generation is released? Again, better example: cost of manufacturing Pentium 200MHz is the same as cost of manufacturing Pentium 75MHz (pleas help yourself and place your favorite current CPU name and frequency). Seems naive and needlessly negative to use such a term. ??? And while on the one hand it's true that specialty engines help keep the cost of z/OS and friends high, No. It is a way to sell cheap CP's for other-than-z/OS purposes. In other words everyone except legacy z/OS users will benefit. No, benefit is bad word. They probably wouldn't choose the platform otherwise. they also allow folks to make more use of same without paying for the MIPS on those specialty engines that they aren't using for z/OS. This is bad? On Thu, Nov 13, 2008 at 5:52 AM, R.S. [EMAIL PROTECTED] MY ADDRES!!! wrote: That's why I translate specialty engines as castrated engines. The only specialization is lack of some abilities, just to support one workload, but not other. Everything behind is sales peach . Big technical effort done to keep high prices on monopoly (MVS) market and be make prices lower for competitive market (JAVA, Linux, etc.). Regards -- Radoslaw Skorupka Lodz, Poland P.S. It is quite good manner to sign the message with the name. I'm pretty sure that neither your first nor last name is zoswork. -- BRE Bank SA ul. Senatorska 18 00-950 Warszawa www.brebank.pl Sd Rejonowy dla m. st. Warszawy XII Wydzia Gospodarczy Krajowego Rejestru Sdowego, nr rejestru przedsibiorców KRS 025237 NIP: 526-021-50-88 Wedug stanu na dzie 01.01.2008 r. kapita zakadowy BRE Banku SA wynosi 118.642.672 zote i zosta w caoci wpacony. -- 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 Register article on HP replacing z
You say that like it's a negative thing. Seriously, what's your point? Do you really think a Cadillac costs 2x as much to build as a Chevy? Do you really think a large Coke at McDonald's costs them a fraction of what they charge? Do you really think that the manufacturing cost of the latest x86 chip drops significantly just because the next generation is released? Seems naive and needlessly negative to use such a term. And while on the one hand it's true that specialty engines help keep the cost of z/OS and friends high, they also allow folks to make more use of same without paying for the MIPS on those specialty engines that they aren't using for z/OS. This is bad? On Thu, Nov 13, 2008 at 5:52 AM, R.S. [EMAIL PROTECTED] wrote: That's why I translate specialty engines as castrated engines. The only specialization is lack of some abilities, just to support one workload, but not other. Everything behind is sales peach . Big technical effort done to keep high prices on monopoly (MVS) market and be make prices lower for competitive market (JAVA, Linux, etc.). -- 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 Register article on HP replacing z
Tom Marchant wrote: On Thu, 13 Nov 2008 07:28:59 -0500, P S wrote: ... specialty engines help keep the cost of z/OS and friends high... They do? Not that I can tell. zAAP and zIIP specialty engines, which take over more and more work as more and more products are enhanced to exploit them, and the technology dividend, which makes the z10 27% less expensive from a software licensing perspective than z900, are elements of an overarching strategy to gradually *lower* the price of z/OS even as MIPS use is exploding. If the price was slashed overnight, the market would crash and IBM would lose too much money. (So would a lot of ISVs.) -- Edward E Jaffe Phoenix Software International, Inc 5200 W Century Blvd, Suite 800 Los Angeles, CA 90045 310-338-0400 x318 [EMAIL PROTECTED] http://www.phoenixsoftware.com/ -- 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 Register article on HP replacing z
Edward Jaffe wrote: [...] zAAP and zIIP specialty engines, which take over more and more work as more and more products are enhanced to exploit them, IMHO it is not enhancement. From financial point of view yes, but technically the product is not better than version working on regular CP. This is definitely some effort, like IBM's effort to create such specialies, but the word enhancement is not the best description IMHO. and the technology dividend, which makes the z10 27% less expensive from a software licensing perspective than z900, are elements of an overarching strategy to gradually *lower* the price of z/OS even as MIPS use is exploding. I really don't want to blame neither mainframe, nor IBM, but 27% is MSU base, not fee. It would be the same percentage if function fee=f(MSU) would be linear. It's not, of course it's growing (is proper math term?). So 27% MSU less means lower fees but not necessarily 27% less. AND IT'S VERY FINE that we have technology dividend. However we can also keep in mind that the divident makes second-hand machines less attractive. Second hand is nowadays the only IBM competitor. If the price was slashed overnight, the market would crash and IBM would lose too much money. (So would a lot of ISVs.) IBM yes, but ISV? I doubt. Imagine: due to lower costs of mainframe (and z/OS) more and more companies buy mainframe as new platform. That means more and more companies need ISV's products, ISV's hardware, services. That also could mean you sold not 1000, but 5000 licenses, so you can lower your prices ...no, not because you can or like customers, just to beat competition. Oh, sweet dreams, mainframes everywhere, everyone need my skills, people schedule my course in 6 months advance, employers compete for me, offer me flexible working hours, company cars, medical care... vbg -- Radoslaw Skorupka Lodz, Poland P.S. When talking about money, it is necessary to mention WLC. IMHO this is the greatest thing IBM did for mainframe pricing. -- BRE Bank SA ul. Senatorska 18 00-950 Warszawa www.brebank.pl Sd Rejonowy dla m. st. Warszawy XII Wydzia Gospodarczy Krajowego Rejestru Sdowego, nr rejestru przedsibiorców KRS 025237 NIP: 526-021-50-88 Wedug stanu na dzie 01.01.2008 r. kapita zakadowy BRE Banku SA wynosi 118.642.672 zote i zosta w caoci wpacony. -- 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 Register article on HP replacing z
On Thu, 2008-11-13 at 08:49 -0800, Edward Jaffe wrote: zAAP and zIIP specialty engines [...] are elements of an overarching strategy to gradually *lower* the price of z/OS even as MIPS use is exploding. If the price was slashed overnight, the market would crash and IBM would lose too much money. (So would a lot of ISVs.) Point taken, but a slash could be gradual (and more than the 10% MSU slide every three year hardware generation). It's easier, and cheaper by far, to reduce a price tag than to do the zXXP engineering. I have to conclude that IBM doesn't *want* more z/OS licenses. I don't understand that, but neither do I have access to the Big Picture. -- David Andrews A. Duda and Sons, Inc. [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 Register article on HP replacing z
