Re: z9 / z10 instruction speed(s)
In ofc8c33b82.6cab769c-on802576e4.004efdf4-802576e4.004f2...@uk.ibm.com, on 03/12/2010 at 02:24 PM, Martin Packer martin_pac...@uk.ibm.com said: Seymour, wrong by about a decade :-) ... Eight years. 3090 was first to have true Expanded Storage, 9021 the last. What about the 9121 and 9221? With CMOS we had the partition real memory thing. Weren't the low end ES/9000 boxen CMOS? Unless it's ME that is wrong by about a decade. :-) No, it's me. Thanks for the correction. They say that the mind is the second thing to go; I can't remember the first. -- Shmuel (Seymour J.) Metz, SysProg and JOAT ISO position; see http://patriot.net/~shmuel/resume/brief.html We don't care. We don't have to care, we're Congress. (S877: The Shut up and Eat Your spam act of 2003) -- 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: z9 / z10 instruction speed(s)
In b53f38421003081137h70e103cfs85330fcffb40e...@mail.gmail.com, on 03/08/2010 at 02:37 PM, George Henke gahe...@gmail.com said: What is not just or equitable is for IBM to view EDS' efficiency and profits from economies of scale as a loss of revenue to themselves (IBM) and then create a pricing scheme that appropriates those profits from such efficiencies for themselves. The hardware and operations sides may be a case of economy of scale, but on the software side the story is very different. It's by no means clear who was appropriating whose profits from whom. Since when have monopolies ever been interested in efficiencies except for their own profitability. That applies to the megacenters as much as it does to IBM. Any well run business is concerned with it's own profitability, and only concerned with the profitability of its customers and suppliers to the extent that they impact its own. Monopolies do not optimize the efficient allocation of resources in the economy. They maximize their own profitablity at the expense of efficient allocation of resources in the economy. Albert Einstein admonished to make things as simple as possible, but no simpler. The Devil is in the details. Google for natural monopoly. As for artificial monopolicies, they can abuse their powers but abuse is not intrinsic. -- Shmuel (Seymour J.) Metz, SysProg and JOAT ISO position; see http://patriot.net/~shmuel/resume/brief.html We don't care. We don't have to care, we're Congress. (S877: The Shut up and Eat Your spam act of 2003) -- 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: z9 / z10 instruction speed(s)
In ofeec49b15.34ea1afd-on852576df.0039741b-852576df.003fb...@us.ibm.com, on 03/07/2010 at 06:35 AM, Timothy Sipples timothy.sipp...@us.ibm.com said: There are, and very many. You're not speaking the same language that he is. In my previous post it should be obvious that smaller means anybody with capacity below a hypothetical non-kneecapped 7-way System z9 BC. Just as it should be obvious that Peter means something *much* smaller than that, and that he is talking about envirnmentals and price, not just CPU capacity. so again I have no idea where you're getting your information. The problem isn't that your information differs, the problem is that your nomenclature differs. Have mainframe technologies ever been more affordable? Heck no. (I've looked up a lot of historical prices.) What was the historical price of a P/390? And it's long past time we stop perpetuating mainframe pricing myths, because it isn't helping anyone. Neither is talking past each other. Perhaps it would be better to stop using the adjective small and start using numbers, e.g., total hardware price $50K. -- Shmuel (Seymour J.) Metz, SysProg and JOAT ISO position; see http://patriot.net/~shmuel/resume/brief.html We don't care. We don't have to care, we're Congress. (S877: The Shut up and Eat Your spam act of 2003) -- 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: z9 / z10 instruction speed(s)
In b53f38421003080735x3b3d4fadt96f4c714c9775...@mail.gmail.com, on 03/08/2010 at 10:35 AM, George Henke gahe...@gmail.com said: I don't see anyone complaining about 64-bit memory being to much and asking to bring back Expanded Storage and paging. Apples and oranges. AFAIK the 308x boxen were the last to have true expanded storage instead of arbitrarily labelling some of the central storage as expanded. Were IBM to design a large BORAM at prices significantly less than RAM prices then I for one would like to see it available as expanded storage for 64-bit systems. What I wouldn't like to see is artificially carving up the central storage as was done from the 3090 on. As to paging, we still have it. With enough central memory the rates are fairly low except during, e.g., dump capture. What nobody wants is excessive paging rates, which you could only bring *back* if you had already experienced them. The last time that I recall seeing excessive paging rates was when a fixed-head disk was out of service for two weeks due to a head crash. -- Shmuel (Seymour J.) Metz, SysProg and JOAT ISO position; see http://patriot.net/~shmuel/resume/brief.html We don't care. We don't have to care, we're Congress. (S877: The Shut up and Eat Your spam act of 2003) -- 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: z9 / z10 instruction speed(s)
In 1228950277-1267928263-cardhu_decombobulator_blackberry.rim.net-10127518...@bda026.bisx.prod.on.blackberry, on 03/07/2010 at 02:17 AM, Ted MacNEIL eamacn...@yahoo.ca said: A rose is a rose is a rose. But a rose is not a sardine. MSUs are just MIPS multiplied by a (marketting) constant or 5. No. -- Shmuel (Seymour J.) Metz, SysProg and JOAT ISO position; see http://patriot.net/~shmuel/resume/brief.html We don't care. We don't have to care, we're Congress. (S877: The Shut up and Eat Your spam act of 2003) -- 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: z9 / z10 instruction speed(s)
Seymour, wrong by about a decade :-) ... 3090 was first to have true Expanded Storage, 9021 the last. With CMOS we had the partition real memory thing. Unless it's ME that is wrong by about a decade. :-) Martin Packer, Mainframe Performance Consultant, Worldwide Banking Center of Excellence, IBM +44-7802-245-584 email: martin_pac...@uk.ibm.com Twitter / Facebook IDs: MartinPacker Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU -- 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: z9 / z10 instruction speed(s) -- OCTANE of CPUs
Steve, So how good is this petrol that avoids detonation? I don't see that idea catching on... (GDR) Ron 100LL (Low Lead -- as in TetraEthyl Lead) is needed to avoid detonation in high horsepower engines. -- 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: z9 / z10 instruction speed(s) -- OCTANE of CPUs
Not so much high-horsepower engines, but rather high-compression engines. Makes a HUGE difference in aircraft reciprocating engines. Higher compression leads to higher heat buildup in the cylinder and that can lead to pre-ignition, with seriously detrimental effect on the engine and the power curve. Higher octane fuels reduce the pre-ignition problems. Rick - Ron Hawkins wrote: Steve, So how good is this petrol that avoids detonation? I don't see that idea catching on... (GDR) Ron 100LL (Low Lead -- as in TetraEthyl Lead) is needed to avoid detonation in high horsepower engines. -- 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 . -- 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: z9 / z10 instruction speed(s) -- OCTANE of CPUs
On Wed, Mar 10, 2010 at 4:11 PM, Rick Fochtman rfocht...@ync.net wrote: Not so much high-horsepower engines, but rather high-compression engines. Makes a HUGE difference in aircraft reciprocating engines. Higher compression leads to higher heat buildup in the cylinder and that can lead to pre-ignition, with seriously detrimental effect on the engine and the power curve. Higher octane fuels reduce the pre-ignition problems. Note that the distinction here is between *controlled combustion* and *detonation*, which is uncontrolled combustion and can cost you cylinder heads, blocks, valves, etc. Not a good thing. Are we far enough OT yet? -- 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: z9 / z10 instruction speed(s) -- OCTANE of CPUs
On 10 March 2010 16:30, zMan zedgarhoo...@gmail.com wrote: On Wed, Mar 10, 2010 at 4:11 PM, Rick Fochtman rfocht...@ync.net wrote: Not so much high-horsepower engines, but rather high-compression engines. Makes a HUGE difference in aircraft reciprocating engines. Higher compression leads to higher heat buildup in the cylinder and that can lead to pre-ignition, with seriously detrimental effect on the engine and the power curve. Higher octane fuels reduce the pre-ignition problems. Note that the distinction here is between *controlled combustion* and *detonation*, which is uncontrolled combustion and can cost you cylinder heads, blocks, valves, etc. Not a good thing. Are we far enough OT yet? We could bring up Diesel engines, with their very high compression ratio (23.5:1 on my old Rabbit), and speculate on why they are almost unheard of as aircraft power plants. But it's not Friday yet, so I won't. Tony H. -- 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: z9 / z10 instruction speed(s)
Steve Comstock writes: I would guess the focus here was on jobs, plain and simple. Maybe installed mainframe MIPS are increasing, but jobs, especially for z/OS staff, appear to be declining (esp. in the US, but some on the list have mentioned similar trends in Europe). I'm not sure if that's true or not, but let's focus on the U.S. labor market and sanity check a couple areas: 1. Have IT jobs *in general* been declining in the U.S. in recent years? Heck yes. It's not as bad as, say, the automotive sector, but it hasn't been good for IT staff in the first decade of the new millenium. The IT job market has yet to recover fully from the dot-com bust. But the U.S. Bureau of Labor Statistics forecasts strong growth in IT jobs over the next decade (mainframe included), with the possible exception of computer programming (i.e. direct writing of code), which is forecast to decline about 3 percent over the decade. (Coding is expected to move outside the U.S. faster than growth in the field, basically.) That said, in 2001 BLS forecast strong growth in IT jobs of as much as 100 percent in the decade for certain categories. That just didn't happen. So even the experts can get the forecasts wrong. But there is a general consensus that there will be future growth in IT jobs, including mainframe-related jobs, in the U.S. 2. Have jobs *in general* been declining in the U.S. in recent years? Heck yes. The U.S. is in the midst of a terrible labor recession, with the U3 unemployment rate sitting at just shy of 10%, or roughly twice the rate of the dot-com years. Moreover, the entire decade consisted of slow job growth at best. It was a terrible decade for nearly all workers in the U.S., and there's no general employment recovery in sight yet. 3. Mainframe labor productivity has improved substantially and continues to improve. Is that bad? Well, it does mean mainframe owners need fewer person-hours than they did in the past to produce the same mainframe-related output value. Then again, most rational employers steer their investments and operations budgets toward higher productivity areas, so that's good for mainframers. If you're more productive than others -- you sure do seem to be -- then to the extent there's competition for resources you'll do better, on average. For what it's worth, I largely disagree with the rest of your comments, Steve. In particular, I've previously (and often) pointed out the fact that IBM has repeatedly and substantially reduced pricing and increased value *across the mainframe board* on a sustained basis for at least 10 years -- heck, it's right in The Mainframe Charter as our CEO's commitment to our customers. Yet people want to deny that reality, too, and even suggest conspiracy theories about z/OS. Sorry, I'm just tired of the conspiracy theories, especially when I've just spent the past several days working full time on serving the needs of another prospective new z/OS customer with as much institutional support as I could ever want or need. Perhaps too much. :-) Good grief, enough already! - - - - - Timothy Sipples IBM Consulting Enterprise Architect for New, Advanced, and/or Innovative Solutions (VCT) Based in Singapore Serving the Growth Markets E-Mail: timothy.sipp...@us.ibm.com -- 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: z9 / z10 instruction speed(s)
No, the real reason *is* cost. Just because you state it is manipulation, doesn't make it a reality. But is does enlighten the rest of us to your line of reasoning. Conspiracy theories, anyone? Bob -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of George Henke Sent: Monday, March 08, 2010 7:15 PM To: IBM-MAIN@bama.ua.edu Subject: Re: z9 / z10 instruction speed(s) And, the real reason (as unpopular as it may be), is cost. No, the real reason is manipulation of supply and demand by a monopoly. snipped -- 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: z9 / z10 instruction speed(s)
Timothy Sipples wrote: Steve Comstock writes: I would guess the focus here was on jobs, plain and simple. Maybe installed mainframe MIPS are increasing, but jobs, especially for z/OS staff, appear to be declining (esp. in the US, but some on the list have mentioned similar trends in Europe). [snip some valid points] For what it's worth, I largely disagree with the rest of your comments, Do you mean: --- It would help if IBM would tell the story, then, wouldn't it? An old thread: IBM doesn't seem interested in winning the hearts and minds battle for the mainframe (esp. z/OS), based on lack of articles and ads promoting and raising awareness of z/OS or even z Series. A strong campaign would go a long way towards getting perceptions caught up with reality, eh? FWIW, I think you do an excellent job presenting an optimistic view of the z/OS world. But on the whole, IBM does not. I'm not calling this a conspiracy, I'm just saying a major reason that mainframes are not really visible to the larger public is that IBM does not do anything to raise visibility. Thus, I am often asked Do they still make mainframes?. The new (a year or two ago) head of computing in Colorado was quoted in the local newspaper as saying No one runs mainframes anymore, despite the fact that the state is a z/OS user. There was quite a debacle here when a new system failed to meet the needs of its clients: people needing food stamps, welfare, unemployment insurance payments, etc. I suspect it was a case of trying to move an app off the aging mainframe to a more modern platform that went awry, but the truth will never be published. It seems to me we want to be in a position where using a mainframe for apps should be included as one of the natural choices to consider, whereas today, the mainframe has the [false] perception of being obsolete, even non-existent, so the assumption by the young IT managers is that the only choices are Unix and WinTel. Steve. In particular, I've previously (and often) pointed out the fact that IBM has repeatedly and substantially reduced pricing and increased value *across the mainframe board* on a sustained basis for at least 10 years -- heck, it's right in The Mainframe Charter as our CEO's commitment to our customers. Yet people want to deny that reality, too, and even suggest conspiracy theories about z/OS. Sorry, I'm just tired of the conspiracy theories, especially when I've just spent the past several days working full time on serving the needs of another prospective new z/OS customer with as much institutional support as I could ever want or need. Perhaps too much. :-) Good grief, enough already! - - - - - Timothy Sipples IBM Consulting Enterprise Architect for New, Advanced, and/or Innovative Solutions (VCT) Based in Singapore Serving the Growth Markets E-Mail: timothy.sipp...@us.ibm.com -- Kind regards, -Steve Comstock The Trainer's Friend, Inc. 303-393-8716 http://www.trainersfriend.com * z/OS application programmer training + Instructor-led on-site classroom based classes + Course materials licensing + Remote contact training + Roadshows + Course development -- 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: z9 / z10 instruction speed(s)
