back to article IBM debuts new Power7 iron this week

It looks like IBM is getting ready to roll out some more Power7-based systems this week, as well as an update to its proprietary midrange operating system, formerly known as OS/400 and now called i for Business, or i for short. The details are a bit thin, but resellers downstream from master distributor Arrow Electronics' …

COMMENTS

This topic is closed for new posts.
Silver badge
WTF?

As expected.....

.... another IBM Power upgrade means another new version of the OS to actually get the benefits of the new CPU.

0
2

Whats that supposed to be?

Hey Matt, is that all of a flamebait you can muster?

Have a decent coffee and and try again.

0
0
Thumb Down

Oh ?

So what OS version does the POWER7, as I have heard it blades, then need ?

AIX version ... what ?

Lets guess 5.3 and 6.1, nothing new there.

// jesper

0
0
Silver badge
Happy

RE: Whats that supposed to be?

"....is that all of a flamebait you can muster?...." Insert faux surprise here that the IBMers aren't keen to admit an upgrade to Power7 is pointless unless you revalidate/upgrade your whole software stack. Just like it as from Power5 to Power6. Whilst the IBM trolls like to froth about fantasy performance comparisons and make all kind of implications about Itanium, they aren't happy to discuss known and admitted issues with Power. Of course, Tukzilla will run all the current hp-ux, OpenVMS, Windows and Linux stacks as currently run on Montvale Itanium kit without the need for a change and with improved performance. And we all know changing the stack is a lot harder and entails more risk than changing the underlying hardware. Discuss!*

/SP&L

*IBM trolls - whilst this is an invite for you to respond, please try and formulate a logical argument, not just the usual "Power is great, Itanic is dead - you know it, you know it, I'll scream and cry until I'm blue in the face if you all don't admit it!"

0
2
Thumb Up

Hi Matt!

Nice to see you back - it's been too quiet ...

"another IBM Power upgrade means another new version of the OS to actually get the benefits of the new CPU"

No actually, Power7 needs either AIX7 (when it appears) or AIX6 with an appropriately modern patch level. If this wasn't the case, then since even El Reg has been saying that the P7 kit has been out there for a while but AIX7 isn't out, what have those early-adopters been using? <grin>

It's the same deal as in the past - e.g. running AIX5.3 on Power6 - you get most of the benefits with an 'old' OS, and the remaining one or two features if you use the 'new' version of AIX.

Got to say that - even as an HPer - I'm quite impressed with what our neighbours in Austin have come up with this time.

0
0

title

"an upgrade to Power7 is pointless unless you revalidate/upgrade your whole software stack"

Wrong.

Aix 6.1 is required. 5.3 will also work, but lack features (quad-SMT).

No discussion neccessary.

EOT

0
0
rch
FAIL

AIX6.1

# another IBM Power upgrade means another new version of the OS to actually get the benefits of the new CPU.#

Nope. The benchmarks are done on AIX6.1 so you get the benefits of the new Power7 CPU now.

It must really suck to be a HP fanboi like you.

0
0
Silver badge
Troll

RE: Oh ?

"So what OS version does the POWER7...need ?...." Aw, ickle Jesper is trying to play all innocent, when he knows perfectly well that P6/6+ was a complete change from P5 (from out-of-order execution to in-order) and that now P7 is a return to little more than lots of P5 cores shrunk onto the same die (so back to out-of-order, because IBM hit the buffers on the P6/6+ design). Which means any apps written for P5 had to have a rewrite to run best on P6 (otherwise they could even perform WORSE on P6 than they did on P5!), and now P6/6+ apps will need a rewrite to be optimised for P7. He also knows that the few new features in P7 won't be unlocked until customers get the next gen of AIX, version 7.x (by the way, when is that likely to be?). All the P6 apps won't work to their best until they're rewritten for P7 and AIX 7.x (and all the iSeries apps too will need updating by the look of this article). And considering that only a fool would put a brand-new version of an OS into production, that means at least six months (probably a year) after the Power7 servers actually become available to customers they will still be stumbling along with AIX 6.1. Who knows when the combination of new CPU, new chipsets, new OS and rewritten apps will be ready for primetime? 2011?