On 13 Nov 2008 09:37:09 -0800, [EMAIL PROTECTED] (David Andrews) wrote: I have to conclude that IBM doesn't *want* more z/OS licenses. I don't understand that, but neither do I have access to the Big Picture. Wild conjecture can come up with some reasons - most of which don't bode well for our future. Does IBM want to sell xNix mainframes? Or are they orienting differently - to sell powerful database machines, powerful application servers, powerful web servers... ? Or would they be happy to be out of the computer hardware business altogether? When PCs became a commodity, they got out of that business. Do they foresee business computer hardware becoming a commodity? -- 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 Register article on HP replacing z
-Original Message- Howard Brazee Does IBM want to sell xNix mainframes? If they sell, yes! Or are they orienting differently - to sell powerful database machines, powerful application servers, powerful web servers... ? Isn't that what a z/OS box is? Or would they be happy to be out of the computer hardware business altogether? When PCs became a commodity, they got out of that business. Do they foresee business computer hardware becoming a commodity? PC's were a commodity long before they sold the business. The question is more of are the (estimated) 10,000 z/OS licenses (equals how many shops?) worth the continued investment in the 360-z architecture? The fact that the z10 was a 'new' platform shows they are/do consider it worthwhile. Of course it could just be a step towards emulating z/Architecture over another instruction set. I heard someone make the comment that even the z10 (and possibly earlier) were RISC with CISC in the milli-micro-etc. code? -- 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 Register article on HP replacing z
On 13 Nov 2008 12:21:31 -0800, [EMAIL PROTECTED] (Ken Porowski) wrote: Does IBM want to sell xNix mainframes? If they sell, yes! Do they think it will sell well enough? Or are they orienting differently - to sell powerful database machines, powerful application servers, powerful web servers... ? Isn't that what a z/OS box is? My orientation may fit better with marketing. Those are three different marketing directions.You and I aren't this market. -- 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 Register article on HP replacing z
Ken Porowski wrote: snip The fact that the z10 was a 'new' platform shows they are/do consider it worthwhile. Of course it could just be a step towards emulating z/Architecture over another instruction set. I heard someone make the comment that even the z10 (and possibly earlier) were RISC with CISC in the milli-micro-etc. code? snip I have _no_ idea where these rumors come from! 668 of the 894 instructions on the z10 EC (about 75%) are implemented entirely in hardware. I don't know about anyone else, but I would not exactly call 668 hardware instructions RISC. For more than you probably want to know, see: http://www.ibm.com/systems/resources/systems_z_news_announcement_pdf_ZSO03018.pdf http://www.redbooks.ibm.com/redpieces/pdfs/sg247515.pdf -- John Eells z/OS Technical Marketing IBM Poughkeepsie [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 Register article on HP replacing z
Can someone pass me the url on the HP article, pls, many thanks in advance Scott Ford Senior Systems Engineer [p] 678.266.3399 x304[m] 609-346-0399 identityforge.com This message is for the designated recipient only and may contain privileged, proprietary, or otherwise private information. If you have received it in error, please notify the sender immediately or let us know at [EMAIL PROTECTED] or [EMAIL PROTECTED], and then delete the original. Any other use of the email by you is prohibited. -Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of John Eells Sent: Thursday, November 13, 2008 4:48 PM To: IBM-MAIN@BAMA.UA.EDU Subject: Re: The Register article on HP replacing z Ken Porowski wrote: snip The fact that the z10 was a 'new' platform shows they are/do consider it worthwhile. Of course it could just be a step towards emulating z/Architecture over another instruction set. I heard someone make the comment that even the z10 (and possibly earlier) were RISC with CISC in the milli-micro-etc. code? snip I have _no_ idea where these rumors come from! 668 of the 894 instructions on the z10 EC (about 75%) are implemented entirely in hardware. I don't know about anyone else, but I would not exactly call 668 hardware instructions RISC. For more than you probably want to know, see: http://www.ibm.com/systems/resources/systems_z_news_announcement_pdf_ZSO0301 8.pdf http://www.redbooks.ibm.com/redpieces/pdfs/sg247515.pdf -- John Eells z/OS Technical Marketing IBM Poughkeepsie [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 -- 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 Register article on HP replacing z
Found where I read it (I did not believe it). I had already heard in other discussions/presentations on the z10 that they were indeed CISC chips. Probably someone assuming z6/Power 6 technology with RISC. Another 'The Register' article titled 'IBM chills mainframe New Coke' http://www.theregister.co.uk/2008/10/21/new_ibm_entry_mainframe/ The article states 'The z10 BC is a cut-down version of the existing z10 Enterprise Class machine, which launched in March 2008 using Big Blue's quad-core z6 CISC mainframe processor.' and continues to mention CISC. One of the comments from readers was CISC processors By Magellan Posted Tuesday 21st October 2008 22:28 GMT Mainframes have not used CISC processors in years. They execute CISC code, but on RISC processors using hardware CISC decoder which converts the CISC operations to RISC operations. This is exactly what AMD and Intel have done for years on x86, ever since AMD's K5 and Intel's P6. I recall reading somewhere there was significant similarity between one of the IBM mainframe RISC processor cores from the early 2000s and IBM's in-order RISC RS64 processor core from the late 1990s. -Original Message- John Eells Ken Porowski wrote: snip I heard someone make the comment that even the z10 (and possibly earlier) were RISC with CISC in the milli-micro-etc. code? snip I have _no_ idea where these rumors come from! 668 of the 894 instructions on the z10 EC (about 75%) are implemented entirely in hardware. I don't know about anyone else, but I would not exactly call 668 hardware instructions RISC. For more than you probably want to know, see: http://www.ibm.com/systems/resources/systems_z_news_announcement_pdf_ZSO 03018.pdf http://www.redbooks.ibm.com/redpieces/pdfs/sg247515.pdf -- John Eells z/OS Technical Marketing IBM Poughkeepsie [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 Register article on HP replacing z