On Mon, 8 Mar 2010 14:43:07 -0600, Rick Fochtman rfocht...@ync.net wrote: But it makes excellent sense in the context of failure and recovery. Having a spare engine to switch over to in the event of a failure in the primaty engine(s) can make a HUGE difference for a shop that needs to maximize availability. Rick Yes agree When it comes to CBU or spare CPU power I guess we can also learn (or at least look at) from other platforms Humility can sometimes be more profitable than mainframe arrogance. On non z/OS platform like x86 with Linux or Windows servers, we often use Dev or Test machines at a purposely less than 50% capacity allowing DR on the test machines as the remaining CPU power is available. Contrary to what a lot of people say about mainframe, using CPU's at a low percentage is a definite advantage as far as you are not paying a different bill,and using a CPU at 95% is just a disadvantage. Yes we can be proud, but we should not be narrow minded. Would we run our own car at 95% power all the time and have no resources left to overtake? Unused power takes care of unwanted peaks in production, planned heavy migration, handles unwanted programmers loop in a dev environment. And the dev environment can be used as a DR platform. Best of all on other platforms it does not change the bill as we are normally charged according to the number of engines. If yesterday x86 were used at 10% capacity, and moving application from one hardware to another was very difficult because of different proc levels, bios etc, virtualisation like VMware is removing all these constraints. Vmotion( example) has simplified so much the way of working, that today we have to slow down the attitude of our system guys because they tend to load these 8 or 16 engines servers to 80 or 90% of their capacity, thus killing the reserved DR capacity. Of course VM was a mainframe invention, but today x86 virtualisation is far more advanced. MSU pricing seems to be accepted mainly (only?) in mainframe world, but I wonder if the same people who believe it is a fair system, would accept paying their windows license on their client PC according to the MIPS or MSU rating. NOT me !! definitely ( sh... I changed my hardware PC 3 months ago) Windows 3.11 in 1985 on a 20 or 40 mhz PC used to cost 100 dollars a license. Today a Window Seven must cost 150$ on a dual or quadri core 3 ghz PC ... and still Microsoft became rich. I am not defending Microsoft policies, but their server software became quite reliable over the years and I am quite happy that their or my own license did not follow the GHZ race on my servers, my home PC or my Thinkpad. And it has helped our company budget ( because I happen to have handled the bills of all platforms and I wonder what we would have done if replacing 5000 pc workstations had meant an increase in software because they were more powerfull). However it seems that the profit is so good in this scheme, that Microsoft, Oracle, IBM are now thinking to apply such a pricing to x86 envt by using PVU's at least in server software.As a customer this is definitely bad news. I am a convinced mainframe person, but it is good to look at the other worlds because they did not become successfull by being incompetent. Reliability of a system comes from the men running it, it is well known that 90% of breakdowns on servers are due to human error ... not because of the OS. On client PC's it is even more because anything is installed at any time. As for the 3 MSU z hardware that we get advertised every now and then on the list, I just wonder how many ISV vendors replaced their T3 PC's by a 3 MSU z/Serie and why some demos from ISV and IBM alike are still being shown on Thinkpads. Is it the price ? or the weigth ? I just wonder how many z/OS mips you can extract from a desktop and at what price ? Obviously there is a market ( ISV's, testers, programmers) but nobody wants to satisfy this part of the market because it would open the door to z/390 emulation (Hercules or T3 alike) under x86 emulation( Vmware alike) and take a part of the market to other hardware platform thus killing this very profitable pricing strategy. This Z entry level story is a joke as far as I am concerned and I do not believe I am the only one. Bruno Sugliani zxnetconsult(at)free(dot)fr -- 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: z9 / z10 instruction speed(s)
Clock speed is only a small piece of the picture. Other major factors are how many clock cycles are required to execute an instruction, how many bits or bytes can be manipulated by a single instruction, how many instructions can concurrently be in the execution pipeline, how many independent ports there are to memory, the width in bits/bytes of the memory and I/O bus paths, etc. Joel, Thank you very much for your excellent analysis of the 2 platforms. I have never seen a clearer, more succinct, comparison of the two platforms in plain English anywhere. You obviously have an in-depth understanding of the hardware and software on both. Thank you very much for sharing it. I learned a lot. On Tue, Mar 9, 2010 at 12:09 AM, Joel C. Ewing jcew...@acm.org wrote: Clock speed is only a small piece of the picture. Other major factors are how many clock cycles are required to execute an instruction, how many bits or bytes can be manipulated by a single instruction, how many instructions can concurrently be in the execution pipeline, how many independent ports there are to memory, the width in bits/bytes of the memory and I/O bus paths, etc. Even 30 years ago, I/O channels on all but the smallest mainframes were essentially independent processors in their own right with independent memory access. The number of independent channels and aggregate I/O transfer rate on a z/10 today can easily outpace a PC by several orders of magnitude and the I/O data transfers don't degrade the general purpose CP's effective speed. When PCs were promoting the 16-bit to 32-bit architecture revolution a number of years back, larger mainframes had already been exploiting internal memory bus widths of 128-bits or more for over a decade. Typically the mainframes performance has been optimized for high data transfer rates, manipulating multi-byte data values (larger integer and floating point values, data records) and multi-tasking, and a single instruction may do some very complex data manipulations that would require many PC instructions. The PC architecture has historically been more concerned with optimizing manipulation of individual bits and bytes, a capability indispensable for image processing and driving pretty graphical user display interfaces. And, before you can talk about doing useful application-level data manipulation, that graphical display interface on the typical PC can suck up a significant portion of the available CPU power and available memory. Consumption of additional PC CPU resources is imposed by limitations of MS Operating systems, for such things as Anti-Virus protection applications which are unnecessary in a mainframe environment. A mainframe, even one with a slower clock speed, can typically run circles around a PC when dealing with a large number of independent users or with applications that are I/O intensive. On the other hand, a PC of comparable clock speed can do a single CPU intensive task where individual bit or byte manipulations predominate in less real time than the typical mainframe. JC Ewing On 03/06/2010 07:44 PM, Lindy Mayfield wrote: This has been a very interesting thread for me. If I remember correctly from the time I saw the z/10 with plexiglass outsides and a hardware guy there to explain what was what, and one of the things he told me was the cpu speed was (IIRC) 4.77GHz. My laptop has a dual core 1.99GHz. You already see where I'm going with this. How does a z/10 get so much more done? Forgive me for my math and analytical skills, but seems like 4 laptops could equal the speed of a 1 cpu z/10. There is a huge chunk of this equation that I'm totally missing. How does a z/10 get so much more done? kind regards, Lindy -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of R.S. Sent: 7. maaliskuuta 2010 2:23 To: IBM-MAIN@bama.ua.edu Subject: Re: z9 / z10 instruction speed(s) Edward Jaffe pisze: [...] People with PC-only experience are always astonished when I tell them about modern mainframe provisioning capabilities. They always assume when your hard drive fills up you need a new one or when your CPU is too slow you need a new one. What we do seems like magic to them. Yes, mainframe capabilities are excellent in this area. From the other hand they solve problems which exist only in mainframe world: CPU power adjustment. CPU shortage is bad thing on any platform, but mainframe is the only one where too much MIPS is not good. Why to downgrade a PC? The same apply to specialty processors. -- Joel C. Ewing, Fort Smith, ARjremoveccapsew...@acm.org -- 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
Re: z9 / z10 instruction speed(s)
Bruno Sugliani wrote: On non z/OS platform like x86 with Linux or Windows servers, we often use Dev or Test machines at a purposely less than 50% capacity allowing DR on the test machines as the remaining CPU power is available. Contrary to what a lot of people say about mainframe, using CPU's at a low percentage is a definite advantage as far as you are not paying a different bill,and using a CPU at 95% is just a disadvantage. And, this is exactly what I'm seeing many of our customers doing with their mainframes. They are setting up LPAR Capacity groups to guarantee a large amount of so-called white space on the CEC--sometimes over 50 percent. They can then use this white space for sudden demand spikes from existing LPARs or for failover from other LPARs or CECs. -- Edward E Jaffe Phoenix Software International, Inc 831 Parkview Drive North El Segundo, CA 90245 310-338-0400 x318 edja...@phoenixsoftware.com http://www.phoenixsoftware.com/ -- 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: z9 / z10 instruction speed(s)
On Sat, 6 Mar 2010 19:25:35 -0500, Timothy Sipples wrote: a fully configured System z9 BC (until end of June). I don't like MIPS metrics, but for those that still do it'd be thousands of MIPS. Thousands? Not quite. Now IBM introduces the System z10 BC, and you upgrade. You get thousands more MIPS if there's no kneecapping available. A z10 BC is thousands more MIPS than a z9 BC? Nope. -- Tom Marchant -- 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: z9 / z10 instruction speed(s)
How can anyone have too many mainframe MIPS? Really? Dahh. Just look at SHADOW, the Data Direct Product that enables a TCB workload, through some niffty fancy footwork, to run workloads on a Ziip processor which handles SRB only workloads. I have a former client who automagically extended the CPU life of his 2 GPP (General Purpose Processor) configuration by installing their product. Ziip/Zaap processors are UNTHROTTLED, UNMETERED. Not like GPP's that are both throttled and metered. And wherefore, the IBM suit against NEON, a competing product? Please note well, the IBM suit is not against SHADOW which was as niffty in its design for the Patent Law as it was for the Ziip processor itself. S, if all this knee-capping is s great, why is everyone installling this product so they can get FREE, UNTHROTTLED MIPS? Let's stop kidding ourselves, IBM and third party vendors are just using every device they can find to get what they want and then ratiionalize it. Textbook definition of rationallzation: Something that is completely logical, just not true. Much of this trail extolling the virtues of knee-capping ad nauseum is nothing more than just that, rationalization. IBM must be chortling at all this, all the way to the bank. On Mon, Mar 8, 2010 at 2:32 AM, R.S. r.skoru...@bremultibank.com.pl wrote: Timothy Sipples pisze: [...] By the way, IBM substantially reduced the price of z/VSE for many/most customers (including smaller ones) via the new MWLC sub-capacity licensing for Version 4, so again I have no idea where you're getting your information. Same with z/VM: IBM has slashed the price, big time. A zNALC z/OS license at 3 MSUs starts way, way, way down in the 3 figures per month U.S. (And then there are the Solution Edition offerings.) IBM has never done anything like that before. Have mainframe technologies ever been more affordable? Heck no. (I've looked up a lot of historical prices.) And it's long past time we stop perpetuating mainframe pricing myths, because it isn't helping anyone. Sounds very reasonable. Can I ask where are new customers of this very affordable platform? Did anyone observed mass migration to z Series? Any crowd in mainframe shop? :-( -- Radoslaw Skorupka Lodz, Poland -- BRE Bank SA ul. Senatorska 18 00-950 Warszawa www.brebank.pl S d Rejonowy dla m. st. Warszawy XII Wydzia Gospodarczy Krajowego Rejestru S dowego, nr rejestru przedsi biorców KRS 025237 NIP: 526-021-50-88 Wed ug stanu na dzie 01.01.2009 r. kapita zak adowy BRE Banku SA (w ca o ci wp acony) wynosi 118.763.528 z otych. W zwi zku z realizacj warunkowego podwy szenia kapita u zak adowego, na podstawie uchwa y XXI WZ z dnia 16 marca 2008r., oraz uchwa y XVI NWZ z dnia 27 pa dziernika 2008r., mo e ulec podwy szeniu do kwoty 123.763.528 z . Akcje w podwy szonym kapitale zak adowym BRE Banku SA b d w ca o ci op acone. -- 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 -- George Henke (C) 845 401 5614 -- 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: z9 / z10 instruction speed(s)
Timothy, Even if there was no kneecapping, there could still be CPUs turned off and on, so the minimal configuration would be a single CPU z9BC, not a 7 way. Timothy Sipples timothy.sipp...@us.ibm.com 3/7/2010 6:35 AM Peter Farley writes: But Tim, there *aren't* any smaller mainframe customers any more (at least not in the USA). There are, and very many. I don't know where you're getting your information. In my previous post it should be obvious that smaller means anybody with capacity below a hypothetical non-kneecapped 7-way System z9 BC. That includes the very smallest customers up through...well, that forms the huge *majority* of customers! By the way, IBM substantially reduced the price of z/VSE for many/most customers (including smaller ones) via the new MWLC sub-capacity licensing for Version 4, so again I have no idea where you're getting your information. Same with z/VM: IBM has slashed the price, big time. A zNALC z/OS license at 3 MSUs starts way, way, way down in the 3 figures per month U.S. (And then there are the Solution Edition offerings.) IBM has never done anything like that before. Have mainframe technologies ever been more affordable? Heck no. (I've looked up a lot of historical prices.) And it's long past time we stop perpetuating mainframe pricing myths, because it isn't helping anyone. AFAIK, there is no way for IBM to bring z/OS on a z10 to a customer with a yearly IT budget in the 5- to (very) low 6-figure range, including software and staffing costs and necessary 3rd-party software. That sounds like an annual IT budget of about $150,000 U.S. (to be generous). And you're right: when it comes to a dedicated/on-site mainframe installation (or, gosh, business computing in general), that's a very tight budget. (However, IBM and/or other organizations could likely provide such an organization with hosted/cloud z/OS services well within that budget.) But I'm not sure when those greener pastures and fonder memories ever actually existed, because $150,000 in 2010 dollars didn't go too far in the 1970s, 1980s, and 1990s either. For example, you would be very fortunate to hire two U.S. FTEs for your IT organization within that budget, but then you'd have no money left over. For reference, when you convert $150,000 in 2009 to 1975, you get $38,040. That is, a total IT budget of $38,040 in 1975 would be equivalent, in Consumer Price Index terms, to a $150,000 budget in 2009. I don't know Did anybody work for an organization in 1975 which had a total annual IT budget of $38,040 or thereabouts...and which owned and operated a dedicated mainframe on-site with licensed software? I think that would have been much *tougher* than it is in 2010. (I remember when Hershey bars cost a nickel) By the way, a lot of businesses cannot afford to buy a tractor-trailer truck either. Between the truck itself, maintenance, driver, fuel, insurance, tolls, etc. it costs a lot more than $150K per year to own and operate one. Yet there are huge numbers of them plying American roads. IBM does not recommend a dedicated/on-site mainframe to *every* business. Just to many/more of them. Charles Mills writes: Now there you go being logical again. Goodness knows I try. :-) But I speak only for myself, so maybe that's why I'm logical. :-) - - - - - Timothy Sipples IBM Consulting Enterprise Architect for New, Advanced, and/or Innovative Solutions (VCT) Based in Singapore Serving the Growth Markets E-Mail: timothy.sipp...@us.ibm.com -- 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 CONFIDENTIALITY/EMAIL NOTICE: The material in this transmission contains confidential and privileged information intended only for the addressee. If you are not the intended recipient, please be advised that you have received this material in error and that any forwarding, copying, printing, distribution, use or disclosure of the material is strictly prohibited. If you have received this material in error, please (i) do not read it, (ii) reply to the sender that you received the message in error, and (iii) erase or destroy the material. Emails are not secure and can be intercepted, amended, lost or destroyed, or contain viruses. You are deemed to have accepted these risks if you communicate with us by email. Thank you. -- 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: z9 / z10 instruction speed(s)
On Mon, Mar 8, 2010 at 10:24 AM, Scott Rowe scott.r...@joann.com wrote: Timothy, Even if there was no kneecapping, there could still be CPUs turned off and on, so the minimal configuration would be a single CPU z9BC, not a 7 way. Given the axioms of the discussion, it would be an entire book. The contention was that IBM shipping capacity that isn't used doesn't make sense. -- 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: z9 / z10 instruction speed(s)
