zSeries IFL speed rating
Is there a site that gives a Mhz rating for the z900 and z9 IFLs? Dennis Roach United Space Alliance 600 Gemini Avenue Mail Code USH-4A3L Houston, Texas 77058 Voice: (281) 282-2975 Page:(713) 736-8275 Fax: (281) 282-3583 E-Mail: [EMAIL PROTECTED] All opinions expressed by me are mine and may not agree with my employer or any person, company, or thing, living or dead, on or near this or any other planet, moon, asteroid, or other spatial object, natural or manufactured, since the beginning of time. -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
> -Original Message- > From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On > Behalf Of Roach, Dennis > Sent: Thursday, July 19, 2007 10:43 AM > To: LINUX-390@VM.MARIST.EDU > Subject: zSeries IFL speed rating > > > Is there a site that gives a Mhz rating for the z900 and z9 IFLs? > > > > Dennis Roach I don't know why you want this, but MHz can be misleading. In fact, all engines on a particular zBox (z800, z890, z900, z990, z9BC, and z9EC) run with the same cycle time. The performance differences within a given box are due to delays in the microcode loaded. IFLs (and zAAPs and zIIPS) all run at "full speed". Eg. all z800 engines have the same cycle speed. It is just that some cycles are "wasted" by the microcode in order to decrease performance. aka "knee capping". -- John McKown Senior Systems Programmer HealthMarkets Keeping the Promise of Affordable Coverage Administrative Services Group Information Technology The information contained in this e-mail message may be privileged and/or confidential. It is for intended addressee(s) only. If you are not the intended recipient, you are hereby notified that any disclosure, reproduction, distribution or other use of this communication is strictly prohibited and could, in certain circumstances, be a criminal offense. If you have received this e-mail in error, please notify the sender by reply and delete this message without copying or disclosing it. -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
On 7/19/07, Roach, Dennis <[EMAIL PROTECTED]> wrote: Is there a site that gives a Mhz rating for the z900 and z9 IFLs? Start here I'd say. There's lots of pointers to various tables. What they talk about is MIPS rate. You can probably also find cycle times published for different generations, but even less useful. Do understand that for some models the real engines have been knee-capped, but the IFL is not. Rob -- Rob van der Heij Velocity Software, Inc http://velocitysoftware.com/ -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
On 7/19/07, Rob van der Heij <[EMAIL PROTECTED]> wrote: Start here I'd say. There's lots of pointers to various tables. http://www.isham-research.co.uk/mips_z9.html -- Rob van der Heij Velocity Software, Inc http://velocitysoftware.com/ -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
On Thursday, 07/19/2007 at 11:45 EDT, "Roach, Dennis" <[EMAIL PROTECTED]> wrote: > Is there a site that gives a Mhz rating for the z900 and z9 IFLs? Since that number would be meaningless, no. I mean, you might find one floating about out there as a factoid, but it has zero value. Why do you care? (Reference recent discussion in IBM-MAIN.) Alan Altmark z/VM Development IBM Endicott -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
Quoting Alan Altmark <[EMAIL PROTECTED]>: > > SLES 10 isn't mentioned. It is a Really Bad Idea to try to install IBM > middleware on anything other than the supported platforms. (a) It may not > install, and (2) it isn't supported by the Support Center. The best bet > is to call it in. Don't worry, they'll tell you if you've wandered off > the farm. :-) > I have a spreadsheet entitled "Tivoli Non-Confidential Platform & Database Support Matrix" that indicates SLES10 on zSeries is supported in 31-bit mode. For sles10 that means you have to install and use the "s390" package to coerce the Tivoli installer into thinking that it is running in a 31-bit environment. Once installed you must also run the TEPS using the s390 command. This also works for a 64-bit sles9 install. The matrix can be had from: http://www-1.ibm.com/support/docview.wss?uid=swg21067036 Leland -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
I have been asked to determine the improvement of an IFL on a z9 BC over a z900. Mhz is something the people being presented to understand. It does not compare to the same speed on say INTEL. Dennis Roach United Space Alliance 600 Gemini Avenue Mail Code USH-4A3L Houston, Texas 77058 Voice: (281) 282-2975 Page:(713) 736-8275 Fax: (281) 282-3583 E-Mail: [EMAIL PROTECTED] All opinions expressed by me are mine and may not agree with my employer or any person, company, or thing, living or dead, on or near this or any other planet, moon, asteroid, or other spatial object, natural or manufactured, since the beginning of time. -Original Message- From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of Alan Altmark Sent: Thursday, July 19, 2007 11:01 AM To: LINUX-390@VM.MARIST.EDU Subject:Re: zSeries IFL speed rating On Thursday, 07/19/2007 at 11:45 EDT, "Roach, Dennis" <[EMAIL PROTECTED]> wrote: > Is there a site that gives a Mhz rating for the z900 and z9 IFLs? Since that number would be meaningless, no. I mean, you might find one floating about out there as a factoid, but it has zero value. Why do you care? (Reference recent discussion in IBM-MAIN.) Alan Altmark z/VM Development IBM Endicott -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