David Andrews writes: It's easier, and cheaper by far, to reduce a price tag than to do the zXXP engineering. *A* price tag, maybe. I have to conclude that IBM doesn't *want* more z/OS licenses. I don't understand that, but neither do I have access to the Big Picture. Okay then. So that would be why IBM has reduced the price per MIPS for z/OS. And keeps increasing its function, adding more freebies every year. And of course that explains zNALC! (New doesn't actually mean new, and application doesn't actually mean application. That NA in the acronym is just a clever bit of misdirection.) And hired all those fresh faced 20-something z/OS developers. That Master the Mainframe contest? That's showing off Linux, isn't it? Why, with all these price reductions, no extra charge enhancements, its hiring spree, and its other nefarious activities, IBM is guaranteed to eliminate all z/OS demand in a mere 6 months. :-) Apologies, David. I shouldn't be sarcastic. But I just had to laugh out loud at your conclusion. :-) By the way, I read your conclusion during a coffee break as I was working on the technical bits of a final bid proposal to a potential new z/OS licensee. So I'm really having fun at your expense. :-) - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Based in Tokyo, Serving IBM Japan / 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 Register article on HP replacing z
Rahim, A. (Ahmed) wrote: IBM Maintains Lead in Server Market URL? -- Edward E Jaffe Phoenix Software International, Inc 5200 W Century Blvd, Suite 800 Los Angeles, CA 90045 310-338-0400 x318 [EMAIL PROTECTED] http://www.phoenixsoftware.com/ -- 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 Register article on HP replacing z
If the application mix and, more importantly the business needs, are such that a move to the HP system is viable, more power to them. All too often, the decision is being made by a PHB that is too busy reading airline magazines to take a realistic look at the costs and benefits of all available solutions. Dilbert is alive and well, and lurking in every manager's office, just waiting to strike. Let's face it, marketting representatives don't always tell the truth. They're like politicians in many ways. And the only time a politician tells the truth is when he calls his opponent a liar! :-) Rick Sebastian Welton wrote: On Tue, 11 Nov 2008 21:16:38 -0600, Rick Fochtman [EMAIL PROTECTED] wrote: Figures don't lie, but liars will figure. Smacks of management by airline magazine, with a generous dose of marketting Male Bovine Excrement. What about all those small VM, VSE, etc shops which were running on low end systems, such as P/390, MP*000, IS, etc. They haven't been able to move to FLEX for the past couple of years, so where did they move to? I know of quite a few that have or are contemplating moving to such systems. Seb. -- 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 -- 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 Register article on HP replacing z
Wow, I ususally just monitor this blog from time to time but that last paragraph says it all! -- Original message from Rick Fochtman [EMAIL PROTECTED]: -- If the application mix and, more importantly the business needs, are such that a move to the HP system is viable, more power to them. All too often, the decision is being made by a PHB that is too busy reading airline magazines to take a realistic look at the costs and benefits of all available solutions. Dilbert is alive and well, and lurking in every manager's office, just waiting to strike. Let's face it, marketting representatives don't always tell the truth. They're like politicians in many ways. And the only time a politician tells the truth is when he calls his opponent a liar! :-) Rick Sebastian Welton wrote: On Tue, 11 Nov 2008 21:16:38 -0600, Rick Fochtman wrote: Figures don't lie, but liars will figure. Smacks of management by airline magazine, with a generous dose of marketting Male Bovine Excrement. What about all those small VM, VSE, etc shops which were running on low end systems, such as P/390, MP*000, IS, etc. They haven't been able to move to FLEX for the past couple of years, so where did they move to? I know of quite a few that have or are contemplating moving to such systems. Seb. -- 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 -- 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 -- 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 Register article on HP replacing z
How about this article from IBM :- IBM Maintains Lead in Server Market Johannesburg - Wednesday, 12 November 2008 - IBM (NYSE: IBM) continues to rank number one in worldwide server revenue market share according to recently released reports by leading independent research firms, Gartner and IDC. Gartner reported that IBM had widened its lead over competition in the second quarter of 2008 capturing 31.2 percent of all server revenue worldwide. The research firm said IBM's server leadership was driven by Power Systems and System z. Power Systems captured 35.7 percent revenue share. All IBM brands in Gartner's over $250 000 server category combined to capture 59.7 percent revenue share. According to the IDC, IBM's share of the server market revenue in the second quarter stood at 33.2 percent - surpassing HP with a 5.8 percent gap. IBM also topped the industry with a 27 percent revenue growth in overall servers. The IDC also noted that IBM ranked number one across all UNIX servers, IDC's server category of $250 000 or more, and the $10 000 or more price point. The firm said IBM's $250 000 plus servers and IBM System z beat the competition with a 37 percent and 24.8 percent revenue share respectively taking 61.8 percent of industry's total revenue share. For IDC's category of $10 000 or more, IBM obtained 48.4 percent revenue share, whereas Gartner put IBM's share at 45.4 percent. IBM maintained the lead across all geographies, which puts us in a great position to continue creating value for our customers in the Sub Saharan Africa market, said Zoaib Hoosen, IBM's Director of Systems and Technology Group, in the Sub Saharan Africa region. IBM recently launched its next generation mainframe for mid-sized businesses, the IBM z10 Business Class (z10 BC) with the capacity of up to 232 x86 servers and 83 percent smaller footprint, and up to 93 percent lower energy costs. Companies in emerging markets (such as South Africa, Nigeria, Kenya, Angola and Tanzania, among others) or in hot industries (such as social networking or mobile commerce) - can now afford IBM's flagship mainframe technology for under $100,000 (about R1 million). Ends For more information: www.ibm.com For Media Enquiries: Nathi Sukazi IBM Communications +27 82 565 0205 +27 11 302 9442 [EMAIL PROTECTED] Regards, Ahmed Rahim System Z Technical support Group Technology (011) 500-6490 Fax 086 539 2113 -Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of John McKown Sent: 11 November 2008 06:17 PM To: IBM-MAIN@BAMA.UA.EDU Subject: The Register article on HP replacing z I don't know how accurate it is, but I found it interesting. http://www.theregister.co.uk/2008/11/11/hp_chases_mainframes/ -- John -- 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 Nedbank Limited Reg No 1951/09/06. The following link displays the names of the Nedbank Board of Directors and Company Secretary. [ http://www.nedbank.co.za/terms/DirectorsNedbank.htm ] This email is confidential and is intended for the addressee only. The following link will take you to Nedbank's legal notice. [ http://www.nedbank.co.za/terms/EmailDisclaimer.htm ] -- 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 Register article on HP replacing z