Let's not be bit wise, and byte foolish. There is merit in intentionally overcapcitating not just in MIPS, but also DASD. I don't see anyone complaining about 64-bit memory being to much and asking to bring back Expanded Storage and paging. On Mon, Mar 8, 2010 at 10:30 AM, zMan zedgarhoo...@gmail.com wrote: On Mon, Mar 8, 2010 at 10:24 AM, Scott Rowe scott.r...@joann.com wrote: Timothy, Even if there was no kneecapping, there could still be CPUs turned off and on, so the minimal configuration would be a single CPU z9BC, not a 7 way. Given the axioms of the discussion, it would be an entire book. The contention was that IBM shipping capacity that isn't used doesn't make sense. -- 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 -- George Henke (C) 845 401 5614 -- 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: z9 / z10 instruction speed(s)
The following message is a courtesy copy of an article that has been posted to bit.listserv.ibm-main,alt.folklore.computers as well. t...@harminc.net (Tony Harminc) writes: I don't know about a total IT budget of $38k, but in 1975 licensed software was pretty much a novelty. The first priced version of MVS (or any other IBM OS except perhaps ACP/TPF?) had yet to appear, and most software was written in house. Some shops were using priced IBM software like PL/I Optimizer, or the COBOL equivalent, and bigger places ran non-free IMS and/or CICS. And there were priced (duh) products from other vendors, like Syncsort. But buying run-the-business application packages was pretty rare. 23jun69 unbundling announcement started charging for software, SE services, other stuff (result of various litigation). however, corporation managed to make the case that kernel software should still be free ... misc. past posts mentioning unbundling http://www.garlic.com/~lynn/submain.html#unbundling the distraction of the FS project allowed the 360/370 product pipeline to go dry ... with the failure of FS ... some past posts http://www.garlic.com/~lynn/submain.html#futuresys there was mad rush to get products back into the 370 product pipeline (both hardware software; aka FS was planned to replace 370 something that was radically different from 370). the lack of products in the pipeline was also used to explain the clone processors being able to gain foothold in the market. as a result, there was decision to also start charging for kernel software. I had been doing 360/370 during the FS period (and even made some less than complimentary observations about the FS activity) ... somewhat as a result ... some of that stuff was picked up as part of basic vm370 release 3. There was then decision made to package up lots of my other stuff and release it as an independent resource manager ... nad it was selected to be the guinea pig for kernel software charging. I got to spend lots of time with business, planning, legal groups on the subject of kernel software charging. Part of the policy was that kernel software directly involved in hardware support was to remain free ... but other stuff could be charged for. The resource manager was shipped as separately charged for product. The pricing of such software had to at least cover the cost of the development (aka pricing couldn't be done at leas than the development cost). Getting close to release ... the work on deciding price for MVS resource manager had been done ... and the direction was that my resource manager couldn't be shipped at lower price than what was being planned for the MVS resource manager (although the MVS development costs had been enormously greater ... all the work that I had done with regard to my costs for setting price ... just went out the window). http://www.garlic.com/~lynn/subtopic.html#fairshare http://www.garlic.com/~lynn/subtopic.html#wsclock During the transition period, the other kernel pricing policy was that free software couldn't have prerequisite on charged for kernel software. For vm370 release 4, there was decision to release multiprocessor support (as support for hardware, would be part of the free kernel base). The problem was that I had included in my resource manager ... a whole bunch of code that multiprocessor support used. The resolution was that something like 90% of the code in my release 3 resource manager ... was moved into the free release 4 kernel (w/o changing the price of the resource manager). http://www.garlic.com/~lynn/subtopic.html#smp More more software (both kernel non-kernel) was being charged for. vm/370 release 6 was still free ... but had a lower-priced kernel add-on (bsepp, for entry level and midrange customers) that was subset of the higher-priced SEPP (which had absorbed my resource manager and added a bunch of other stuff). As previously mentioned, VM/SP Release 1 marked the end of transition in kernel software pricing, BSEPP/SEPP were merged back into the base kernel ... and the whole thing became charged for (the name change also reflected change, instead of vm/370 release 7 ... it was vm system product release 1 ... i.e. a charged-for product). recent posts: http://www.garlic.com/~lynn/2010d.html#14 Happy DEC-10 Day http://www.garlic.com/~lynn/2010d.html#27 HONE VMSHARE http://www.garlic.com/~lynn/2010d.html#39 search engine history, was Happy DEC-10 Day http://www.garlic.com/~lynn/2010d.html#42 search engine history, was Happy DEC-10 Day http://www.garlic.com/~lynn/2010d.html#43 What was old is new again (water chilled) http://www.garlic.com/~lynn/2010d.html#60 LPARs: More or Less? http://www.garlic.com/~lynn/2010d.html#62 LPARs: More or Less? http://www.garlic.com/~lynn/2010d.html#66 LPARs: More or Less? http://www.garlic.com/~lynn/2010d.html#79 LPARs: More or Less? http://www.garlic.com/~lynn/2010e.html#17 Senior Java Developer vs. MVS Systems Programmer (warning: Conley rant)
Re: z9 / z10 instruction speed(s)
The contention was that IBM shipping capacity that isn't used doesn't make sense. A contention which I disagree with. It's cheaper to build one type of chip/card, and use other methods to limit capacity, which is what software pricing is based on. I knew, in the mid-1980's, when IBM introduced model groups, that capacity-based pricing was going to introduce headaches. I told IBM this, at the time, but they didn't appear to care. My job has been, for over 30 years, to manage, forecast, and configure processors (and other hardware), and I do not like performing unnatural acts to conform to flawed pricing models. That takes away any value I can bring to the business. IBM, and the ISV's, don't get this. Each pricing solution, IBM comes up with, seems to be more flawed (and complex) than the previous one. Unfortunately, when it's the only game in town, you have no choice but to comply. Complaining (especially to IBM), gets you nothing but a chance to vent your spleen. - Too busy driving to stop for gas! -- 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: z9 / z10 instruction speed(s)
Agreed. So the next time I fill up at the gas station the price should be based on horsepower. All the SUV's should pay vastlly more for the same gas that I use for my Honda Civic. I always use high-test since high-octane is always better even for small cars, better mileage and cooler running engines. The grade of gas used affects the temperature of the engine. It is one reason, why airplanes use special high octane gas. Capacity based pricing has nothing but greed written all over it. How can IBM even keep a straight face when they say shipping capacity that isn't used 'doesn't make sense' What doesn't make sense is capacity-based pricing. Rationalization: Completely logical, just not true. On Mon, Mar 8, 2010 at 11:55 AM, Ted MacNEIL eamacn...@yahoo.ca wrote: The contention was that IBM shipping capacity that isn't used doesn't make sense. A contention which I disagree with. It's cheaper to build one type of chip/card, and use other methods to limit capacity, which is what software pricing is based on. I knew, in the mid-1980's, when IBM introduced model groups, that capacity-based pricing was going to introduce headaches. I told IBM this, at the time, but they didn't appear to care. My job has been, for over 30 years, to manage, forecast, and configure processors (and other hardware), and I do not like performing unnatural acts to conform to flawed pricing models. That takes away any value I can bring to the business. IBM, and the ISV's, don't get this. Each pricing solution, IBM comes up with, seems to be more flawed (and complex) than the previous one. Unfortunately, when it's the only game in town, you have no choice but to comply. Complaining (especially to IBM), gets you nothing but a chance to vent your spleen. - Too busy driving to stop for gas! -- 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 -- George Henke (C) 845 401 5614 -- 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: z9 / z10 instruction speed(s)
-Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of George Henke Sent: Monday, March 08, 2010 9:08 AM To: IBM-MAIN@bama.ua.edu Subject: Re: z9 / z10 instruction speed(s) How can anyone have too many mainframe MIPS? Really? Dahh. Just look at SHADOW, the Data Direct Product that enables a TCB workload, through some niffty fancy footwork, to run workloads on a Ziip processor which handles SRB only workloads. I have a former client who automagically extended the CPU life of his 2 GPP (General Purpose Processor) configuration by installing their product. Ziip/Zaap processors are UNTHROTTLED, UNMETERED. Not like GPP's that are both throttled and metered. And wherefore, the IBM suit against NEON, a competing product? Please note well, the IBM suit is not against SHADOW which was as niffty in its design for the Patent Law as it was for the Ziip processor itself. S, if all this knee-capping is s great, why is everyone installling this product so they can get FREE, UNTHROTTLED MIPS? Because everybody wants something for nothing? The basic greediness of most business managers? Let's stop kidding ourselves, IBM and third party vendors are just using every device they can find to get what they want and then ratiionalize it. Textbook definition of rationallzation: Something that is completely logical, just not true. Much of this trail extolling the virtues of knee-capping ad nauseum is nothing more than just that, rationalization. IBM must be chortling at all this, all the way to the bank. I think that see your point. And somewhat agree with it. However, as I recall it being explained to me by another, the tier based pricings started when a big outsourcer (EDS?) would bid for small companies' IT business. EDS was cheaper because they had a huge data center which they would parcel up. But that big data center paid the same software cost as the smaller ones. So, instead on n small accounts (all paying they same), the software vendor ended up with one big account, again paying one license instead of the n small accounts. Is this fair? IMO, no, it is not. Of course, now that software is a bigger slice of the income, the effect is magnified. The solution? Hum, how about charging by the core (aka CPU)? The more CPUs you run a product on, the more it costs. Oh, of course, this means that it would be more profitable to make the CEC consist of a large number of really weak processors. Oh, well, it was an idea. (yeah, a stupid one). Well, then how about a rather small ILC (initial license cost) for a product. Then a maintenance fee for scheduled maintenance and upgrades. And, lastly, a per incident fee for problem resolution. Hum, but then the vendor ships bugger software in order to up the per incident fee. And make you sign an NDA for all solutions otherwise the first person to get a solution would post it here and cut the vendor's repeat business for that solution/incident. Why do (generic) you think Linux is getting to be so popular? __Just__ because it works fairly well? Hell, no. If you really want, you can run it with a total software investment of US $0.00 . You may need to buy customized software. Or write your own. But the entire infrastructure is there for gratis. This means a small business can invest in better hardware instead of balancing hardware vs. software costs. And, once you get big enough to really need support, you're in a better position to get it from Novell, RedHat, or even a number of local consultants. -- John McKown Systems Engineer IV IT Administrative Services Group HealthMarkets(r) 9151 Boulevard 26 * N. Richland Hills * TX 76010 (817) 255-3225 phone * (817)-961-6183 cell john.mck...@healthmarkets.com * www.HealthMarkets.com Confidentiality Notice: This e-mail message may contain confidential or proprietary information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. HealthMarkets(r) is the brand name for products underwritten and issued by the insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake Life Insurance Company(r), Mid-West National Life Insurance Company of TennesseeSM and The MEGA Life and Health Insurance Company.SM -- 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: z9 / z10 instruction speed(s)
-Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of George Henke Sent: Monday, March 08, 2010 11:24 AM To: IBM-MAIN@bama.ua.edu Subject: Re: z9 / z10 instruction speed(s) Agreed. So the next time I fill up at the gas station the price should be based on horsepower. All the SUV's should pay vastlly more for the same gas that I use for my Honda Civic. I always use high-test since high-octane is always better even for small cars, better mileage and cooler running engines. The grade of gas used affects the temperature of the engine. It is one reason, why airplanes use special high octane gas. Capacity based pricing has nothing but greed written all over it. How can IBM even keep a straight face when they say shipping capacity that isn't used 'doesn't make sense' What doesn't make sense is capacity-based pricing. Rationalization: Completely logical, just not true. Well, I think you've hit the nail on the head! We need software pricing to be like gas pricing. The more gas your car uses, the more you pay. I guess the quality of the software would relate to the octane of the gas. You want better software, you'd better want to pay more for it than for the lesser software. Any metering of software needs to be done on the basis of actual usage, not on the theoretical maximum use that you could have run it if you had run it continuously and nothing else. Examples that come to mind: CA-7 price based on number of jobs scheduled per month. CA-1 price based on total number of tape volumes managed. CICS price based on number of transactions, file and database I/O, etc. per month. Hum, how to cost out z/OS itself? -- John McKown Systems Engineer IV IT Administrative Services Group HealthMarkets(r) 9151 Boulevard 26 * N. Richland Hills * TX 76010 (817) 255-3225 phone * (817)-961-6183 cell john.mck...@healthmarkets.com * www.HealthMarkets.com Confidentiality Notice: This e-mail message may contain confidential or proprietary information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. HealthMarkets(r) is the brand name for products underwritten and issued by the insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake Life Insurance Company(r), Mid-West National Life Insurance Company of TennesseeSM and The MEGA Life and Health Insurance Company.SM -- 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: z9 / z10 instruction speed(s)
On Mon, Mar 8, 2010 at 12:24 PM, George Henke gahe...@gmail.com wrote: So the next time I fill up at the gas station the price should be based on horsepower. All the SUV's should pay vastlly more for the same gas that I use for my Honda Civic. Well, they do pay more for their vehicles -- and engine upgrades are more expensive. More CPUs use more electricity, and that costs more. I always use high-test since high-octane is always better even for small cars, better mileage and cooler running engines. The grade of gas used affects the temperature of the engine. It is one reason, why airplanes use special high octane gas. Myth. If you're not getting preignition, you're wasting money. Capacity based pricing has nothing but greed written all over it. How can IBM even keep a straight face when they say shipping capacity that isn't used 'doesn't make sense' If you're quoting me, (a) I'm not IBM and (b) you're the one making that assertion. What doesn't make sense is capacity-based pricing. Well, it does to everyone but you. Not sure how to address that. You seem to want more for less; again, that's not how the world works. Rationalization: Completely logical, just not true. -- 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: z9 / z10 instruction speed(s)