> -Original Message- > From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On > Behalf Of Roach, Dennis > Sent: Thursday, July 19, 2007 11:59 AM > To: LINUX-390@VM.MARIST.EDU > Subject: Re: zSeries IFL speed rating > > > I have been asked to determine the improvement of an IFL on a > z9 BC over a z900. > Mhz is something the people being presented to understand. > It does not compare to the same speed on say INTEL. > > Dennis Roach I'd just use MSUs. Eg. a 40MSU processor is 2x a 20MSU processor. They should be able to understand that, if all they want is a relative comparison. Of course, your-mileage-may-vary because you are not only getting a faster cycle time, you are getting a different microcode implementation. So some instructions may be faster while others could even be slower (not likely). That is, each instruction might vary in performance relative to the same instruction on the other machine. If they get cranky, ask them if they think it is correct to compare the MHz rating of a Celeron, Pentium, and Xeon processor. If they do, then they're dangerously ignorant. A Xeon CPU with a lower MHz is much more productive than a Celeron with a faster MHz. -- John McKown Senior Systems Programmer HealthMarkets Keeping the Promise of Affordable Coverage Administrative Services Group Information Technology The information contained in this e-mail message may be privileged and/or confidential. It is for intended addressee(s) only. If you are not the intended recipient, you are hereby notified that any disclosure, reproduction, distribution or other use of this communication is strictly prohibited and could, in certain circumstances, be a criminal offense. If you have received this e-mail in error, please notify the sender by reply and delete this message without copying or disclosing it. -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
Roach, Dennis wrote: I have been asked to determine the improvement of an IFL on a z9 BC over a z900. Mhz is something the people being presented to understand. It does not compare to the same speed on say INTEL. It's still bogus.. If you take Intel processors for example, even then, WITHIN their own brand, MHz is not a good indicator of measurement (Ex: a Core 2 @ 2.66 MHz will most probably outperform a Pentium IV @ 3.6 MHz) There has to be some other form of measurement and "hard numbers" that you may be able to use to grade improvement from z900 to z9 - but I'm sure MHz is not one of them ! --Ivan -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
Its been a long time, but I seem to remember that IBM had a computer model that the marketing people used. You could take performance data from your system and then plug in different changes, such as increasing real memory size, additional channels, additional processors, etc. and it would predict the performance improvement. I don't remember all the details. It may have been specific to MVS, since I think it used RMF data as a baseline. Paul Noble Cuyahoga County Information Service Center >>> "Roach, Dennis" <[EMAIL PROTECTED]> 7/19/2007 12:58 PM >>> I have been asked to determine the improvement of an IFL on a z9 BC over a z900. Mhz is something the people being presented to understand. It does not compare to the same speed on say INTEL. Dennis Roach United Space Alliance 600 Gemini Avenue Mail Code USH-4A3L Houston, Texas 77058 Voice: (281) 282-2975 Page:(713) 736-8275 Fax: (281) 282-3583 E-Mail: [EMAIL PROTECTED] All opinions expressed by me are mine and may not agree with my employer or any person, company, or thing, living or dead, on or near this or any other planet, moon, asteroid, or other spatial object, natural or manufactured, since the beginning of time. -Original Message- From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of Alan Altmark Sent: Thursday, July 19, 2007 11:01 AM To: LINUX-390@VM.MARIST.EDU Subject:Re: zSeries IFL speed rating On Thursday, 07/19/2007 at 11:45 EDT, "Roach, Dennis" <[EMAIL PROTECTED]> wrote: > Is there a site that gives a Mhz rating for the z900 and z9 IFLs? Since that number would be meaningless, no. I mean, you might find one floating about out there as a factoid, but it has zero value. Why do you care? (Reference recent discussion in IBM-MAIN.) Alan Altmark z/VM Development IBM Endicott -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
Unfortunately, it's not something that will translate in a meaninful way. Thats
why MIPS ratings are somewhat arbitrary as well. Meaningless
Inidicator of Processor Speed has been used to define the acronym.