On Tue, 11 Nov 2008 10:29:32 -0800, Edward Jaffe [EMAIL PROTECTED] wrote: HPs statement seem to contradict IDC's trackers. But, it's possible both are correct. HP might be bottom feeding in the $250K marketplace. Another possibility is that Sun and other competitors might be in wholesale collapse in the $250K server marketplace, making IBM's overall market penetration higher. (The only problem with that theory is that IDC trackers also show HP shrinking in that space.) Last, but not least, HP could be flat out lying. We report. You decide... I think this article was very interesting but all these extra comments made me think. I just wonder if IDC or Ferrari was running a market study of the automotive industry and would conclude that Ferrari has 46 % of the world automotive industry for cars with an above than 2000 k$ price tag. Would we conclude that this is a tremendous result for Ferrari in the market share ? Or would we conclude that these Ferrari cars are too expensive ? Or that this kind of study and numbers is just meaningless? ( my favorite) OK i know there is a big effort from IBM to show that we are not talking about Ferrari's but about big freight train (z10 annoucements in Paris lately) but the logic of this kind of study is beyond my own logic. There is a simple rule in our industry, We use what we can afford to use according to criteria.(DR,TTR,availability,costs etc..), The TSM example few posts ago is showing exactly that . Some ( very few) use it on mainframe some (the majority) use it on Unix. Why ? The first ones generally are not night batch intensive and can afford the run without extra costs. The second because they are fully busy at night,would increase their MVS/RACF/etc MLC charges by using TSM at night. When you have a budget in your hand, the choice is very quickly made Is (UNIX+ TSM + Hardware P or else + robot +Maintenance) cheaper or more expensive than (TSM OTC + MLC increase + maintenance +K7 on existing robot) When you know that, you take a decision and don't give a damn about anything else. And please IBM spare me the song about the cost of people, the same person can run TSM on any platform, he will not even know it apart from the cartridge hardware! my 0.02 cents ( i had to take these kind of decisions ) Bruno Sugliani zxnetconsult(at)free(dot)fr http://zxnetconsult.free.fr -- 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 Register article on HP replacing z
2008/11/11 Edward Jaffe [EMAIL PROTECTED]: Tony Harminc wrote: ... The Reg has also bought into IBM's line about speciality engines being some sort of performance enhancers for Java, Linux, and so on, rather than just a scheme for maintaining market differentiation between Classic and New workloads. But, specialty engines *do* enhance performance of these workloads! Many users (us included) have sub-capacity CPs. Our current hardware is a 2096-L03. Each CP generates about 66 MIPS. The IFLs, zAAPs and zIIPs run at 470 MIPS -- or about seven times faster. We plan to get a z10 BC before the end of the year. On that machine, the specialty engines run at an astonishing 713 MIPS! OK, OK - you (and IBM) win! But you and most others here understand that the specialty engines are not different in any way from normal engines; indeed it's the normal engines that are artificially slowed down based on what you pay. There's nothing wrong with this, of course, except that people generally don't like price differentiation based on things like how they use a product, or some kind of perceived willingness to pay. It's no different from how airlines sell the same seat to business vs vacation travellers so that you can find you have paid 10 times as much as your seatmate, or why the business editions of Windows Vista cost more than the corresponding home editions, though there's almost certainly more code needed to implement the fancy media features in home than the backup and such in business. The Reg, and others who should be able to dig a little deeper, have accepted the implication that an IFL is somehow optimized or specialized for running Linux, a zAAP for running Java, and so on, when they're really all identical. Of course there's no reason IBM couldn't make true specialty engines, and it won't surprise anyone if they do at some point. And if that happens, doubtless IBM marketing will treat them as just an addition to the current line of innovative specials. Tony H. -- 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 Register article on HP replacing z
On Wed, 12 Nov 2008 12:42:31 -0500, Tony Harminc [EMAIL PROTECTED] wrote: snip The Reg, and others who should be able to dig a little deeper, have accepted the implication that an IFL is somehow optimized or specialized for running Linux, a zAAP for running Java, and so on, when they're really all identical. Of course there's no reason IBM couldn't make true specialty engines, and it won't surprise anyone if they do at some point. And if that happens, doubtless IBM marketing will treat them as just an addition to the current line of innovative specials. Tony H. There are such now. The crypto coprocessors are specialized hardware. Or, at least, they are not zArch engines with a special microcode load as the IFL, zIIP, and zAAP are. I could imagine, some day, a hardware engine which has the Java byte code interpreter as replaceable firmware. Maybe a CELL processor. Or maybe even a x86_64 board to run a headless version of Windows or Linux. -- John -- 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 Register article on HP replacing z
On Tue, 11 Nov 2008 21:16:38 -0600, Rick Fochtman [EMAIL PROTECTED] wrote: Figures don't lie, but liars will figure. Smacks of management by airline magazine, with a generous dose of marketting Male Bovine Excrement. What about all those small VM, VSE, etc shops which were running on low end systems, such as P/390, MP*000, IS, etc. They haven't been able to move to FLEX for the past couple of years, so where did they move to? I know of quite a few that have or are contemplating moving to such systems. Seb. -- 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 Register article on HP replacing z