Wow, what an inappropriate analogy. George Henke gahe...@gmail.com 3/8/2010 12:24 PM Agreed. So the next time I fill up at the gas station the price should be based on horsepower. All the SUV's should pay vastlly more for the same gas that I use for my Honda Civic. I always use high-test since high-octane is always better even for small cars, better mileage and cooler running engines. The grade of gas used affects the temperature of the engine. It is one reason, why airplanes use special high octane gas. Capacity based pricing has nothing but greed written all over it. How can IBM even keep a straight face when they say shipping capacity that isn't used 'doesn't make sense' What doesn't make sense is capacity-based pricing. Rationalization: Completely logical, just not true. On Mon, Mar 8, 2010 at 11:55 AM, Ted MacNEIL eamacn...@yahoo.ca wrote: The contention was that IBM shipping capacity that isn't used doesn't make sense. A contention which I disagree with. It's cheaper to build one type of chip/card, and use other methods to limit capacity, which is what software pricing is based on. I knew, in the mid-1980's, when IBM introduced model groups, that capacity-based pricing was going to introduce headaches. I told IBM this, at the time, but they didn't appear to care. My job has been, for over 30 years, to manage, forecast, and configure processors (and other hardware), and I do not like performing unnatural acts to conform to flawed pricing models. That takes away any value I can bring to the business. IBM, and the ISV's, don't get this. Each pricing solution, IBM comes up with, seems to be more flawed (and complex) than the previous one. Unfortunately, when it's the only game in town, you have no choice but to comply. Complaining (especially to IBM), gets you nothing but a chance to vent your spleen. - Too busy driving to stop for gas! -- 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 -- George Henke (C) 845 401 5614 -- 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 CONFIDENTIALITY/EMAIL NOTICE: The material in this transmission contains confidential and privileged information intended only for the addressee. If you are not the intended recipient, please be advised that you have received this material in error and that any forwarding, copying, printing, distribution, use or disclosure of the material is strictly prohibited. If you have received this material in error, please (i) do not read it, (ii) reply to the sender that you received the message in error, and (iii) erase or destroy the material. Emails are not secure and can be intercepted, amended, lost or destroyed, or contain viruses. You are deemed to have accepted these risks if you communicate with us by email. Thank you. -- 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: z9 / z10 instruction speed(s)
George Henke wrote: Capacity based pricing has nothing but greed written all over it. I'm glad IBM and many ISVs offer such steep discounts to smaller customers that choose to enable/use only a subset of the potential processing capacity available on the z10. Without those discounts, many customers would be unable to afford their mainframes. -- Edward E Jaffe Phoenix Software International, Inc 831 Parkview Drive North El Segundo, CA 90245 310-338-0400 x318 edja...@phoenixsoftware.com http://www.phoenixsoftware.com/ -- 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: z9 / z10 instruction speed(s)
Hey Scott, Actually I thought it was quite appropriate of an analogy, just not what George expected. :-) To George's assertion that capacity-based pricing doesn't make sense: I DO pay more for gas for my SUV than I do for my smaller car (not per gallon, but definitely per trip) - simply because I use more gas to get from point A to point B. And to an extent, it is based on capacity, not just the fact that I'm going from A to B. I pay more partly for the capacity to take more people in one trip (i.e. the capacity is higher). Do I always use that extra capacity? No, but it is there when I want to use it. The bigger the capacity of the vehicle, the more I paid for it up front, and the more I pay on an on-going basis to keep it running, supported, and maintained. That's different from the computer kneecapping, how? If knee-capped boxes weren't made available for smaller shops like mine, we would have been off the mainframe years ago. And capacity-based pricing is a metric used across the board, not just by IBM on mainframes. Go and price out a windows server and see if you get an 8-way server for the same price as a one-way. Or try to buy a 128 core SUN server for the same price as a 2 core SUN workstation. Or back to your car analogy, why on earth would I pay for a 48 passenger commercial bus to take my family of 4 to church Sunday morning? You buy what you need to get the job done, preferably without paying a bunch of extra money for capacity and features you will never need. Back to sitting on the sidelines watching the debate. :-) Rex -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Scott Rowe Sent: Monday, March 08, 2010 12:32 PM To: IBM-MAIN@bama.ua.edu Subject: Re: z9 / z10 instruction speed(s) Wow, what an inappropriate analogy. George Henke gahe...@gmail.com 3/8/2010 12:24 PM Agreed. So the next time I fill up at the gas station the price should be based on horsepower. All the SUV's should pay vastlly more for the same gas that I use for my Honda Civic. I always use high-test since high-octane is always better even for small cars, better mileage and cooler running engines. The grade of gas used affects the temperature of the engine. It is one reason, why airplanes use special high octane gas. Capacity based pricing has nothing but greed written all over it. How can IBM even keep a straight face when they say shipping capacity that isn't used 'doesn't make sense' What doesn't make sense is capacity-based pricing. Rationalization: Completely logical, just not true. -- 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: z9 / z10 instruction speed(s)
It is comforting to know that satire is alive and well in IT. How boring life would be without it. However, as I recall it being explained to me by another, the tier based pricings started when a big outsourcer (EDS?) would bid for small companies' IT business. EDS was cheaper because they had a huge data center which they would parcel up. But that big data center paid the same software cost as the smaller ones. So, instead on n small accounts (all paying they same), the software vendor ended up with one big account, again paying one license instead of the n small accounts. Is this fair? IMO, no, it is not. Of course, now that software is a bigger slice of the income, the effect is magnified. It is called economies of scale, not price gouging, and yes, it is fair for EDS (as much as I don't like to admit it) to realize the efficiencies of such economies of scale because they have incurred the business risk of building said large installations and are justified in trying to improve their utilization, efficiency, to defray that risk. What is not just or equitable is for IBM to view EDS' efficiency and profits from economies of scale as a loss of revenue to themselves (IBM) and then create a pricing scheme that appropriates those profits from such efficiencies for themselves. Since when have monopolies ever been interested in efficiencies except for their own profitability. Monopolies do not optimize the efficient allocation of resources in the economy. They maximize their own profitablity at the expense of efficient allocation of resources in the economy. The is the basic evil of monopolies (Microeconmics 101). It is opposite of The greatest good for the greatest number. On Mon, Mar 8, 2010 at 1:32 PM, Scott Rowe scott.r...@joann.com wrote: Wow, what an inappropriate analogy. George Henke gahe...@gmail.com 3/8/2010 12:24 PM Agreed. So the next time I fill up at the gas station the price should be based on horsepower. All the SUV's should pay vastlly more for the same gas that I use for my Honda Civic. I always use high-test since high-octane is always better even for small cars, better mileage and cooler running engines. The grade of gas used affects the temperature of the engine. It is one reason, why airplanes use special high octane gas. Capacity based pricing has nothing but greed written all over it. How can IBM even keep a straight face when they say shipping capacity that isn't used 'doesn't make sense' What doesn't make sense is capacity-based pricing. Rationalization: Completely logical, just not true. On Mon, Mar 8, 2010 at 11:55 AM, Ted MacNEIL eamacn...@yahoo.ca wrote: The contention was that IBM shipping capacity that isn't used doesn't make sense. A contention which I disagree with. It's cheaper to build one type of chip/card, and use other methods to limit capacity, which is what software pricing is based on. I knew, in the mid-1980's, when IBM introduced model groups, that capacity-based pricing was going to introduce headaches. I told IBM this, at the time, but they didn't appear to care. My job has been, for over 30 years, to manage, forecast, and configure processors (and other hardware), and I do not like performing unnatural acts to conform to flawed pricing models. That takes away any value I can bring to the business. IBM, and the ISV's, don't get this. Each pricing solution, IBM comes up with, seems to be more flawed (and complex) than the previous one. Unfortunately, when it's the only game in town, you have no choice but to comply. Complaining (especially to IBM), gets you nothing but a chance to vent your spleen. - Too busy driving to stop for gas! -- 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 -- George Henke (C) 845 401 5614 -- 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 CONFIDENTIALITY/EMAIL NOTICE: The material in this transmission contains confidential and privileged information intended only for the addressee. If you are not the intended recipient, please be advised that you have received this material in error and that any forwarding, copying, printing, distribution, use or disclosure of the material is strictly prohibited. If you have received this material in error, please (i) do not read it, (ii) reply to the sender that you received the message in error, and (iii) erase or destroy the material. Emails are not secure and can be intercepted, amended, lost or destroyed, or contain viruses. You are
Re: z9 / z10 instruction speed(s) -- OCTANE of CPUs
SNIPPAGE So the next time I fill up at the gas station the price should be based on horsepower. All the SUV's should pay vastlly more for the same gas that I use for my Honda Civic. I always use high-test since high-octane is always better even for small cars, better mileage and cooler running engines. The grade of gas used affects the temperature of the engine. It is one reason, why airplanes use special high octane gas. SNIPPAGE It may be one reason, but it is not the defining reason. 80 Octane is not readily available in the US any longer, so 100LL is what is used. 100LL (Low Lead -- as in TetraEthyl Lead) is needed to avoid detonation in high horsepower engines. The other octane levels are gone (unless on a military field, but since I don't operate there, I can't really say what their octane ratings are). And this is why there are people pushing for AutoGas STCs, but requiring that the autogas NOT have alcohol in it (since it attracts moisture, and climbing out to 10,000 ft MSL can mean the outside temp is -30C in the winter guess what happens). sore subject for me But today, even the low end auto engines have detonation sensors to change the timing and mixture. So running premium is a waste. The idea that we need different CPUs to allow for different workloads is a bit twisted. That the clock speed would be different, or the internal cache would be different, etc. is understandable. But, using certain arguments, and apparently IBM does, we do not see mainframes in the 10-20 MIP range with embedded disk drives, etc. (similar to the P/390 on steroids, MP/3000, or the FLEX/ES machines, or PSIs offerings). We don't see this because it is not in IBM's best interest. So we sell you the same machine, but at a lower cost because we disabled it, than the one we sell to the top level power users. Isn't this socialism in effect in computer systems? Seems that Bausch ran into this problem when they were doing the same thing (effectively) for contacts. If you paid more for your contacts, you could wear them for 3 months. Otherwise, you paid less for a box of contacts, but could only wear them for a week at a time. Regards, Steve Thompson -- Opinions expressed by this poster do not necessarily reflect those of poster's employer -- -- 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: z9 / z10 instruction speed(s)
So the next time I fill up at the gas station the price should be based on horsepower. Actually, that analogy is a bit flawed. Higher horsepower requires, usually, more gas/octane. So, at the same price per gallon, the more powerful vehicles are already 'penalised'. - Too busy driving to stop for gas! -- 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: z9 / z10 instruction speed(s)
The higher octane is usually required to prevent preignition/detonation. A high performance engine (hp/displacement) can require a higher octane even with todays engine management computers. Note that this is not limited to the high horsepower engines, even a economy car can require a high octane fuel. Many engines today with forced induction (superchargers/turbochargers) and even those with a high compression ratio will require something better than regular gas (US terms). -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Ted MacNEIL Sent: Monday, March 08, 2010 3:15 PM To: IBM-MAIN@bama.ua.edu Subject: Re: [IBM-MAIN] z9 / z10 instruction speed(s) So the next time I fill up at the gas station the price should be based on horsepower. Actually, that analogy is a bit flawed. Higher horsepower requires, usually, more gas/octane. So, at the same price per gallon, the more powerful vehicles are already 'penalised'. - Too busy driving to stop for gas! -- 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 -- 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: z9 / z10 instruction speed(s)
The following message is a courtesy copy of an article that has been posted to bit.listserv.ibm-main,alt.folklore.computers as well. eamacn...@yahoo.ca (Ted MacNEIL) writes: A contention which I disagree with. It's cheaper to build one type of chip/card, and use other methods to limit capacity, which is what software pricing is based on. aka there is a lot of upfront fixed costs ... but volume manufacturing techniques frequently drop the bottom out of per unit costs (i.e. per unit price can be dominated by the upfront fixed costs, leveraging common unit built in larger volumes can easily offset having multiple custom designed items). it actually costs both the vendor and the customers quite a bit to physically change item ... potentially significantly more than bare bones per unit (volume) manufacturing costs ... as a result having large number of units prestaged ... is trade-off of the extra volume manufacturing cost of each of the units against the vendorcustomer change cost of physically adding/replacing each individual item. it is somewhat the change-over to 3rd wave (information age). Earlier, the cost ... and therefor perceived value, was mostly in the actual building of something. moving into the 3rd wave, much more of the value has moved to the design of something ... and volume manufacturing techniques has frequently reduced the per unit building cost as close as possible to zero. They are now doing multi-billion dollar chip plants that are obsolete in a few years. Manufacturing cost is the actual creation of the wafer ... with thousands of chips cut from each wafer (motivating move from 8in wafer to 12in wafer, getting more chips per wafer). The bare-bones cost for building one additional chip ... can be a couple pennies ... however, the chip price may be set at a couple hundred (or more) in order to recover the cost of the upfront chip design as well as the cost of the plant. It may then cost the vendorcustomer, tens (or hundreds) of dollars to actually physically deploy each chip where it is useful. An economic alternative is to package a large number of chips in a single deployment ... potentially at loss of a few cents per chip ... in the anticipation that the extra chips might be needed at some point (possibly being able to eliminate cost of actually having to physical deploy each individual chip). note that the pharmaceutical industry has been going thru similar scenarios with brand drugs (with upfront development costs) and generic drugs. something similar was used as justification for the FS project ... the corporate RD costs was significantly higher than the vendors turning out clone controllers ... including the one I worked on as undergraduate http://www.garlic.com/~lynn/subtopic.html#360pcm then the distraction of the FS effort (and drying up 370 product pipelines) is then blamed for allowing clone processors to gain market foothold. http://www.garlic.com/~lynn/submain.html#futuresys Some discussion here: http://www.jfsowa.com/computer/memo125.htm Article by corporate executive involved in FS effort: http://www.ecole.org/Crisis_and_change_1995_1.htm quote from above: IBM tried to react by launching a major project called the 'Future System' (FS) in the early 1970's. The idea was to get so far ahead that the competition would never be able to keep up, and to have such a high level of integration that it would be impossible for competitors to follow a compatible niche strategy. However, the project failed because the objectives were too ambitious for the available technology. Many of the ideas that were developed were nevertheless adapted for later generations. Once IBM had acknowledged this failure, it launched its 'box strategy', which called for competitiveness with all the different types of compatible sub-systems. But this proved to be difficult because of IBM's cost structure and its RD spending, and the strategy only resulted in a partial narrowing of the price gap between IBM and its rivals. ... snip ... There have been some comments that the baroque nature of the pu4/pu5 (vtam/3705ncp) interface, did try approximate the FS high level of integration objective. -- 42yrs virtualization experience (since Jan68), online at home since Mar1970 -- 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: z9 / z10 instruction speed(s)