What you CAN tell them is the speed difference between a z9 BC vs a z900 as an
expression of pecent improvement, or as a multiplier of improvement.
IBM has those* numbers availble and it is still management meaningful.
"Roach, Dennis" <[EMAIL PROTECTED]>
Sent by: Linux on 390 Port
To
LINUX-390@VM.MARIST.EDU
cc
07/19/2007 11:58 AM
Subject
Re:
zSeries IFL speed rating
Please respond to
Linux on 390 Port
I have been asked to determine the improvement of an IFL on a z9 BC over a z900.
Mhz is something the people being presented to understand.
It does not compare to the same speed on say INTEL.
Dennis Roach
United Space Alliance
600 Gemini Avenue
Mail Code USH-4A3L
Houston, Texas 77058
Voice: (281) 282-2975
Page:(713) 736-8275
Fax: (281) 282-3583
E-Mail: [EMAIL PROTECTED]
All opinions expressed by me are mine and may not agree with my employer or any
person, company, or thing, living or dead, on or near this or any
other
planet, moon, asteroid, or other spatial object, natural or manufactured, since
the beginning of time.
-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of
Alan Altmark
Sent:Thursday, July 19, 2007 11:01 AM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: zSeries IFL speed rating
On Thursday, 07/19/2007 at 11:45 EDT, "Roach, Dennis"
<[EMAIL PROTECTED]> wrote:
> Is there a site that gives a Mhz rating for the z900 and z9 IFLs?
Since that number would be meaningless, no. I mean, you might find one
floating about out there as a factoid, but it has zero value. Why do you
care? (Reference recent discussion in IBM-MAIN.)
Alan Altmark
z/VM Development
IBM Endicott
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
On 7/19/07, Roach, Dennis <[EMAIL PROTECTED]> wrote: I have been asked to determine the improvement of an IFL on a z9 BC over a z900. Mhz is something the people being presented to understand. It does not compare to the same speed on say INTEL. It does not help you a think that people understand cycle time. As if you could compare a Pentium-M and a Celeron with the same clock speed. With previous S/390 generations I had a program to determine cycle time. With modern CPU's the criterion turned out to be how many instructions you can put extra in the loop without slowing it down... Apart from MIPS or MSU numbers that you will find, be aware - MSU ratings have been tweaked to make software license of z9 more competitive - speed increase from z900 to z9 is non-linear, in particular instructions popular with Linux have been optimized beyond that - when the application is not compiled to exploit z9 architecture, some advantage disappears - many ratings are for single-image z/OS workload on nn CPU's, which may not be what you do Rob -- Rob van der Heij Velocity Software, Inc http://velocitysoftware.com/ -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
> Is there a site that gives a Mhz rating for the z900 and z9 > IFLs? Get the zPCR tool from IBM. This is the external version of the same tool the IBM reps use to compare performance of various mainframe systems. http://www-03.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/PRS1381 Jim -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
Generally the IFL's are not capped, and the number 450 MIPS seems to ring a bell. Doug At 11:56 AM 7/19/2007, you wrote: On 7/19/07, Roach, Dennis <[EMAIL PROTECTED]> wrote: Is there a site that gives a Mhz rating for the z900 and z9 IFLs? Start here I'd say. There's lots of pointers to various tables. What they talk about is MIPS rate. You can probably also find cycle times published for different generations, but even less useful. Do understand that for some models the real engines have been knee-capped, but the IFL is not. Rob -- Rob van der Heij Velocity Software, Inc http://velocitysoftware.com/ -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 Doug Fuerst Consultant BK Associates Brooklyn, NY (718) 921-2620 (Office) (718) 921-0952 (Fax) (917) 572-7364 (Cell) [EMAIL PROTECTED] -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
>IBM has those* numbers availble and it is still management meaningful. Huh? What does that statement even mean? IBM has numbers, but they're no longer meaningful! - Too busy driving to stop for gas! -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