Tony Harminc wrote: The Reg, and others who should be able to dig a little deeper, have accepted the implication that an IFL is somehow optimized or specialized for running Linux, a zAAP for running Java, and so on, when they're really all identical. Agreed 100%. There is no difference between the engines on full-capacity models. There is no special optimization of any kind. An engine is an engine. They run exactly the same instructions through exactly the same instruction pipeline in exactly the same way. -- Edward E Jaffe Phoenix Software International, Inc 5200 W Century Blvd, Suite 800 Los Angeles, CA 90045 310-338-0400 x318 [EMAIL PROTECTED] http://www.phoenixsoftware.com/ -- 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 Register article on HP replacing z
Edward Jaffe wrote: Agreed 100%. There is no difference between the engines on full-capacity models. There is no special optimization of any kind. An engine is an engine. They run exactly the same instructions through exactly the same instruction pipeline in exactly the same way. C'mon Ed, you know that ain't true - at least one exception already exists. Try IPL'ing z/OS native on one of your IFLs. Different [micro,milli,macro]-code and the fact that they are now pooled certainly allows for the prospect of such optimization without the users being any the wiser. Maybe in the future, maybe already. Shane ... -- 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 Register article on HP replacing z
Shane wrote: Edward Jaffe wrote: Agreed 100%. There is no difference between the engines on full-capacity models. There is no special optimization of any kind. An engine is an engine. They run exactly the same instructions through exactly the same instruction pipeline in exactly the same way. C'mon Ed, you know that ain't true - at least one exception already exists. Try IPL'ing z/OS native on one of your IFLs. OK. There is that one command, used early in z/OS IPL, that is deliberately disallowed on specialty engines to prevent users from accidentally IPLing z/OS on them. (z/VM can IPL no problem because it doesn't use that command.) But, my point is that there is no optimization that allows them to process Linux for z instruction streams, Java, or other specialty engine-eligible work any faster. -- Edward E Jaffe Phoenix Software International, Inc 5200 W Century Blvd, Suite 800 Los Angeles, CA 90045 310-338-0400 x318 [EMAIL PROTECTED] http://www.phoenixsoftware.com/ -- 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 Register article on HP replacing z
Wouldn't a specialty engine be excluded from the normal load sharing? By dispatching only its designated work and excluding other stuff (such as I/O interrupts) it ought to accomplish more of the targeted work than a general purpose engine. Is that reasonable? True, there need not be anything special about the engine proper, only in the way it is managed and exploited. But, then, that begins to infringe on how we define 'optimization'. -Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Edward Jaffe Sent: Wednesday, November 12, 2008 3:17 PM To: IBM-MAIN@BAMA.UA.EDU Subject: Re: The Register article on HP replacing z Shane wrote: Edward Jaffe wrote: Agreed 100%. There is no difference between the engines on full-capacity models. There is no special optimization of any kind. An engine is an engine. They run exactly the same instructions through exactly the same instruction pipeline in exactly the same way. C'mon Ed, you know that ain't true - at least one exception already exists. Try IPL'ing z/OS native on one of your IFLs. OK. There is that one command, used early in z/OS IPL, that is deliberately disallowed on specialty engines to prevent users from accidentally IPLing z/OS on them. (z/VM can IPL no problem because it doesn't use that command.) But, my point is that there is no optimization that allows them to process Linux for z instruction streams, Java, or other specialty engine-eligible work any faster. -- Edward E Jaffe Phoenix Software International, Inc 5200 W Century Blvd, Suite 800 Los Angeles, CA 90045 310-338-0400 x318 [EMAIL PROTECTED] http://www.phoenixsoftware.com/ -- 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 NOTICE: This electronic mail message and any files transmitted with it are intended exclusively for the individual or entity to which it is addressed. The message, together with any attachment, may contain confidential and/or privileged information. Any unauthorized review, use, printing, saving, copying, disclosure or distribution is strictly prohibited. If you have received this message in error, please immediately advise the sender by reply email and delete all copies. -- 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 Register article on HP replacing z
On 11/12/2008 at 4:33 PM, Hal Merritt [EMAIL PROTECTED] wrote: Wouldn't a specialty engine be excluded from the normal load sharing? By dispatching only its designated work and excluding other stuff (such as I/O interrupts) it ought to accomplish more of the targeted work than a general purpose engine. Is that reasonable? In the case of IFLs, all those things are run on the IFL. For zIIPs and zAAPs, that's a different story. Ed's right. There is one and only one instruction that is disabled on an IFL. Everything else is identical to a standard engine. Mark Post -- 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 Register article on HP replacing z
On Wed, 12 Nov 2008 13:17:09 -0800, Edward Jaffe [EMAIL PROTECTED] wrote: snip OK. There is that one command, used early in z/OS IPL, that is deliberately disallowed on specialty engines to prevent users from accidentally IPLing z/OS on them. (z/VM can IPL no problem because it doesn't use that command.) But, my point is that there is no optimization that allows them to process Linux for z instruction streams, Java, or other specialty engine-eligible work any faster. -- Edward E Jaffe I know that I won't be told, but I'll ask anyway. What is the instruction? Does it do anything other than blow up on an non-CP engine? I.e. does it have some side effect such that z/OS will fail if it is not issued? Is it only used once during IPL, or is it strategically placed in various modules so that it could not simply be zapped out? OK, I know that I could simply look at the Hercules emulation code to determine some of that. -- John -- 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 Register article on HP replacing z
Mark Post wrote: On 11/12/2008 at 4:33 PM, Hal Merritt [EMAIL PROTECTED] wrote: Wouldn't a specialty engine be excluded from the normal load sharing? By dispatching only its designated work and excluding other stuff (such as I/O interrupts) it ought to accomplish more of the targeted work than a general purpose engine. Is that reasonable? In the case of IFLs, all those things are run on the IFL. For zIIPs and zAAPs, that's a different story. Ed's right. There is one and only one instruction that is disabled on an IFL. Everything else is identical to a standard engine. Right. And, given enough of them, not all general CPs used by z/OS will field interrupts either. See the CPENABLE parameter. -- Edward E Jaffe Phoenix Software International, Inc 5200 W Century Blvd, Suite 800 Los Angeles, CA 90045 310-338-0400 x318 [EMAIL PROTECTED] http://www.phoenixsoftware.com/ -- 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 Register article on HP replacing z