Yes, using the quantity of gas required, rather than the unit cost makes much more sense. Pommier, Rex R. rex.pomm...@cnasurety.com 3/8/2010 2:22 PM Hey Scott, Actually I thought it was quite appropriate of an analogy, just not what George expected. :-) To George's assertion that capacity-based pricing doesn't make sense: I DO pay more for gas for my SUV than I do for my smaller car (not per gallon, but definitely per trip) - simply because I use more gas to get from point A to point B. And to an extent, it is based on capacity, not just the fact that I'm going from A to B. I pay more partly for the capacity to take more people in one trip (i.e. the capacity is higher). Do I always use that extra capacity? No, but it is there when I want to use it. The bigger the capacity of the vehicle, the more I paid for it up front, and the more I pay on an on-going basis to keep it running, supported, and maintained. That's different from the computer kneecapping, how? If knee-capped boxes weren't made available for smaller shops like mine, we would have been off the mainframe years ago. And capacity-based pricing is a metric used across the board, not just by IBM on mainframes. Go and price out a windows server and see if you get an 8-way server for the same price as a one-way. Or try to buy a 128 core SUN server for the same price as a 2 core SUN workstation. Or back to your car analogy, why on earth would I pay for a 48 passenger commercial bus to take my family of 4 to church Sunday morning? You buy what you need to get the job done, preferably without paying a bunch of extra money for capacity and features you will never need. Back to sitting on the sidelines watching the debate. :-) Rex -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Scott Rowe Sent: Monday, March 08, 2010 12:32 PM To: IBM-MAIN@bama.ua.edu Subject: Re: z9 / z10 instruction speed(s) Wow, what an inappropriate analogy. George Henke gahe...@gmail.com 3/8/2010 12:24 PM Agreed. So the next time I fill up at the gas station the price should be based on horsepower. All the SUV's should pay vastlly more for the same gas that I use for my Honda Civic. I always use high-test since high-octane is always better even for small cars, better mileage and cooler running engines. The grade of gas used affects the temperature of the engine. It is one reason, why airplanes use special high octane gas. Capacity based pricing has nothing but greed written all over it. How can IBM even keep a straight face when they say shipping capacity that isn't used 'doesn't make sense' What doesn't make sense is capacity-based pricing. Rationalization: Completely logical, just not true. -- 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 CONFIDENTIALITY/EMAIL NOTICE: The material in this transmission contains confidential and privileged information intended only for the addressee. If you are not the intended recipient, please be advised that you have received this material in error and that any forwarding, copying, printing, distribution, use or disclosure of the material is strictly prohibited. If you have received this material in error, please (i) do not read it, (ii) reply to the sender that you received the message in error, and (iii) erase or destroy the material. Emails are not secure and can be intercepted, amended, lost or destroyed, or contain viruses. You are deemed to have accepted these risks if you communicate with us by email. Thank you. -- 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: z9 / z10 instruction speed(s)
-snip Given the axioms of the discussion, it would be an entire book. The contention was that IBM shipping capacity that isn't used doesn't make sense. unsnip- But it makes excellent sense in the context of failure and recovery. Having a spare engine to switch over to in the event of a failure in the primaty engine(s) can make a HUGE difference for a shop that needs to maximize availability. Rick -- 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: z9 / z10 instruction speed(s)
On Mon, Mar 8, 2010 at 3:43 PM, Rick Fochtman rfocht...@ync.net wrote: -snip Given the axioms of the discussion, it would be an entire book. The contention was that IBM shipping capacity that isn't used doesn't make sense. unsnip- But it makes excellent sense in the context of failure and recovery. Having a spare engine to switch over to in the event of a failure in the primaty engine(s) can make a HUGE difference for a shop that needs to maximize availability. Sure, and for the other reasons elaborated upon at length here. The point is, that seemed to be George Henke's contention. -- 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: z9 / z10 instruction speed(s)
Radoslaw Skorupka writes: Can I ask where are new customers of this very affordable platform? Many (though not all) are in the developing world. That's why it's called developing, I guess. :-) Some places need new credit card processing systems (to pick an example) because they don't have them. Other places have been processing credit cards for decades, and if they are growing it's horizontally (adding functions, optimizing, etc. -- new workloads). To use a somewhat flawed analogy, President Obama recently announced that a company in the U.S. intends to build 2 new nuclear power plants. That's not so impressive when you consider that China alone has about 50 new nuclear plants either under construction or in the advanced planning stage. Of course, when a few of the new mainframe customers choose to reveal themselves (HDFC, BC Card...) there are about 20 posts questioning reality and another 20 concerning the pros and cons of international trade (or something like that). :-) Also, I think it's safe to say IBM would always like more new mainframe customers. All that said, I should repeat that it's long past time mainframe pricing perceptions, here and elsewhere, caught up with reality. - - - - - Timothy Sipples IBM Consulting Enterprise Architect for New, Advanced, and/or Innovative Solutions (VCT) Based in Singapore Serving the Growth Markets E-Mail: timothy.sipp...@us.ibm.com -- 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: z9 / z10 instruction speed(s)
Timothy Sipples wrote: Radoslaw Skorupka writes: Can I ask where are new customers of this very affordable platform? Many (though not all) are in the developing world. That's why it's called developing, I guess. :-) Some places need new credit card processing systems (to pick an example) because they don't have them. Other places have been processing credit cards for decades, and if they are growing it's horizontally (adding functions, optimizing, etc. -- new workloads). But the OP wrote: - But Tim, there *aren't* any smaller mainframe customers any more (at least not in the USA). - I would guess the focus here was on jobs, plain and simple. Maybe installed mainframe MIPS are increasing, but jobs, especially for z/OS staff, appear to be declining (esp. in the US, but some on the list have mentioned similar trends in Europe). To use a somewhat flawed analogy, President Obama recently announced that a company in the U.S. intends to build 2 new nuclear power plants. That's not so impressive when you consider that China alone has about 50 new nuclear plants either under construction or in the advanced planning stage. Of course, when a few of the new mainframe customers choose to reveal themselves (HDFC, BC Card...) there are about 20 posts questioning reality and another 20 concerning the pros and cons of international trade (or something like that). :-) Also, I think it's safe to say IBM would always like more new mainframe customers. I'm sure. But clearly they would prefer them to be Linux on z mainframe customers. I say clearly just reflecting the crazy pricing wrt zIIPs and zAAPs and zNALC which provide lower prices for use of Java, WebSphere, DB2 etc. than for existing apps in COBOL, Assembler, PL/I, and the like. This seems like a good place to get your customers in preparation for migrating off z/OS because I suspect apps are easier to port to Linux if they are of the Java / WebSphere / DB2 ilk than if they are in the COBOL / IMS / CICS style. All that said, I should repeat that it's long past time mainframe pricing perceptions, here and elsewhere, caught up with reality. It would help if IBM would tell the story, then, wouldn't it? An old thread: IBM doesn't seem interested in winning the hearts and minds battle for the mainframe (esp. z/OS), based on lack of articles and ads promoting and raising awareness of z/OS or even z Series. A strong campaign would go a long way towards getting perceptions caught up with reality, eh? - - - - - Timothy Sipples IBM Consulting Enterprise Architect for New, Advanced, and/or Innovative Solutions (VCT) Based in Singapore Serving the Growth Markets E-Mail: timothy.sipp...@us.ibm.com -- 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 -- Kind regards, -Steve Comstock The Trainer's Friend, Inc. 303-393-8716 http://www.trainersfriend.com * z/OS application programmer training + Instructor-led on-site classroom based classes + Course materials licensing + Remote contact training + Roadshows + Course development -- 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: z9 / z10 instruction speed(s)
On 8 March 2010 15:43, Rick Fochtman rfocht...@ync.net wrote: -snip Given the axioms of the discussion, it would be an entire book. The contention was that IBM shipping capacity that isn't used doesn't make sense. unsnip- But it makes excellent sense in the context of failure and recovery. Having a spare engine to switch over to in the event of a failure in the primaty engine(s) can make a HUGE difference for a shop that needs to maximize availability. Clearly you are thinking of the Citroën Sahara... One engine in the front, and one in the back, independently driving front and rear wheels, but with one set of controls. Wikipedia claims there are only 27 left in the world. Tony H. -- 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: z9 / z10 instruction speed(s)
On Mon, Mar 8, 2010 at 3:45 PM, zMan zedgarhoo...@gmail.com wrote: On Mon, Mar 8, 2010 at 3:43 PM, Rick Fochtman rfocht...@ync.net wrote: -snip Given the axioms of the discussion, it would be an entire book. The contention was that IBM shipping capacity that isn't used doesn't make sense. unsnip- But it makes excellent sense in the context of failure and recovery. Having a spare engine to switch over to in the event of a failure in the primaty engine(s) can make a HUGE difference for a shop that needs to maximize availability. Sure, and for the other reasons elaborated upon at length here. The point is, that seemed to be George Henke's contention. OK, too many pronouns even for me to understand my own post. Let me try again: Failover is but one of many reasons why not using every CPU in the book makes sense. However, George Henke's contention is that kneecapping CPUs is somehow wrong -- that not using all the available speed is a bad idea. Taken to its logical conclusion, not firing up every CPU in the book is also somehow wrong. -- 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: z9 / z10 instruction speed(s)
Failover is but one of many reasons why not using every CPU in the book makes sense. However, George Henke's contention is that kneecapping CPUs is somehow wrong -- that not using all the available speed is a bad idea. Taken to its logical conclusion, not firing up every CPU in the book is also somehow wrong From a purely operational point of view, setting aside any pricing constraints, why would I NOT use every CP available? Why would I NOT use all the memory available? Shall we vary some memory offline now because we have too much 64-bit memory. There was a time, not too long ago, when Directors of Computer Resources would indeed vary resources like DASD and CPs offline and hold them in reserve for future growth and to avoid the needless proliferation of data, wasting of DASD and MIPS. Data, like nature, seems to abhor a vacuum and occupy any empty space and gobble up any spare CPU cycles just because it is there.. But that practice is fast disappearing now and, in any case, it was always a discretionary choice on the part of IT management and not an outside constraint imposed by IBM or software vendors. On Mon, Mar 8, 2010 at 6:16 PM, zMan zedgarhoo...@gmail.com wrote: On Mon, Mar 8, 2010 at 3:45 PM, zMan zedgarhoo...@gmail.com wrote: On Mon, Mar 8, 2010 at 3:43 PM, Rick Fochtman rfocht...@ync.net wrote: -snip Given the axioms of the discussion, it would be an entire book. The contention was that IBM shipping capacity that isn't used doesn't make sense. unsnip- But it makes excellent sense in the context of failure and recovery. Having a spare engine to switch over to in the event of a failure in the primaty engine(s) can make a HUGE difference for a shop that needs to maximize availability. Sure, and for the other reasons elaborated upon at length here. The point is, that seemed to be George Henke's contention. OK, too many pronouns even for me to understand my own post. Let me try again: Failover is but one of many reasons why not using every CPU in the book makes sense. However, George Henke's contention is that kneecapping CPUs is somehow wrong -- that not using all the available speed is a bad idea. Taken to its logical conclusion, not firing up every CPU in the book is also somehow wrong. -- 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 -- George Henke (C) 845 401 5614 -- 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: z9 / z10 instruction speed(s)
why would I NOT use every CP available? I think this is a strawman argument. The CP's are installed, but not available. The microcode determines what is available. And, the real reason (as unpopular as it may be), is cost. You didn't pay to see those cards. - Too busy driving to stop for gas! -- 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: z9 / z10 instruction speed(s)
And, the real reason (as unpopular as it may be), is cost. No, the real reason is manipulation of supply and demand by a monopoly. Perhaps some remember the days of the Cabbage Patch dolls. The manufacturer limited production while aggressively creating within every young girl the uncontrollable desire for a Cabbage Patch doll. The result: A sky high price for a rather odd looking doll that no one would have otherwise ever bought in the first place. There has always been limited editions of all kinds of things: Hummel figures, China, artistic prints of paintings, etc. But do we really want the mainframe world to work that way? On Mon, Mar 8, 2010 at 6:48 PM, Ted MacNEIL eamacn...@yahoo.ca wrote: why would I NOT use every CP available? I think this is a strawman argument. The CP's are installed, but not available. The microcode determines what is available. And, the real reason (as unpopular as it may be), is cost. You didn't pay to see those cards. - Too busy driving to stop for gas! -- 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 -- George Henke (C) 845 401 5614 -- 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: z9 / z10 instruction speed(s)