Since Mhz and MIPS are such misused values, I prefer to run the same program on old and new engines to compare the performance change. I use an old FORTRAN (no flames please) program that computes pi to 5000 places. A boss once needed something to see if the vendor really did upgrade our processor, just after our Performance and Capacity Planning group was disbanded. So on an idle system, this just gets into memory and runs with a high Total/Virtual CPU ratio. On my z890 IFL, 20 iterations takes about 8.571 sec virtual, 8.578 sec total and 8.944 sec elapsed time. Find a favorite program that is repeatable and keep using that to test the speed of your engines. But all numbers must be taken as relative and with a big grain of salt. /Tom Kern /301-903-2211 -- Original Message --- From: "Roach, Dennis" <[EMAIL PROTECTED]> Subject: Re: zSeries IFL speed rating I have been asked to determine the improvement of an IFL on a z9 BC over a z900. Mhz is something the people being presented to understand. It does not compare to the same speed on say INTEL. Dennis Roach United Space Alliance 600 Gemini Avenue Mail Code USH-4A3L Houston, Texas 77058 Voice: (281) 282-2975 Page:(713) 736-8275 Fax: (281) 282-3583 E-Mail: [EMAIL PROTECTED] -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
>>> On Thu, Jul 19, 2007 at 6:01 PM, in message <[EMAIL PROTECTED]>, Doug Fuerst <[EMAIL PROTECTED]> wrote: > Generally the IFL's are not capped, and the number 450 MIPS seems to > ring a bell. IFLs, zIIPs, and zAAPs are _never_ capped. They always run at their full rated speed, even if the CPs in the box don't. Mark Post -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
> Generally the IFL's are not capped, and the number 450 MIPS seems to ring a > bell. But, does it chime in tune? 450 MIPS on a z/990. 585 MIPS on a z/9 EC. Now, you know! But, what does it mean? If you had a workload to compare, it might tell you something. Otherwise, the 'M' still stands for 'Meaningless'. You need to look at more than just MIPS to understand what the environment does. - Too busy driving to stop for gas! -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
It means that IBM has z9 BC vs z900 performance benchmarks. MHZ ratings are not
meaningful across processor architetures, and are not management
meaningful to any useful degree. IBM's benchmarks of performance between a z9
BC and a z900 are well understood by IBM and that compartive
information is management meaningful.
Ted MacNEIL <[EMAIL PROTECTED]>
Sent by: Linux on 390 Port
To
LINUX-390@VM.MARIST.EDU
cc
07/19/2007 03:42 PM
Subject
Re:
zSeries IFL speed rating
Please respond to
Linux on 390 Port
>IBM has those* numbers availble and it is still management meaningful.
Huh? What does that statement even mean?
IBM has numbers, but they're no longer meaningful!
-
Too busy driving to stop for gas!
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
From Dennis Roach: "I'd just use MSUs. Eg. a 40MSU processor is 2x a 20MSU processor." Be careful using MSUs to report on processor speed. MSUs are strictly for software pricing and are tweaked with each processor upgrade to keep software costs somewhat within line. For example a 2064-1C4 processor of 843 MIPS is rated at 153 MSUs for a 5.51 factor, a 2086-470 processor is 1313 MIPS/208 MSUs (6.31), a 2096-S07/U04 processor is 1004 MIPS/138 MSUs (7.28), and a 2094-404 processor is 740 MIPS/103 MSUs (7.18). So a 40 MSU processor is not necessarily 2x a 20 MSU processor. Tom Kelman Commerce Bank of Kansas City (816) 760-7632 -Original Message- From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of McKown, John Sent: Thursday, July 19, 2007 12:13 PM To: LINUX-390@VM.MARIST.EDU Subject: Re: zSeries IFL speed rating > -Original Message- > From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On > Behalf Of Roach, Dennis > Sent: Thursday, July 19, 2007 11:59 AM > To: LINUX-390@VM.MARIST.EDU > Subject: Re: zSeries IFL speed rating > > > I have been asked to determine the improvement of an IFL on a > z9 BC over a z900. > Mhz is something the people being presented to understand. > It does not compare to the same speed on say INTEL. > > Dennis Roach I'd just use MSUs. Eg. a 40MSU processor is 2x a 20MSU processor. They should be able to understand that, if all they want is a relative comparison. Of course, your-mileage-may-vary because you are not only getting a faster cycle time, you are getting a different microcode implementation. So some instructions may be faster while others could even be slower (not likely). That is, each instruction might vary in performance relative to the same instruction on the other machine. If they get cranky, ask them if they think it is correct to compare the MHz rating of a Celeron, Pentium, and Xeon processor. If they do, then they're dangerously ignorant. A Xeon CPU with a lower MHz is much more productive than a Celeron with a faster MHz. -- John McKown Senior Systems Programmer HealthMarkets Keeping the Promise of Affordable Coverage Administrative Services Group Information Technology The information contained in this e-mail message may be privileged and/or confidential. It is for intended addressee(s) only. If you are not the intended