On Tue, 11 Nov 2008 10:29:32 -0800, Edward Jaffe wrote: Dave Salt wrote: So if HP alone is converting more than 250 mainframe shops every 2 years, then at least 1,000 mainframe shops will be gone over the next 8 years? Ouch! Someone please tell me this isn't true? Let me state up front that I do not trust the Register. Never have... ... Last, but not least, HP could be flat out lying. We report. You decide... -- Here's another take on the same HP report... http://itknowledgeexchange.techtarget.com/mainframe-blog/hp-250-users- have-moved-from-mainframe-to-itanium/?track=NL- 576ad=673317USCAasrc=EM_NLN_4996661uid=279318 Mind the wrap. Interesting note in this article, Unfortunately, HP couldnt provide a single customer out of a reported 250 that could speak to me about their migration. -- 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 Register article on HP replacing z
Look out for the zHYBRID A system z able to run blade servers as well as IBM's General Purpose CP's Regards, Ahmed Rahim System Z Technical support Group Technology (011) 500-6490 Fax 086 539 2113 -Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of John McKown Sent: 12 November 2008 08:04 PM To: IBM-MAIN@BAMA.UA.EDU Subject: Re: The Register article on HP replacing z On Wed, 12 Nov 2008 12:42:31 -0500, Tony Harminc [EMAIL PROTECTED] wrote: snip The Reg, and others who should be able to dig a little deeper, have accepted the implication that an IFL is somehow optimized or specialized for running Linux, a zAAP for running Java, and so on, when they're really all identical. Of course there's no reason IBM couldn't make true specialty engines, and it won't surprise anyone if they do at some point. And if that happens, doubtless IBM marketing will treat them as just an addition to the current line of innovative specials. Tony H. There are such now. The crypto coprocessors are specialized hardware. Or, at least, they are not zArch engines with a special microcode load as the IFL, zIIP, and zAAP are. I could imagine, some day, a hardware engine which has the Java byte code interpreter as replaceable firmware. Maybe a CELL processor. Or maybe even a x86_64 board to run a headless version of Windows or Linux. -- John -- 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 Nedbank Limited Reg No 1951/09/06. The following link displays the names of the Nedbank Board of Directors and Company Secretary. [ http://www.nedbank.co.za/terms/DirectorsNedbank.htm ] This email is confidential and is intended for the addressee only. The following link will take you to Nedbank's legal notice. [ http://www.nedbank.co.za/terms/EmailDisclaimer.htm ] -- 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 Register article on HP replacing z
I don't know how accurate it is, but I found it interesting. http://www.theregister.co.uk/2008/11/11/hp_chases_mainframes/ -- John -- 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 Register article on HP replacing z
An interesting article indeed. For those who haven't read it, here are some statements that caught my eye: - HP announced today that in the past two years, it has helped more than 250 customers worldwide to migrate from mainframes to Integrity-based servers. - HP has been chasing IBM and other mainframes for 25 years, and ... has averaged around 100 per year for a while, but is accelerating - While IBM has specialty engines on mainframes that lower the cost of supporting Linux, Java, and DB2 workloads compared to regular mainframe engines, the lower price still leaves a big gap, even on the new z10 BC boxes. Even when you move down to a Linux specialty engine, it is an order of magnitude more expensive than Linux on an Integrity machines - HP contends that customers can save up to 70 per cent by getting off the mainframe So if HP alone is converting more than 250 mainframe shops every 2 years, then at least 1,000 mainframe shops will be gone over the next 8 years? Ouch! Someone please tell me this isn't true? Dave Salt SimpList(tm) - try it; you'll get it! http://www.mackinney.com/products/SIM/simplist.htm Date: Tue, 11 Nov 2008 10:16:42 -0600 From: [EMAIL PROTECTED] Subject: The Register article on HP replacing z To: IBM-MAIN@BAMA.UA.EDU I don't know how accurate it is, but I found it interesting. http://www.theregister.co.uk/2008/11/11/hp_chases_mainframes/ -- John -- 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 _ -- 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 Register article on HP replacing z
So if HP alone is converting more than 250 mainframe shops every 2 years, then at least 1,000 mainframe shops will be gone over the next 8 years? Ouch! Someone please tell me this isn't true? Dave Salt hmmm, sounds like more management by airport magazine hype. Don't believe everything you read. -- 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 Register article on HP replacing z
Dave Salt wrote: So if HP alone is converting more than 250 mainframe shops every 2 years, then at least 1,000 mainframe shops will be gone over the next 8 years? Ouch! Someone please tell me this isn't true? Let me state up front that I do not trust the Register. Never have... The article quotes statements from HP and not from an independent source. Most analysts trust the IDC trackers http://www.idc.com/prodserv/trackers.jsp to monitor market movement. Over the past several years, IBM has been citing the IDC Quarterly Tracker to show that System z has been consistently and significantly gaining market share in the $250K server marketplace. That's not a measure of MIPS shipped or anything like that. It's a measure of actual percentage of total market ownership. HPs statement seem to contradict IDC's trackers. But, it's possible both are correct. HP might be bottom feeding in the $250K marketplace. Another possibility is that Sun and other competitors might be in wholesale collapse in the $250K server marketplace, making IBM's overall market penetration higher. (The only problem with that theory is that IDC trackers also show HP shrinking in that space.) Last, but not least, HP could be flat out lying. We report. You decide... -- Edward E Jaffe Phoenix Software International, Inc 5200 W Century Blvd, Suite 800 Los Angeles, CA 90045 310-338-0400 x318 [EMAIL PROTECTED] http://www.phoenixsoftware.com/ -- 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 Register article on HP replacing z