Clock speed is only a small piece of the picture. Other major factors are how many clock cycles are required to execute an instruction, how many bits or bytes can be manipulated by a single instruction, how many instructions can concurrently be in the execution pipeline, how many independent ports there are to memory, the width in bits/bytes of the memory and I/O bus paths, etc. Even 30 years ago, I/O channels on all but the smallest mainframes were essentially independent processors in their own right with independent memory access. The number of independent channels and aggregate I/O transfer rate on a z/10 today can easily outpace a PC by several orders of magnitude and the I/O data transfers don't degrade the general purpose CP's effective speed. When PCs were promoting the 16-bit to 32-bit architecture revolution a number of years back, larger mainframes had already been exploiting internal memory bus widths of 128-bits or more for over a decade. Typically the mainframes performance has been optimized for high data transfer rates, manipulating multi-byte data values (larger integer and floating point values, data records) and multi-tasking, and a single instruction may do some very complex data manipulations that would require many PC instructions. The PC architecture has historically been more concerned with optimizing manipulation of individual bits and bytes, a capability indispensable for image processing and driving pretty graphical user display interfaces. And, before you can talk about doing useful application-level data manipulation, that graphical display interface on the typical PC can suck up a significant portion of the available CPU power and available memory. Consumption of additional PC CPU resources is imposed by limitations of MS Operating systems, for such things as Anti-Virus protection applications which are unnecessary in a mainframe environment. A mainframe, even one with a slower clock speed, can typically run circles around a PC when dealing with a large number of independent users or with applications that are I/O intensive. On the other hand, a PC of comparable clock speed can do a single CPU intensive task where individual bit or byte manipulations predominate in less real time than the typical mainframe. JC Ewing On 03/06/2010 07:44 PM, Lindy Mayfield wrote: This has been a very interesting thread for me. If I remember correctly from the time I saw the z/10 with plexiglass outsides and a hardware guy there to explain what was what, and one of the things he told me was the cpu speed was (IIRC) 4.77GHz. My laptop has a dual core 1.99GHz. You already see where I'm going with this. How does a z/10 get so much more done? Forgive me for my math and analytical skills, but seems like 4 laptops could equal the speed of a 1 cpu z/10. There is a huge chunk of this equation that I'm totally missing. How does a z/10 get so much more done? kind regards, Lindy -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of R.S. Sent: 7. maaliskuuta 2010 2:23 To: IBM-MAIN@bama.ua.edu Subject: Re: z9 / z10 instruction speed(s) Edward Jaffe pisze: [...] People with PC-only experience are always astonished when I tell them about modern mainframe provisioning capabilities. They always assume when your hard drive fills up you need a new one or when your CPU is too slow you need a new one. What we do seems like magic to them. Yes, mainframe capabilities are excellent in this area. From the other hand they solve problems which exist only in mainframe world: CPU power adjustment. CPU shortage is bad thing on any platform, but mainframe is the only one where too much MIPS is not good. Why to downgrade a PC? The same apply to specialty processors. -- Joel C. Ewing, Fort Smith, ARjremoveccapsew...@acm.org -- 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: z9 / z10 instruction speed(s)
On 3/8/2010 7:15 PM, George Henke wrote: And, the real reason (as unpopular as it may be), is cost. No, the real reason is manipulation of supply and demand by a monopoly. Perhaps some remember the days of the Cabbage Patch dolls. The manufacturer limited production while aggressively creating within every young girl the uncontrollable desire for a Cabbage Patch doll. The result: A sky high price for a rather odd looking doll that no one would have otherwise ever bought in the first place. In the eighties I worked for CCSI in Washington, DC, and we had a contract to print batches of personalized birth/adoption certificates for these dolls (on a Xerox 4050). I can't help but wonder whether this personalization had something to do with the popularity of the dolls, as well as the fact that each doll was unique; I'm not aware of other dolls using that strategy. Gerhard Postpischil Bradford, VT -- 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: z9 / z10 instruction speed(s)
Peter Farley writes: But Tim, there *aren't* any smaller mainframe customers any more (at least not in the USA). There are, and very many. I don't know where you're getting your information. In my previous post it should be obvious that smaller means anybody with capacity below a hypothetical non-kneecapped 7-way System z9 BC. That includes the very smallest customers up through...well, that forms the huge *majority* of customers! By the way, IBM substantially reduced the price of z/VSE for many/most customers (including smaller ones) via the new MWLC sub-capacity licensing for Version 4, so again I have no idea where you're getting your information. Same with z/VM: IBM has slashed the price, big time. A zNALC z/OS license at 3 MSUs starts way, way, way down in the 3 figures per month U.S. (And then there are the Solution Edition offerings.) IBM has never done anything like that before. Have mainframe technologies ever been more affordable? Heck no. (I've looked up a lot of historical prices.) And it's long past time we stop perpetuating mainframe pricing myths, because it isn't helping anyone. AFAIK, there is no way for IBM to bring z/OS on a z10 to a customer with a yearly IT budget in the 5- to (very) low 6-figure range, including software and staffing costs and necessary 3rd-party software. That sounds like an annual IT budget of about $150,000 U.S. (to be generous). And you're right: when it comes to a dedicated/on-site mainframe installation (or, gosh, business computing in general), that's a very tight budget. (However, IBM and/or other organizations could likely provide such an organization with hosted/cloud z/OS services well within that budget.) But I'm not sure when those greener pastures and fonder memories ever actually existed, because $150,000 in 2010 dollars didn't go too far in the 1970s, 1980s, and 1990s either. For example, you would be very fortunate to hire two U.S. FTEs for your IT organization within that budget, but then you'd have no money left over. For reference, when you convert $150,000 in 2009 to 1975, you get $38,040. That is, a total IT budget of $38,040 in 1975 would be equivalent, in Consumer Price Index terms, to a $150,000 budget in 2009. I don't know Did anybody work for an organization in 1975 which had a total annual IT budget of $38,040 or thereabouts...and which owned and operated a dedicated mainframe on-site with licensed software? I think that would have been much *tougher* than it is in 2010. (I remember when Hershey bars cost a nickel) By the way, a lot of businesses cannot afford to buy a tractor-trailer truck either. Between the truck itself, maintenance, driver, fuel, insurance, tolls, etc. it costs a lot more than $150K per year to own and operate one. Yet there are huge numbers of them plying American roads. IBM does not recommend a dedicated/on-site mainframe to *every* business. Just to many/more of them. Charles Mills writes: Now there you go being logical again. Goodness knows I try. :-) But I speak only for myself, so maybe that's why I'm logical. :-) - - - - - Timothy Sipples IBM Consulting Enterprise Architect for New, Advanced, and/or Innovative Solutions (VCT) Based in Singapore Serving the Growth Markets E-Mail: timothy.sipp...@us.ibm.com -- 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: z9 / z10 instruction speed(s)
On 7 March 2010 06:35, Timothy Sipples timothy.sipp...@us.ibm.com wrote: For reference, when you convert $150,000 in 2009 to 1975, you get $38,040. That is, a total IT budget of $38,040 in 1975 would be equivalent, in Consumer Price Index terms, to a $150,000 budget in 2009. I don't know Did anybody work for an organization in 1975 which had a total annual IT budget of $38,040 or thereabouts...and which owned and operated a dedicated mainframe on-site with licensed software? I think that would have been much *tougher* than it is in 2010. (I remember when Hershey bars cost a nickel) I don't know about a total IT budget of $38k, but in 1975 licensed software was pretty much a novelty. The first priced version of MVS (or any other IBM OS except perhaps ACP/TPF?) had yet to appear, and most software was written in house. Some shops were using priced IBM software like PL/I Optimizer, or the COBOL equivalent, and bigger places ran non-free IMS and/or CICS. And there were priced (duh) products from other vendors, like Syncsort. But buying run-the-business application packages was pretty rare. Tony H. -- 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: z9 / z10 instruction speed(s)
--SNIP- Hershey bars cost a nickel) I don't know about a total IT budget of $38k, but in 1975 licensed software was pretty much a novelty. The first priced version of MVS (or any other IBM OS except perhaps ACP/TPF?) had yet to appear, and most software was written in house. Some shops were using priced IBM software like PL/I Optimizer, or the COBOL equivalent, and bigger places ran non-free IMS and/or CICS. And there were priced (duh) products from other vendors, like Syncsort. But buying run-the-business application packages was pretty rare. Tony H. Tony: What about payroll applications. I seem to recall back then companies ran that application and just about nothing else. My memory is iffy here but I think UCC (or was it a precursor?) had a huge payroll package. Russ could verify that. My memory is iffy here but most(all?) of the UCC payroll package was COBOL. There may have been 1 or 2 other PAYROLL applications but couldn't come up with names. I worked at a bank and we had a dedicated mod 50 (IIRC) just for online savings TP system and that was all that was run on it. We also had a mod 40 or 30 (do not remember) we had for credit card system which I think was part of a nationwide system (precursor to Master card Visa) Town Country charge seems right but do not remember. There were other systems besides CICS IMS is what I am saying doing decent amount of work. Ed -- 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: z9 / z10 instruction speed(s)
On 3/8/2010 1:03 AM, Tony Harminc wrote: I don't know about a total IT budget of $38k, but in 1975 licensed software was pretty much a novelty. The first priced version of MVS (or any other IBM OS except perhaps ACP/TPF?) had yet to appear, and most software was written in house. I worked for ADR from 1966 to 1972, and in that time Autoflow, the Librarian, and ROSCOE sold pretty well. There were also some niche products (e.g., Tally) that did well. Gerhard Postpischil Bradford, VT -- 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: z9 / z10 instruction speed(s)
Timothy Sipples pisze: [...] By the way, IBM substantially reduced the price of z/VSE for many/most customers (including smaller ones) via the new MWLC sub-capacity licensing for Version 4, so again I have no idea where you're getting your information. Same with z/VM: IBM has slashed the price, big time. A zNALC z/OS license at 3 MSUs starts way, way, way down in the 3 figures per month U.S. (And then there are the Solution Edition offerings.) IBM has never done anything like that before. Have mainframe technologies ever been more affordable? Heck no. (I've looked up a lot of historical prices.) And it's long past time we stop perpetuating mainframe pricing myths, because it isn't helping anyone. Sounds very reasonable. Can I ask where are new customers of this very affordable platform? Did anyone observed mass migration to z Series? Any crowd in mainframe shop? :-( -- 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 01.01.2009 r. kapita zakadowy BRE Banku SA (w caoci wpacony) wynosi 118.763.528 zotych. W zwizku z realizacj warunkowego podwyszenia kapitau zakadowego, na podstawie uchway XXI WZ z dnia 16 marca 2008r., oraz uchway XVI NWZ z dnia 27 padziernika 2008r., moe ulec podwyszeniu do kwoty 123.763.528 z. Akcje w podwyszonym kapitale zakadowym BRE Banku SA bd w caoci opacone. -- 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: z9 / z10 instruction speed(s)
The following message is a courtesy copy of an article that has been posted to bit.listserv.ibm-main,alt.folklore.computers as well. jmfbahciv jmfbah...@aol writes: ROTFLMAO. A typing fo-paw? re: http://www.garlic.com/~lynn/2010e.html#47 z9 / z10 instruction speed(s) yep ... oh well .. s/invented/invited/ note that it was in 1990 ... twenty years ago. we would go by somers and discuss with some of the occupants that the company (especially mainframe business) was facing similar issues; they would all essentially agree and (also) be able to clearly articulate the issues ... and then we would go back the next month or a couple months ... and nothing had changed. there seemed to be a strong sense that they were (also) trying to preserve the status quo until their retirement, leaving corrective action to somebody else. then the company went into the red ... and some of the status quo and vested interests were being forced to change (compared to auto industry that managed to preserve status guo vested interests across years and years in the red). misc posts mentioning auto C4 /or past protectionism import quotas http://www.garlic.com/~lynn/2000f.html#43 Reason Japanese cars are assembled in the US (was Re: American bigotry) http://www.garlic.com/~lynn/2003i.html#61 TGV in the USA? http://www.garlic.com/~lynn/2004c.html#51 [OT] Lockheed puts F-16 manuals online http://www.garlic.com/~lynn/2004h.html#22 Vintage computers are better than modern crap ! http://www.garlic.com/~lynn/2006g.html#20 The Pankian Metaphor http://www.garlic.com/~lynn/2006m.html#49 The Pankian Metaphor (redux) http://www.garlic.com/~lynn/2007f.html#50 The Perfect Computer - 36 bits? http://www.garlic.com/~lynn/2007g.html#29 The Perfect Computer - 36 bits? http://www.garlic.com/~lynn/2007g.html#34 U.S. Cedes Top Spot in Global IT Competitiveness http://www.garlic.com/~lynn/2007g.html#52 U.S. Cedes Top Spot in Global IT Competitiveness http://www.garlic.com/~lynn/2007i.html#13 U.S. Cedes Top Spot in Global IT Competitiveness http://www.garlic.com/~lynn/2007j.html#33 IBM Unionization http://www.garlic.com/~lynn/2007n.html#31 IBM obsoleting mainframe hardware http://www.garlic.com/~lynn/2008.html#84 Toyota Sales for 2007 May Surpass GM http://www.garlic.com/~lynn/2008.html#85 Toyota Sales for 2007 May Surpass GM http://www.garlic.com/~lynn/2008c.html#22 Toyota Beats GM in Global Production http://www.garlic.com/~lynn/2008c.html#68 Toyota Beats GM in Global Production http://www.garlic.com/~lynn/2008e.html#31 IBM announced z10 ..why so fast...any problem on z 9 http://www.garlic.com/~lynn/2008f.html#50 Toyota's Value Innovation: The Art of Tension http://www.garlic.com/~lynn/2008h.html#65 Is a military model of leadership adequate to any company, as far as it based most on authority and discipline? http://www.garlic.com/~lynn/2008k.html#2 Republican accomplishments and Hoover http://www.garlic.com/~lynn/2008k.html#50 update on old (GM) competitiveness thread http://www.garlic.com/~lynn/2008k.html#58 Mulally motors on at Ford http://www.garlic.com/~lynn/2008m.html#21 Fraud due to stupid failure to test for negative http://www.garlic.com/~lynn/2008m.html#52 Are family businesses unfair competition? http://www.garlic.com/~lynn/2008n.html#4 Michigan industry http://www.garlic.com/~lynn/2008p.html#77 Tell me why the taxpayer should be saving GM and Chrysler (and Ford) managers shareholders at this stage of the game? http://www.garlic.com/~lynn/2008q.html#22 Is Pride going to decimate the auto Industry? http://www.garlic.com/~lynn/2009i.html#2 China-US Insights on the Future of the Auto Industry http://www.garlic.com/~lynn/2009i.html#3 IBM interprets Lean development's Kaizen with new MCIF product http://www.garlic.com/~lynn/2009i.html#10 64 Cores -- IBM is showing a prototype already http://www.garlic.com/~lynn/2009i.html#31 Why are z/OS people reluctant to use z/OS UNIX? http://www.garlic.com/~lynn/2010b.html#14 360 programs on a z/10 -- 42yrs virtualization experience (since Jan68), online at home since Mar1970 -- 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: z9 / z10 instruction speed(s)
Edward Jaffe pisze: [...] People with PC-only experience are always astonished when I tell them about modern mainframe provisioning capabilities. They always assume when your hard drive fills up you need a new one or when your CPU is too slow you need a new one. What we do seems like magic to them. Yes, mainframe capabilities are excellent in this area. From the other hand they solve problems which exist only in mainframe world: CPU power adjustment. CPU shortage is bad thing on any platform, but mainframe is the only one where too much MIPS is not good. Why to downgrade a PC? The same apply to specialty processors. -- 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 01.01.2009 r. kapita zakadowy BRE Banku SA (w caoci wpacony) wynosi 118.763.528 zotych. W zwizku z realizacj warunkowego podwyszenia kapitau zakadowego, na podstawie uchway XXI WZ z dnia 16 marca 2008r., oraz uchway XVI NWZ z dnia 27 padziernika 2008r., moe ulec podwyszeniu do kwoty 123.763.528 z. Akcje w podwyszonym kapitale zakadowym BRE Banku SA bd w caoci opacone. -- 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: z9 / z10 instruction speed(s)