recipient, you are hereby notified that any disclosure, reproduction, distribution or other use of this communication is strictly prohibited and could, in certain circumstances, be a criminal offense. If you have received this e-mail in error, please notify the sender by reply and delete this message without copying or disclosing it. -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 * If you wish to communicate securely with Commerce Bank and its affiliates, you must log into your account under Online Services at http://www.commercebank.com or use the Commerce Bank Secure Email Message Center at https://securemail.commercebank.com NOTICE: This electronic mail message and any attached files are confidential. The information is exclusively for the use of the individual or entity intended as the recipient. If you are not the intended recipient, any use, copying, printing, reviewing, retention, disclosure, distribution or forwarding of the message or any attached file is not authorized and is strictly prohibited. If you have received this electronic mail message in error, please advise the sender by reply electronic mail immediately and permanently delete the original transmission, any attachments and any copies of this message from your computer system. * -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
> -Original Message- > From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On > Behalf Of Kelman, Tom > Sent: Friday, July 20, 2007 8:32 AM > To: LINUX-390@VM.MARIST.EDU > Subject: Re: zSeries IFL speed rating > > > From Dennis Roach: > > "I'd just use MSUs. Eg. a 40MSU processor is 2x a 20MSU processor." > > > Be careful using MSUs to report on processor speed. MSUs are strictly > for software pricing and are tweaked with each processor > upgrade to keep > software costs somewhat within line. For example a 2064-1C4 processor > of 843 MIPS is rated at 153 MSUs for a 5.51 factor, a > 2086-470 processor > is 1313 MIPS/208 MSUs (6.31), a 2096-S07/U04 processor is > 1004 MIPS/138 > MSUs (7.28), and a 2094-404 processor is 740 MIPS/103 MSUs > (7.18). So a > 40 MSU processor is not necessarily 2x a 20 MSU processor. > > Tom Kelman > Commerce Bank of Kansas City > (816) 760-7632 Good point. The marketing people took a very good measure of performance, which was originally designed to rationalize "work" across multiple different processors and royally . I am really beginning to HATE software vendors and their marketing techniques. -- John McKown Senior Systems Programmer HealthMarkets Keeping the Promise of Affordable Coverage Administrative Services Group Information Technology The information contained in this e-mail message may be privileged and/or confidential. It is for intended addressee(s) only. If you are not the intended recipient, you are hereby notified that any disclosure, reproduction, distribution or other use of this communication is strictly prohibited and could, in certain circumstances, be a criminal offense. If you have received this e-mail in error, please notify the sender by reply and delete this message without copying or disclosing it. -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
On 7/20/07, James Melin <[EMAIL PROTECTED]> wrote: > It means that IBM has z9 BC vs z900 performance benchmarks. MHZ ratings are > not meaningful across processor architetures, and are not management > meaningful to any useful degree. IBM's benchmarks of performance between a > z9 BC and a z900 are well understood by IBM and that compartive > information is management meaningful. So I guess "management meaningful" means we know it's BS but the boss will buy it and it gets him off your back? Yes, sometimes you give in - like I did in the following anecdote :-) A large outsourcing company was using the IBM published ratings for various CPU models to normalize the CPU usage of each image. Simple - you divide the total cost of the computing center by the "normalized used MIPS" to give you the $ per normalized CPU hour, and then cut the bill for each LPAR. As you should know, the models are rated on "single z/OS image throughput" and MP overhead in MVS means a 20-way z/OS image does not do 10 times more than a 2-way image. The bean counters now conclude that normalization of the 20-way is different from the 2-way, so a single CPU hours on a 20-way will be cheaper than on a 2-way... Among others, we ran a production LPAR and a small test LPAR on two different machines: a 6-way and a 16-way. By swapping the two LPARs so that production runs on the 16-way, we saved money... Both were z990 so we used the same amount of CPU hours. But since the bean counters had decided the per-CPU capacity of a 16-way was only 75% of the CPU in a 6-way, we reduce our CPU cost with 25%. The test system did not use a lot, so we could afford to run that on a more expensive CPU. I have also defended this claiming the tests would be better on an expensive CPU :-) Rob -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