Concur. When I read the article my 'male bovine excrement' alarms went off. It is clearly (to me, anyway) based on facts as perceived by HP marketing. And we all know how marketing perceives and conveys 'facts' ;-) As would most Inquiring Minds, I'd like to know the reality. But I doubt that's going to happen any time soon. PS: I find The Register to be pretty much what it claims to be. With that in mind, its credibility rates somewhat above a typical US politico, way above a typical software salesperson, but well below the collective wisdom of this august group :-) -Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Bobbie Justice Sent: Tuesday, November 11, 2008 12:17 PM To: IBM-MAIN@BAMA.UA.EDU Subject: Re: The Register article on HP replacing z So if HP alone is converting more than 250 mainframe shops every 2 years, then at least 1,000 mainframe shops will be gone over the next 8 years? Ouch! Someone please tell me this isn't true? Dave Salt hmmm, sounds like more management by airport magazine hype. Don't believe everything you read. -- 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 NOTICE: This electronic mail message and any files transmitted with it are intended exclusively for the individual or entity to which it is addressed. The message, together with any attachment, may contain confidential and/or privileged information. Any unauthorized review, use, printing, saving, copying, disclosure or distribution is strictly prohibited. If you have received this message in error, please immediately advise the sender by reply email and delete all copies. -- 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 Register article on HP replacing z
I concur also. I wonder what year my company was included in the 250 per 2 years statistic. We brought in a brand-spanking-new HP superdome back in 2001 as our mainframe (a 7060H50!!) killer. As per my CTO back then don't do any maintenance to the mainframe because it will be gone in 3 years. You can guess the rest. Our z9-BC is running happily; the superdome is in the process of being replaced by newer, smaller (physically) HP-UX boxes, and the get off the mainframe project has been shelved. Rex -Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Hal Merritt Sent: Tuesday, November 11, 2008 12:52 PM To: IBM-MAIN@BAMA.UA.EDU Subject: Re: The Register article on HP replacing z Concur. When I read the article my 'male bovine excrement' alarms went off. It is clearly (to me, anyway) based on facts as perceived by HP marketing. And we all know how marketing perceives and conveys 'facts' ;-) As would most Inquiring Minds, I'd like to know the reality. But I doubt that's going to happen any time soon. PS: I find The Register to be pretty much what it claims to be. With that in mind, its credibility rates somewhat above a typical US politico, way above a typical software salesperson, but well below the collective wisdom of this august group :-) -- 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 Register article on HP replacing z
Counter-Point http://www.clabbyanalytics.com/uploads/HPvsIBMFinalFinalFinal.pdf http://www.clabbyanalytics.com/Free_Reports_Critiques.html Best Regards, Sam Knutson, GEICO System z Performance and Availability Management mailto:[EMAIL PROTECTED] (office) 301.986.3574 (cell) 301.996.1318 Think big, act bold, start simple, grow fast... This email/fax message is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution of this email/fax is prohibited. If you are not the intended recipient, please destroy all paper and electronic copies of the original message. -- 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 Register article on HP replacing z
2008/11/11 Dave Salt [EMAIL PROTECTED]: So if HP alone is converting more than 250 mainframe shops every 2 years, then at least 1,000 mainframe shops will be gone over the next 8 years? Ouch! Someone please tell me this isn't true? When HP claims it has helped more than 250 customers worldwide to migrate from mainframes to Integrity-based servers, I'm not at all sure that means that shop no longer runs a mainframe. It may well be that HP has had some success in converting apps to other platforms, and they will doubtless count that as a migration, even if the mainframe remains, or has even been upgraded. But it's a world filled with marketing claims, and HP is far from the only offender. Notice that as well as swallowing whatever HP has said, The Reg has also bought into IBM's line about speciality engines being some sort of performance enhancers for Java, Linux, and so on, rather than just a scheme for maintaining market differentiation between Classic and New workloads. Tony H. -- 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 Register article on HP replacing z
Going to have to go with IBM on this one. Those workloads are very CPU intensive and would be hugely expensive under classic mainframe pricing schemes. Not only would you need a lot more raw horse power, but your other vendors will want to charge you more as well. Designating some engines as outside the normal pricing equation not only annoys those vendors no end, but the bottom line is an improvement in the bang per buck ratios. I think that 'performance enhancement' is a fair claim. My $0.02 (before taxes) -Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Tony Harminc Sent: Tuesday, November 11, 2008 1:24 PM To: IBM-MAIN@BAMA.UA.EDU Subject: Re: The Register article on HP replacing z 2008/11/11 Dave Salt [EMAIL PROTECTED]: ..snip But it's a world filled with marketing claims, and HP is far from the only offender. Notice that as well as swallowing whatever HP has said, The Reg has also bought into IBM's line about speciality engines being some sort of performance enhancers for Java, Linux, and so on, rather than just a scheme for maintaining market differentiation between Classic and New workloads. Tony H. NOTICE: This electronic mail message and any files transmitted with it are intended exclusively for the individual or entity to which it is addressed. The message, together with any attachment, may contain confidential and/or privileged information. Any unauthorized review, use, printing, saving, copying, disclosure or distribution is strictly prohibited. If you have received this message in error, please immediately advise the sender by reply email and delete all copies. -- 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 Register article on HP replacing z