Leaving aside the important point that System z is certainly not the only CPU to offer the option of kneecapped CPUs -- Intel Celeron, anyone? -- imagine (briefly) a world in which CPU kneecapping was not available. Now, what would be the smallest capacity System z available? Answer: a fully configured System z9 BC (until end of June). I don't like MIPS metrics, but for those that still do it'd be thousands of MIPS. And let's further assume that all IBM software is available with sub-capacity (kneecapped) licensing (i.e. VWLC). (That's an oversimplification but effectively true.) You could license 3 MSUs of IBM software on your ginormous machine. However (and here's the big catch), an awful lot of non-IBM software isn't sub-capacity licensed. So what would be your software license charge for a whopper machine? Now IBM introduces the System z10 BC, and you upgrade. You get thousands more MIPS if there's no kneecapping available. Leaving aside many other important issues, what happens to your software bill? Ooops. Note that I don't speak for IBM, but let me come right out and say this: kneecapping is one of the best things IBM has ever done to help the smaller mainframe customer manage their costs while enjoying the same benefits in staying current with technology the way that the big shops do. Without kneecapping there probably wouldn't be very many smaller mainframe customers, either current ones or new ones. In fact, I suspect that System z is the only CPU offering kneecapping where there's such a clear benefit to the customer. (Not only to the vendor.) Intel's 486SX, for example, didn't seem to offer the customer any benefits. (Was AutoCAD less pricey on a 486SX versus a 486DX? No, it was just slower.) - - - - - Timothy Sipples IBM Consulting Enterprise Architect for New, Advanced, and/or Innovative Solutions (VCT) Based in Singapore Serving the Growth Markets E-Mail: timothy.sipp...@us.ibm.com -- 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: z9 / z10 instruction speed(s)
Now there you go being logical again. Charles -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Timothy Sipples Sent: Saturday, March 06, 2010 4:26 PM To: IBM-MAIN@bama.ua.edu Subject: Re: z9 / z10 instruction speed(s) Leaving aside the important point that System z is certainly not the only CPU to offer the option of kneecapped CPUs -- Intel Celeron, anyone? -- imagine (briefly) a world in which CPU kneecapping was not available. Now, what would be the smallest capacity System z available? Answer: a fully configured System z9 BC (until end of June). I don't like MIPS metrics, but for those that still do it'd be thousands of MIPS. And let's further assume that all IBM software is available with sub-capacity (kneecapped) licensing (i.e. VWLC). (That's an oversimplification but effectively true.) You could license 3 MSUs of IBM software on your ginormous machine. However (and here's the big catch), an awful lot of non-IBM software isn't sub-capacity licensed. So what would be your software license charge for a whopper machine? Now IBM introduces the System z10 BC, and you upgrade. You get thousands more MIPS if there's no kneecapping available. Leaving aside many other important issues, what happens to your software bill? Ooops. Note that I don't speak for IBM, but let me come right out and say this: kneecapping is one of the best things IBM has ever done to help the smaller mainframe customer manage their costs while enjoying the same benefits in staying current with technology the way that the big shops do. Without kneecapping there probably wouldn't be very many smaller mainframe customers, either current ones or new ones. In fact, I suspect that System z is the only CPU offering kneecapping where there's such a clear benefit to the customer. (Not only to the vendor.) Intel's 486SX, for example, didn't seem to offer the customer any benefits. (Was AutoCAD less pricey on a 486SX versus a 486DX? No, it was just slower.) - - - - - Timothy Sipples IBM Consulting Enterprise Architect for New, Advanced, and/or Innovative Solutions (VCT) Based in Singapore Serving the Growth Markets E-Mail: timothy.sipp...@us.ibm.com -- 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 -- 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: z9 / z10 instruction speed(s)
R.S. wrote: Edward Jaffe pisze: [...] People with PC-only experience are always astonished when I tell them about modern mainframe provisioning capabilities. They always assume when your hard drive fills up you need a new one or when your CPU is too slow you need a new one. What we do seems like magic to them. Yes, mainframe capabilities are excellent in this area. From the other hand they solve problems which exist only in mainframe world: CPU power adjustment. CPU shortage is bad thing on any platform, but mainframe is the only one where too much MIPS is not good. Why to downgrade a PC? The same apply to specialty processors. That is not completely true. Some amount of Unix and even Windows enterprise software (and not from IBM) is licensed by the capacity on the machine. -- Radoslaw Skorupka Lodz, Poland -- BRE Bank SA ul. Senatorska 18 00-950 Warszawa www.brebank.pl Regards, Henry -- 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: z9 / z10 instruction speed(s)
Henry Willard pisze: R.S. wrote: Edward Jaffe pisze: [...] People with PC-only experience are always astonished when I tell them about modern mainframe provisioning capabilities. They always assume when your hard drive fills up you need a new one or when your CPU is too slow you need a new one. What we do seems like magic to them. Yes, mainframe capabilities are excellent in this area. From the other hand they solve problems which exist only in mainframe world: CPU power adjustment. CPU shortage is bad thing on any platform, but mainframe is the only one where too much MIPS is not good. Why to downgrade a PC? The same apply to specialty processors. That is not completely true. Some amount of Unix and even Windows enterprise software (and not from IBM) is licensed by the capacity on the machine. Yes, it is PART of the picture. While usually non-mainframe software does NOT depend on CPU speed (don't confuse with NUMBER of CPUs), there is other side of the coin. The other side is you pay for what you use and you get MORE (not enabled) - just for future upgrades. It is convenient for customer and for IBM. Other HW vendors don't do it for various reasons. -- 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 01.01.2009 r. kapita zakadowy BRE Banku SA (w caoci wpacony) wynosi 118.763.528 zotych. W zwizku z realizacj warunkowego podwyszenia kapitau zakadowego, na podstawie uchway XXI WZ z dnia 16 marca 2008r., oraz uchway XVI NWZ z dnia 27 padziernika 2008r., moe ulec podwyszeniu do kwoty 123.763.528 z. Akcje w podwyszonym kapitale zakadowym BRE Banku SA bd w caoci opacone. -- 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: z9 / z10 instruction speed(s)
-Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Timothy Sipples Sent: Saturday, March 06, 2010 7:26 PM To: IBM-MAIN@bama.ua.edu Subject: Re: z9 / z10 instruction speed(s) Snipped Note that I don't speak for IBM, but let me come right out and say this: kneecapping is one of the best things IBM has ever done to help the smaller mainframe customer manage their costs while enjoying the same benefits in staying current with technology the way that the big shops do. Without kneecapping there probably wouldn't be very many smaller mainframe customers, either current ones or new ones. But Tim, there *aren't* any smaller mainframe customers any more (at least not in the USA). IMHO, IBM abandoned the truly small customer when it abandoned VM/VSE for this once-upon-a-time viable customer base. z/OS is far too expensive (even with kneecapping) for (let's say) a small manufacturer (a regional baker for example). AFAIK, there is no way for IBM to bring z/OS on a z10 to a customer with a yearly IT budget in the 5- to (very) low 6-figure range, including software and staffing costs and necessary 3rd-party software. This is the once-upon-a-time mainframe customer base which IBM has deliberately decided to ignore, despite the fact that many such small operations can (and do) grow into larger ones. IBM has made and continues to make clear that its promise to investors to be present only in high-margin businesses is not just talk. If it's not at least a 7-figure contract for a mainframe, IBM just isn't that interested. Just my USD$0.02, FWIW. Peter This message and any attachments are intended only for the use of the addressee and may contain information that is privileged and confidential. If the reader of the message is not the intended recipient or an authorized representative of the intended recipient, you are hereby notified that any dissemination of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by e-mail and delete the message and any attachments from your system. -- 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: z9 / z10 instruction speed(s)
This has been a very interesting thread for me. If I remember correctly from the time I saw the z/10 with plexiglass outsides and a hardware guy there to explain what was what, and one of the things he told me was the cpu speed was (IIRC) 4.77GHz. My laptop has a dual core 1.99GHz. You already see where I'm going with this. How does a z/10 get so much more done? Forgive me for my math and analytical skills, but seems like 4 laptops could equal the speed of a 1 cpu z/10. There is a huge chunk of this equation that I'm totally missing. How does a z/10 get so much more done? kind regards, Lindy -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of R.S. Sent: 7. maaliskuuta 2010 2:23 To: IBM-MAIN@bama.ua.edu Subject: Re: z9 / z10 instruction speed(s) Edward Jaffe pisze: [...] People with PC-only experience are always astonished when I tell them about modern mainframe provisioning capabilities. They always assume when your hard drive fills up you need a new one or when your CPU is too slow you need a new one. What we do seems like magic to them. Yes, mainframe capabilities are excellent in this area. From the other hand they solve problems which exist only in mainframe world: CPU power adjustment. CPU shortage is bad thing on any platform, but mainframe is the only one where too much MIPS is not good. Why to downgrade a PC? The same apply to specialty processors. -- Radoslaw Skorupka Lodz, Poland -- 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: z9 / z10 instruction speed(s)
I don't like MIPS metrics A rose is a rose is a rose. MSUs are just MIPS multiplied by a (marketting) constant or 5. - Too busy driving to stop for gas! -- 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: z9 / z10 instruction speed(s)
On Sat, Mar 6, 2010 at 8:43 PM, Lindy Mayfield lindy.mayfi...@ssf.sas.com wrote: This has been a very interesting thread for me. If I remember correctly from the time I saw the z/10 with plexiglass outsides and a hardware guy there to explain what was what, and one of the things he told me was the cpu speed was (IIRC) 4.77GHz. My laptop has a dual core 1.99GHz. You already see where I'm going with this. How does a z/10 get so much more done? Forgive me for my math and analytical skills, but seems like 4 laptops could equal the speed of a 1 cpu z/10. There is a huge chunk of this equation that I'm totally missing. How does a z/10 get so much more done? The usual metaphor is that four Tauruses do not a semi make, even though the acreage and gas mileage are comparable (no, I'm not prepared to claim that 4 or Tauruses are necessarily the right substitutions, but you get my point). (Note that the converse [inverse? perverse?] is also true: most people wouldn't *want* to use a z10 for all daily computing use -- Web browsing, photo editing, etc.) z10s (not z/10s -- hardware not software, no slash) get more* done because of: - hardware design: I/O is offloaded better, multitasking is more efficient, etc. - software design: those of us who grew up counting bits and storing flags in the top byte of a 24-bit address to save memory are pretty far from the GUI kids whose hello world application occupies 200KB+, and so are our applications I could go on, but the bottom line is, different design, different targets, different results. If you look at carefully written PC software like, say, Steve Gibson's stuff (www.grc.com -- not a plug, just an example that comes to mind), you'll see incredibly rich and powerful stuff that fits in the palm of your PC's hand, so to speak. http://www.grc.com/freepopular.htm has dozens of apps, most in the 25K (yes, K!) range. So it isn't impossible to write tight software for PCs, just discouraged by the apparent lack of need and ubiquity of IDEs that produce bloat. ...phsiii *FSVO more: often just different. As noted above, playing MP3s or editing photos on a z would be suboptimal at best. -- 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: z9 / z10 instruction speed(s)
li...@akphs.com (Phil Smith III) writes: If you look at carefully written PC software like, say, Steve Gibson's stuff (www.grc.com -- not a plug, just an example that comes to mind), you'll see incredibly rich and powerful stuff that fits in the palm of your PC's hand, so to speak. http://www.grc.com/freepopular.htm has dozens of apps, most in the 25K (yes, K!) range. So it isn't impossible to write tight software for PCs, just discouraged by the apparent lack of need and ubiquity of IDEs that produce bloat. recent thread about redoing part of res system ... sized that something like ten rs/6000 580s could have handled the full world-wide activity way outperforming large datacenter of TPF systems (some load projected to take a couple hundred es9000 processors) and current treo (xscale processor) theoritically has approx. compute power of those ten 580s. http://www.garlic.com/~lynn/2010b.html#80 Happy DEC-10 Day http://www.garlic.com/~lynn/2010c.html#19 Processes' memory -- 42yrs virtualization experience (since Jan68), online at home since Mar1970 -- 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: z9 / z10 instruction speed(s)
This is simply incredible, to think that IBM would deliberately run BCT loops to throttle, slowdown, CPs. It is one thing to cut back the CPU cache. It is quite another to deliberate slow things down. When will they ever learn that quality sells more than anything else and that making anything less than the best possible product just hurts themselves. Detroit is in shambles because of built-in obsolescence. My 1997 Honda Civic with the original radiator, air conditioner, heater coil, starter, fuel pump, gas tank, etc has 374,000 miles on it and about the only thing I have ever had to replace, aside from regular wear and tear like tires, muffler, brakes, battery etc, is the ignition coil. Sakichi Toyoda, (1867-1930) the greatest inventor and industrialist of Japan, built his company in the early 1900's which later spawned the current Toyota auto company, currently the world's largest automobile company, despite recent problems. BTW: the name Toyoda was changed to Toyota for PR reasons; the former has bad connotations in Japanese. Sakichi Toyoda, built is company and success on what is know as the 5 Whys. When doing PD he required his people to ask a minimum of 5 Whys to get at the root cause of the problem: My car will not start. (the problem) 1. *Why?* - The battery is dead. (first why) 2. *Why?* - The alternator is not functioning. (second why) 3. *Why?* - The alternator belt has broken. (third why) 4. *Why?* - The alternator belt was well beyond its useful service life and has never been replaced. (fourth why) 5. *Why?* - I have not been maintaining my car according to the recommended service schedule. (fifth why, a root cause) The current trend towards CMMI and the Six Sigma standard of quality, 6 standard deviations (3.4 defects in a million) instead of the typical 3 standards deviations (1 defect in 370) points to the demand for quality, excellence, and perfection in everything. No one will settle for less. It is only a question of time before IBM goes the way of Detroit, if they do not wake up. On Wed, Mar 3, 2010 at 4:03 AM, Edward Jaffe edja...@phoenixsoftware.comwrote: McKown, John wrote: There are multiple z9 models. Each model has its own MSU rating, which is basically related to the number of CPs enabled and their speed. Now, I know that all the CPs on all z9 run same hardware speed. So, I'm wondering how they are knee capped? Now, I know that the knee capping is done by loading in a specific MCL. So, I'm thinking that this somehow does something like inserts a wait state during instruction processing. That is, the XYZ instruction on all z9s run in the same amount of time. But there is something extra done at the end of the XYZ instruction which causes a wait before the next instruction is actually executed. Am I on the right track? Or is it done is some other strange manner? There is a hardware timer pop (think STIMER REAL) that occurs every 'n' milliseconds on every CP that passes control to a millicode routine that does important housekeeping tasks for the CP, such as noticing and responding to SIGP requests. Before exiting this routine, they load a model-dependent integer value into a millicode register (Rx) and execute BCT Rx,* which chews up the prescribed amount of processor cycles. -- Edward E Jaffe Phoenix Software International, Inc 831 Parkview Drive North El Segundo, CA 90245 310-338-0400 x318 edja...@phoenixsoftware.com http://www.phoenixsoftware.com/ -- 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 -- George Henke (C) 845 401 5614 -- 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: z9 / z10 instruction speed(s)