>>> On Fri, Jul 20, 2007 at 2:16 AM, in message <[EMAIL PROTECTED] .bisx.prod.on.blackberry>, Ted MacNEIL <[EMAIL PROTECTED]> wrote: -snip- > 585 MIPS on a z/9 EC. What about on a z9 BC? > Now, you know! > But, what does it mean? > If you had a workload to compare, it might tell you something. > Otherwise, the 'M' still stands for 'Meaningless'. I think just about every experienced mainframe technician in the world knows this. Harassing them because they have to deal with non-technicians that _don't_ understand it doesn't make this a very friendly place to come for help. (Not picking on you specifically, a number of people jumped on that bandwagon.) It's also difficult to compare workloads on two different generations of hardware, when you only have one of those generations at hand. Not everyone has the time, budget, management patience, or whatever to do what is needed to get around that. Mark Post -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
On Friday, 07/20/2007 at 10:51 EDT, "McKown, John" <[EMAIL PROTECTED]> wrote: > Good point. The marketing people took a very good measure of > performance, which was originally designed to rationalize "work" across > multiple different processors and royally . I am > really beginning to HATE software vendors and their marketing > techniques. Nah. What you hate is that technology has surpassed our collective ability to rationalize it. When MHz had a linear correlation to MIPS and no one had heard of NUMA or pipelining, life was good. We went from measurements that reflected cycle time (MHz) to instruction execution power (MIPS), even though it was best expressed as a range, to one that tries to represent computing capacity and throughput (MSUs). Why? As was said, to bring some sanity to software pricing. Allowing our bosses to believe they can do capacity planning on the back of a paper napkin borders on professsional suicide. They will ultimately make the wrong decision using bad or misleading data and rolls downhill. [BTW, paper napkins are de rigeur for all significant planning exercises. Beware.] If the boss wants to know how much money she needs to budget in 2007 to allow for organic growth and new projects, help her. If she wants to know the cycle time of the CPU, ask why. While it's an entertaining trivia question asked at all the *best* cocktail parties thrown by engineers, it's like nails on a chalkboard to hear an IT manager ask it. (Correct answer: "For what model?", resulting in knowing nods from around the room. Depending on how much the party-goers get out, you might even quip "Laden or unladen?".) I'm trying to get folks to quite charging for each CPU-minute a z/VM guest consumes. It doesn't accurately (or even remotely) measure the IT resources (people, power, space, machines) it really does chew up. And buying newer, faster machines just spins the meter dial faster, ultimately making the user go buy their own system. "There's a hole in my bucket, dear Liza" It ain't 1980 any more and it hasn't been for quite a while. :-) Alan Altmark & Co. -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
And they use the small cocktail napkins not the large lap covering napkins Bob Lawrence DBA Boscovs Dept Stores LLC -Original Message- From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of Alan Altmark Sent: Friday, July 20, 2007 3:16 PM To: LINUX-390@VM.MARIST.EDU Subject: Re: zSeries IFL speed rating On Friday, 07/20/2007 at 10:51 EDT, "McKown, John" <[EMAIL PROTECTED]> wrote: > Good point. The marketing people took a very good measure of > performance, which was originally designed to rationalize "work" across > multiple different processors and royally . I am > really beginning to HATE software vendors and their marketing > techniques. Nah. What you hate is that technology has surpassed our collective ability to rationalize it. When MHz had a linear correlation to MIPS and no one had heard of NUMA or pipelining, life was good. We went from measurements that reflected cycle time (MHz) to instruction execution power (MIPS), even though it was best expressed as a range, to one that tries to represent computing capacity and throughput (MSUs). Why? As was said, to bring some sanity to software pricing. Allowing our bosses to believe they can do capacity planning on the back of a paper napkin borders on professsional suicide. They will ultimately make the wrong decision using bad or misleading data and rolls downhill. [BTW, paper napkins are de rigeur for all significant