Dave Salt pisze: An interesting article indeed. For those who haven't read it, here are some statements that caught my eye: - HP announced today that in the past two years, it has helped more than 250 customers worldwide to migrate from mainframes to Integrity-based servers. - HP has been chasing IBM and other mainframes for 25 years, and ... has averaged around 100 per year for a while, but is accelerating - While IBM has specialty engines on mainframes that lower the cost of supporting Linux, Java, and DB2 workloads compared to regular mainframe engines, the lower price still leaves a big gap, even on the new z10 BC boxes. Even when you move down to a Linux specialty engine, it is an order of magnitude more expensive than Linux on an Integrity machines - HP contends that customers can save up to 70 per cent by getting off the mainframe So if HP alone is converting more than 250 mainframe shops every 2 years, then at least 1,000 mainframe shops will be gone over the next 8 years? Ouch! Someone please tell me this isn't true? Pay me, and I'll tell you what you want vbg More seriously, I believe that HP's words are true *in some sense*. I also believe it is definitely not true, that avg 100 mainframes shops per year were gone. No. Rather avg. 100 customers *tried* to get rid of mainframe and bought something from HP. With what results ? Well... You know... Reboot Hill... However, please keep in mind that ANY system conversion is risky: off mainframe, to mainframe, VSE-z/OS, TPF-z/OS any kind - it is *always* risky. The older, the bigger, the more specific application the bigger risk is. I also pay attention that they talk about small mainframes. Yes, it's easier to migrate off small mainframe, because you should not worry about CPU power. Last but not least: we use mainframes, but our business use *applications*. Yes, indeed, our lovely mainframes are *not* for Parallel Sysplex, are not for GDPS, are not for IMS, DB2, DFSMS. They are for applications. I doubt HP has skills to migrate off *any* application, regardless what kind of applications it is. I was involved in several migrations in bank world. Trust me, it was never matter of platform, it was always matter of application functionality. My $0.02 (before crisis) -- Radoslaw Skorupka Lodz, Poland -- BRE Bank SA ul. Senatorska 18 00-950 Warszawa www.brebank.pl Sd Rejonowy dla m. st. Warszawy XII Wydzia Gospodarczy Krajowego Rejestru Sdowego, nr rejestru przedsibiorców KRS 025237 NIP: 526-021-50-88 Wedug stanu na dzie 01.01.2008 r. kapita zakadowy BRE Banku SA wynosi 118.642.672 zote i zosta w caoci wpacony. -- 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 Register article on HP replacing z
Pesky users. Keeping the REAL reason for Mainframes obscured. Ken Porowski AVP Systems Software CIT Group E: [EMAIL PROTECTED] -Original Message- R.S. snip Last but not least: we use mainframes, but our business use *applications*. Yes, indeed, our lovely mainframes are *not* for Parallel Sysplex, are not for GDPS, are not for IMS, DB2, DFSMS. They are for applications. -- 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 Register article on HP replacing z
Pesky users. Keeping the REAL reason for Mainframes obscured. END USER is still missing in the z/Arch instruction set. - Too busy driving to stop for gas! -- 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 Register article on HP replacing z
The following message is a courtesy copy of an article that has been posted to bit.listserv.ibm-main as well. [EMAIL PROTECTED] (Pommier, Rex R.) writes: I concur also. I wonder what year my company was included in the 250 per 2 years statistic. We brought in a brand-spanking-new HP superdome back in 2001 as our mainframe (a 7060H50!!) killer. As per my CTO back then don't do any maintenance to the mainframe because it will be gone in 3 years. You can guess the rest. Our z9-BC is running happily; the superdome is in the process of being replaced by newer, smaller (physically) HP-UX boxes, and the get off the mainframe project has been shelved. sumperdome involved some people that had been involved in ibm risc group. it was somewhat positioned as more cost-effective (convex) examplar. SCI was commodity (NUMA) shared memory scaleup technology ... somewhat out of SLAC. DG Sequent had done NUMA 256 processor machines (64-port SCI, with 64 boards four 486 processors per board). IBM later bought Sequent. Convex had done NUMA 128 processor machines (64-port SCI, with 64 boards two HP RISC processors per board). HP bought Convex ... and superdome was somewhat positioned as a more cost-effective Examplar. SGI also did SCI NUMA machines with MIPS RISC processors. Part of the issue has been the programming complexity to take advantage of NUMA architectures ... not unlike all the current stuff about how to migrate traditional desktop software to take advantage of multi-core processors. There are also still a large number of issues with regard to maturity level of all the u*ix systems for business critical dataprocessing vis-a-vis legacy commercial systems. This is less of an issue when there is a large DBMS or other large application subsystem (possibly in a single, dedicated environment) that masks underlying operating system characteristics. -- 40+yrs virtualization experience (since Jan68), online at home since Mar70 -- 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 Register article on HP replacing z
Tony Harminc wrote: ... The Reg has also bought into IBM's line about speciality engines being some sort of performance enhancers for Java, Linux, and so on, rather than just a scheme for maintaining market differentiation between Classic and New workloads. But, specialty engines *do* enhance performance of these workloads! Many users (us included) have sub-capacity CPs. Our current hardware is a 2096-L03. Each CP generates about 66 MIPS. The IFLs, zAAPs and zIIPs run at 470 MIPS -- or about seven times faster. We plan to get a z10 BC before the end of the year. On that machine, the specialty engines run at an astonishing 713 MIPS! -- Edward E Jaffe Phoenix Software International, Inc 5200 W Century Blvd, Suite 800 Los Angeles, CA 90045 310-338-0400 x318 [EMAIL PROTECTED] http://www.phoenixsoftware.com/ -- 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 Register article on HP replacing z
On 11 Nov 2008 12:44:31 -0800, [EMAIL PROTECTED] (R.S.) wrote: Pay me, and I'll tell you what you want vbg More seriously, I believe that HP's words are true *in some sense*. I also believe it is definitely not true, that avg 100 mainframes shops per year were gone. No. Rather avg. 100 customers *tried* to get rid of mainframe and bought something from HP. With what results ? Well... You know... Reboot Hill... Also, there are always companies whose needs are increasing and always companies whose needs are decreasing. With a pure equilibrium, some mainframe shops will downgrade - and some mini shops will upgrade. A one sided stat doesn't even indicate trends, much less which choice is best for a particular company. -- 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 Register article on HP replacing z
Figures don't lie, but liars will figure. Smacks of management by airline magazine, with a generous dose of marketting Male Bovine Excrement. John McKown wrote: I don't know how accurate it is, but I found it interesting. http://www.theregister.co.uk/2008/11/11/hp_chases_mainframes/ -- John -- 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 -- 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