On Fri, Mar 5, 2010 at 11:02 AM, George Henke gahe...@gmail.com wrote: This is simply incredible, to think that IBM would deliberately run BCT loops to throttle, slowdown, CPs. It is one thing to cut back the CPU cache. It is quite another to deliberate slow things down. I thought this debate ended with the 486SX chips. Using your logic, they should offer CPUs in groups of 4, since they're shipping them anyway. Using short words: YOU PAY LESS, YOU GET LESS. What's tricky about that? -- 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: z9 / z10 instruction speed(s)
The following message is a courtesy copy of an article that has been posted to bit.listserv.ibm-main,alt.folklore.computers as well. gahe...@gmail.com (George Henke) writes: The current trend towards CMMI and the Six Sigma standard of quality, 6 standard deviations (3.4 defects in a million) instead of the typical 3 standards deviations (1 defect in 370) points to the demand for quality, excellence, and perfection in everything. Toyota doing Lean manufacturing http://en.wikipedia.org/wiki/Lean_manufacturing some of this has cross-over with Boyd's OODA-loops ... I had sponsored Boyd's briefings at IBM http://en.wikipedia.org/wiki/OODA_loop Toyota Production System http://en.wikipedia.org/wiki/Toyota_Production_System In the early 90s, one of the big-3 had C4 task force that was looking at improving their competitive position ... and invented in some number of technology vendors to participate. They went thru majority of the issues with respect to their current state and foreign competitors. One of the big issues was major foreign competitor had reduced elapsed cycle to produce a (totally new) product (from idea to rolling off the line) from 7-8 years to 2-3 years (and looking at dropping below 2 years). Big part of C4 was leveraging technology as part of drastically reducing elapsed product cycle. I chided some of the mainframe brethern attending the meetings about being there to offer advice on reducing product cycle from 7-8yrs to 2-3yrs (when they were still on long product cycle). Within the domesitic auto industry ... although they could very clearly articulate all the important issues ... the status quo was so entrenched that they found it difficult to change. recent thread in greater ibm blog mentioning Boyd http://www.garlic.com/~lynn/2010e.html#39 Agile Workforce http://www.garlic.com/~lynn/20103.html#43 Boyd's Briefings misc. past posts mentioning Boyd /or OODA-loops http://www.garlic.com/~lynn/subboyd.html -- 42yrs virtualization experience (since Jan68), online at home since Mar1970 -- 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: z9 / z10 instruction speed(s)
This is simply incredible, to think that IBM would deliberately run BCT loops to throttle, slowdown, CPs. Why? This kind of thing has been around nearly as long as commercial computing. I remember when upgrades consisted of removing circuitry that slowed the processor down. The AMD470 had built-in decelerators. It is one thing to cut back the CPU cache. It is quite another to deliberate slow things down. Why? If you don't need the capacity, what's the issue? Would you rather pay full hardware software costs for capacity you don't need. Also, this way, IBM has to build just one processor chip. - Too busy driving to stop for gas! -- 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: z9 / z10 instruction speed(s)
On Fri, 5 Mar 2010 11:18:08 -0500, zMan zedgarhoo...@gmail.com wrote: On Fri, Mar 5, 2010 at 11:02 AM, George Henke gahe...@gmail.com wrote: This is simply incredible, to think that IBM would deliberately run BCT loops to throttle, slowdown, CPs. It is one thing to cut back the CPU cache. It is quite another to deliberate slow things down. I thought this debate ended with the 486SX chips. Using your logic, they should offer CPUs in groups of 4, since they're shipping them anyway. Using short words: YOU PAY LESS, YOU GET LESS. What's tricky about that? Exactly. Didn't we have this discussion a couple of weeks ago? Mark -- Mark Zelden - Zelden Consulting Services - z/OS, OS/390 and MVS mailto:mzel...@flash.net Mark's MVS Utilities: http://home.flash.net/~mzelden/mvsutil.html Systems Programming expert at http://expertanswercenter.techtarget.com/ -- 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: z9 / z10 instruction speed(s)
On Fri, 5 Mar 2010 11:02:22 -0500, George Henke wrote: This is simply incredible, to think that IBM would deliberately run BCT loops to throttle, slowdown, CPs. Check the archives for discussions of kneecapped processors. This has been covered many times. It is one thing to cut back the CPU cache. It is quite another to deliberate slow things down. That's absurd. Reducing CPU cache will certainly cause a system to slow down, but not in a very predictable way. And why is it more reasonable to use only part of the available cache than it is to run the CPU at a slower speed? -- Tom Marchant -- 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: z9 / z10 instruction speed(s)
The following message is a courtesy copy of an article that has been posted to bit.listserv.ibm-main,alt.folklore.computers as well. eamacn...@yahoo.ca (Ted MacNEIL) writes: Why? If you don't need the capacity, what's the issue? Would you rather pay full hardware software costs for capacity you don't need. Also, this way, IBM has to build just one processor chip. i.e. aggregate computing cost (for everybody) is actually less with single part number ... than if there were large number of different parts. in early 80s, major analysis of vm/4341s going into every nook cranny versus big iron in the datacenter, was the enormously greater big iron expense involved in adding capacity. this can somewhat also be seen with returning to the old timesharing days with cloud computing. having extra capacity already available at the customer site ... is analogous to having on-site spare part depot /or on-site CE. recent reference to 3033N ... slower than 168/3032 ... but able to be field upgraded to full speed 3033: http://www.garlic.com/~lynn/2010e.html#27 SHAREWARE at Its Finest -- 42yrs virtualization experience (since Jan68), online at home since Mar1970 -- 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: z9 / z10 instruction speed(s)
George Henke wrote: This is simply incredible, to think that IBM would deliberately run BCT loops to throttle, slowdown, CPs. It is one thing to cut back the CPU cache. It is quite another to deliberate slow things down. IBM's current knee-capping approach is far superior to the old adjusted-clocking approach. It has given rise to the most granular sizing and flexible upgrade paths of any hardware platform on the planet. Today's mainframe dynamic provisioning capabilities are truly leading-edge, and improving with each new generation. We can dynamically grow any DASD volume--on the fly--up to 226GB in size. We can download and dynamically apply a patch that makes our CPUs run faster, adds new CPUs, or both. Expect to see even more such capabilities in the future... People with PC-only experience are always astonished when I tell them about modern mainframe provisioning capabilities. They always assume when your hard drive fills up you need a new one or when your CPU is too slow you need a new one. What we do seems like magic to them. Any sufficiently advanced technology is indistinguishable from magic. -- Arthur C. Clarke, Profiles of The Future, 1961 -- Edward E Jaffe Phoenix Software International, Inc 831 Parkview Drive North El Segundo, CA 90245 310-338-0400 x318 edja...@phoenixsoftware.com http://www.phoenixsoftware.com/ -- 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: z9 / z10 instruction speed(s)
On Fri, Mar 5, 2010 at 2:01 PM, Edward Jaffe edja...@phoenixsoftware.comwrote: Today's mainframe dynamic provisioning capabilities are truly leading-edge, and improving with each new generation. We can dynamically grow any DASD volume--on the fly--up to 226GB in size. We can download and dynamically apply a patch that makes our CPUs run faster, adds new CPUs, or both. Expect to see even more such capabilities in the future... People with PC-only experience are always astonished when I tell them about modern mainframe provisioning capabilities. They always assume when your hard drive fills up you need a new one or when your CPU is too slow you need a new one. What we do seems like magic to them. You know *I* don't disagree with your position here, but there is a disconnect -- distributed folks don't understand issues like small volumes (226GB being smaller than the hard drive in my laptop), much less JCL. So many/most of them *still* see the mainframe as slow, old, difficult, crippled. That's something that needs to be rectified, but I'm ed if I know how to do it. -- 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: z9 / z10 instruction speed(s)
But, can you connect 10,000 dasd up to your PC? Now, as for getting them to understand where a PC or a mainframe makes sense, that is harder. PC people think in terms of 10's of transactions. Mainframer's think in terms of 10's of thousands of transactions. I know of a company that has a table that is coming up on crossing the 2 trillion rows in it mark. I know another company that does over 100 million inserts a day. As far as hobbling a CPU, if you want it to run flat out then pay for it. I'm sure IBM would rather that all boxes run full speed with all the processors turned on. $ Chris Blaicher Phone: 512-340-6154 Mobile: 512-627-3803 -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of zMan Sent: Friday, March 05, 2010 1:46 PM To: IBM-MAIN@bama.ua.edu Subject: Re: z9 / z10 instruction speed(s) distributed folks don't understand issues like small volumes (226GB being smaller than the hard drive in my laptop) -- 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: z9 / z10 instruction speed(s)
zMan wrote: ... distributed folks don't understand issues like small volumes (226GB being smaller than the hard drive in my laptop)... It has already been said by IBM, but obviously bears repeating... The 226GB per volume EAV limit is nowhere near the *architectural* limit of EAV--which is over 220TB per volume. IBM artificially limited the initial per volume EAV size to 226GB to ensure any production performance bottlenecks can be addressed before raising the limit to higher values. IBM is very aware of the airline magazine mentality that prevails today. They realize it's difficult to be on the defensive all the time trying to explain capacity and throughput issues to ignorant people. It's better to be the ones' putting the others on the defensive. That's why we have 64-bit processors (and not 63-bit) and a 4.4Ghz chip in the z10 (and not the 2+Ghz value we should have had if we had continued down the z9 path). I would not be surprised to see a 1TB or even 2TB EAV in the not-too-distant future. My prediction is that IBM will continually raise the per volume EAV size to a value at or above the prevailing off-platform volume size for exactly the reasons stated... -- Edward E Jaffe Phoenix Software International, Inc 831 Parkview Drive North El Segundo, CA 90245 310-338-0400 x318 edja...@phoenixsoftware.com http://www.phoenixsoftware.com/ -- 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: z9 / z10 instruction speed(s)
Until you tell them the (upfront) cost to get into a mainframe - or the cost to turn on those CPs. Then the look of awe turns to derisive laughter. And growing DASD is only a big deal to us because of our history - ask gil about ZFS; ask a Linux admin about LVM (or even EVMS). Who cares about the underlying hardware - it's the filesystem (your data) that matters. Whack in some new disk, grow your data (dynamically) across it. Even the mainframe vendors know this - they keep swapping out the drives for bigger ones. I say we still have a fight on our hands. Shane ... On Sat, Mar 6th, 2010 at 6:01 AM, Edward Jaffe wrote: People with PC-only experience are always astonished when I tell them about modern mainframe provisioning capabilities. They always assume when your hard drive fills up you need a new one or when your CPU is too slow you need a new one. What we do seems like magic to them. Any sufficiently advanced technology is indistinguishable from magic. -- Arthur C. Clarke, Profiles of The Future, 1961 -- 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: z9 / z10 instruction speed(s)
McKown, John wrote: There are multiple z9 models. Each model has its own MSU rating, which is basically related to the number of CPs enabled and their speed. Now, I know that all the CPs on all z9 run same hardware speed. So, I'm wondering how they are knee capped? Now, I know that the knee capping is done by loading in a specific MCL. So, I'm thinking that this somehow does something like inserts a wait state during instruction processing. That is, the XYZ instruction on all z9s run in the same amount of time. But there is something extra done at the end of the XYZ instruction which causes a wait before the next instruction is actually executed. Am I on the right track? Or is it done is some other strange manner? There is a hardware timer pop (think STIMER REAL) that occurs every 'n' milliseconds on every CP that passes control to a millicode routine that does important housekeeping tasks for the CP, such as noticing and responding to SIGP requests. Before exiting this routine, they load a model-dependent integer value into a millicode register (Rx) and execute BCT Rx,* which chews up the prescribed amount of processor cycles. -- Edward E Jaffe Phoenix Software International, Inc 831 Parkview Drive North El Segundo, CA 90245 310-338-0400 x318 edja...@phoenixsoftware.com http://www.phoenixsoftware.com/ -- 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: z9 / z10 instruction speed(s)
I was told (years ago,different hardware) that the pipeline was filled with the appropriate number of NOPs. Might have just been an engineer finding a convenient explanation for a sysprog though ... I always wondered how that worked across different workloads - with all the smarts built into the hardware to optimize overlapped pipeline stages and branch prediction. No doubt a better answer will arise. Shane ... On Wed, Mar 3rd, 2010 at 8:59 AM, McKown, John john.mck...@healthmarkets.com wrote: There are multiple z9 models. Each model has its own MSU rating, which is basically related to the number of CPs enabled and their speed. Now, I know that all the CPs on all z9 run same hardware speed. So, I'm wondering how they are knee capped? Now, I know that the knee capping is done by loading in a specific MCL. So, I'm thinking that this somehow does something like inserts a wait state during instruction processing. That is, the XYZ instruction on all z9s run in the same amount of time. But there is something extra done at the end of the XYZ instruction which causes a wait before the next instruction is actually executed. Am I on the right track? Or is it done is some other strange manner? -- 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: z9 / z10 instruction speed(s)
-Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of McKown, John Sent: Tuesday, March 02, 2010 3:59 PM To: IBM-MAIN@bama.ua.edu Subject: z9 / z10 instruction speed(s) There are multiple z9 models. Each model has its own MSU rating, which is= basically related to the number of CPs enabled and their speed. Now, I k= now that all the CPs on all z9 run same hardware speed. So, I'm wondering h= ow they are knee capped? Now, I know that the knee capping is done by l= oading in a specific MCL. So, I'm thinking that this somehow does something= like inserts a wait state during instruction processing. That is, the XY= Z instruction on all z9s run in the same amount of time. But there is some= thing extra done at the end of the XYZ instruction which causes a wait b= efore the next instruction is actually executed. Am I on the right track? O= r is it done is some other strange manner? SNIP From somewhere in the hazy past, it had something to do with instruction fetch. So while instruction fetch was being held up for n microcode cycles, the pipe was being filled, effectively, with NOPR instructions. Regards, Steve Thompson -- Opinions expressed by this poster may not reflect those of poster's employer -- -- 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: z9 / z10 instruction speed(s)
On 2 March 2010 16:59, McKown, John john.mck...@healthmarkets.com wrote: There are multiple z9 models. Each model has its own MSU rating, which is basically related to the number of CPs enabled and their speed. Now, I know that all the CPs on all z9 run same hardware speed. So, I'm wondering how they are knee capped? Now, I know that the knee capping is done by loading in a specific MCL. So, I'm thinking that this somehow does something like inserts a wait state during instruction processing. That is, the XYZ instruction on all z9s run in the same amount of time. But there is something extra done at the end of the XYZ instruction which causes a wait before the next instruction is actually executed. Am I on the right track? Or is it done is some other strange manner? Could be a simple as disabling (or just not using, or flushing) a certain amount of cache. It would be instructive to see how fast a program that stays in one cache line runs on a less-than-full-speed processor. But no doubt as soon as someone figures out how to exploit that, they'll put a stop to it. Tony H. -- 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