planning exercises. Beware.] If the boss wants to know how much money she needs to budget in 2007 to allow for organic growth and new projects, help her. If she wants to know the cycle time of the CPU, ask why. While it's an entertaining trivia question asked at all the *best* cocktail parties thrown by engineers, it's like nails on a chalkboard to hear an IT manager ask it. (Correct answer: "For what model?", resulting in knowing nods from around the room. Depending on how much the party-goers get out, you might even quip "Laden or unladen?".) I'm trying to get folks to quite charging for each CPU-minute a z/VM guest consumes. It doesn't accurately (or even remotely) measure the IT resources (people, power, space, machines) it really does chew up. And buying newer, faster machines just spins the meter dial faster, ultimately making the user go buy their own system. "There's a hole in my bucket, dear Liza" It ain't 1980 any more and it hasn't been for quite a while. :-) Alan Altmark & Co. -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
Here is good site that lists some performance guesstimates. Remember that the specialty engines run at full speed. So in the case of a z/9 EC the IFL can be compared to the engine on the model 701, on your z/900 the engine can be compared to the model 101,1C1 or 2C1 depending on your current model and the processor card that is installed. http://www.tech-news.com/publib/ John -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
Yeah, I'm fresh out of whatever. And I'm running a bit low on stuff, thingummies, and whatsits, too. Jon Not everyone has the time, budget, management patience, or whatever to do what is needed to get around that. -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
Thomas Kern wrote: Since Mhz and MIPS are such misused values, I prefer to run the same program on old and new engines to compare the performance change. I use an old FORTRAN (no flames please) program that computes pi to 5000 places. A boss once needed something to see if the vendor really did upgrade our processor, just after our Performance and Capacity Planning group was disbanded. So on an idle system, this just gets into memory and runs with a high Total/Virtual CPU ratio. On my z890 IFL, 20 iterations takes about 8.571 sec virtual, 8.578 sec total and 8.944 sec elapsed time. Find a favorite program that is repeatable and keep using that to test the speed of your engines. But all numbers must be taken as relative and with a big grain of salt. CPU power is only ever part of the story; if it's all that matters, you'd all be using Xeons or Opterons. Do a proper benchmark, one that reflects what you want to do. Even if your current workload is constrained by CPU, doubling the speed of the CPU or doubling the number may well do nothing than find the next bottleneck. In a queue at the theatre, everyone might be lining up to have their tickets checked at the door and thereafter being shown quickly to their seats. If there's a big queue at the door, getting more ticket checkers won't help if you can't also show people to their seats more quickly. Improving one component of a balanced system just makes the system unbalanced. -- Cheers John -- spambait [EMAIL PROTECTED] [EMAIL PROTECTED] Please do not reply off-list -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: zSeries IFL speed rating
I didn't need a full throughput benchmark, nor did I need to benchmark the I/O subsystem or the tape drives. The boss asked a specific question about the CPU power. I found a program that answered his question to his complete satisfaction. Now if he had asked for a throughput benchmark, an orchestrated set of OfficeVision activity would have shown him that IBM had badly understated the need for increased real memory. But that isn't what he wanted. /Tom Kern --- John Summerfield <[EMAIL PROTECTED]> wrote: > CPU power is only ever part of the story; if it's all that matters, > you'd all be using Xeons or Opterons. > > Do a proper benchmark, one that reflects what you want to do. Even if > your current workload is constrained by CPU, doubling the speed of the > CPU or doubling the number may well do nothing than find the next > bottleneck. > > In a queue at the theatre, everyone might be lining up to have their > tickets checked at the door and thereafter being shown quickly to their > seats. If there's a big queue at the door, getting more ticket checkers > won't help if you can't also show people to their seats more quickly. > > Improving one component of a balanced system just makes the system > unbalanced. > Pinpoint customers who are looking for what you sell. http://searchmarketing.yahoo.com/ -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