".....Lets guess 5.3 and 6.1....." Not if they want to get what they paid for! In fact, if the app they run has been optimised for AIX 6.1 on P6/6+ then it may even go SLOWER on P7! And does IBM have longterm plans to develop into 5.3 the features of 6.1 or planned for 7.x? More to the point, will they really be happy to support AIX 5.x, 6.x and 7.x concurrently? And what's the point of upgrading to P7 instead of just buying the current (and price-reduced) P6/6+ range if you're not getting what IBM will be charging you the extra for?

The truth is IBM rushed forward the Power7 release becasue Nehalem and Tukzilla were going to arrive first and they wanted to have something to announce. So they had to announce it long before the high-end servers were ready, and long before AIX7 was even close to release. Jesper knows that, despite his pretending it has no impact.

/SP&L

0
2
Silver badge
FAIL

@Matt

WTF are you talking about!

Any AIX code that is compiled with the compiler defaults will work. It just does, and has done for years.

There is a good chance (greater than 80%) that if you pulled a binary compiled for RS64 or Power2, or even the original RIOS chipsets from a system running AIX 4.3.2 or later, and placed it *WITHOUT CHANGE* on an AIX 6.1 system running Power7, it would run.

*IF* you are talking about extracting every ounce of performance for some code, then I agree that if you have optimized the code for cach size and processor affinity or the particular properties of the floating point pipeline (as is being done where I work currently) or any number of other factors, you will need to re-compile it with the relevant options to get the best performance, but even then the Power6 optimized code will probably run on Power7. This is not new nor specific to IBM processors, and hasn't been ever since I was working on PDP/11's and VAXes.

It is acknowledged that there will be a different execution profiles for Power7 from Power6. The design requirements for the processor were different, and I talk to some worried people talking about how scalable their model is at the moment. We currently have some code that tops out at about 768 processors, and gets worse if you increase the number above this. If the overall clock speed is dropped without a corresponding increase in the number of instructions per clock cycle, then this code may well be slower on Power7 than Power6.

But it's not certain. The speed of the level 3 cache, together with the less deep pipeline required for slower clock speeds may just offset the drop in overall clock speed (pipeline stalls become less of a problem). And as the bottleneck with the code is in inter-thread communication, the high bandwidth between cores on the same die, and on the same QCM may also offer realistic hope of performance gain. And the interconnect between the nodes in an IH supernode should also perform better than what is currently used. And I believe that the number of possible in-flight speculative execution threads possible with more available execution units may reduce pipeline stalls even more.

Remember the Pentium 4 vs. Pentium 3 debacle, where a Pentium 3 at the same clock speed out performed the Pentium 4 when it was first launched, and the follow up Pentium M and D processors dropped the clock speed and again outperformed a Pentium 4.

The talk of in-order vs. out-of-order is bogus. The results for the same stream of instructions should be the same regardless of whether it is in-order or out-of-order. If it's not, the processor is broken. The difference is that out-of-order may allow the hardware instruction scheduler to better utilize the available execution units, leading to more-instructions per clock.

And anyway, the only applications that really are affected by the clock speed are speed-daemon floating point hungry single threaded research types of workloads. From my 30+ years of experience, this is a very small (but admittedly valuable) part of the AIX customer base.

For your commercial workload customer, the difference between Power5, Power6 and Power7 architecture and instruction sets will be largely ignorable. What will be more important is the number of cores, the number of simultaneous threads that can be executed and the memory constraints of the hardware. For these customers, the application providers may not even have optimized versions of their code for the different processors, but a one-size-fits-all distribution. Power7 is probably a big step forward for them. I know application providers who still compile on Power4 hardware with a generic set of compiler switches to allow the code to run on the entire processor set.

In fact, AIX is like this. The version of AIX 5.3 that runs on the Power6 575's (the current speed freak machine) is EXACTLY THE SAME as that which runs on an RS64 44P 170 (same install disks, I know, because I have done it!). There is no distinction on FixCentral for patches. It is just all the same.

Your comments about AIX5.3 are interesting. Yes, there are some features that will not be available to AIX5.3. Things like Turbo mode. But AIX 5.3 is still supported (not sure if there is an end-of-life date yet, would expect it if and when AIX 6.2 or AIX 7.1 is available), and there is likely to be a new Technology Level (TL12?) for Power7 that will add some of the new features. IBM have always kept the latest 2 releases of AIX under active support, and normally publish withdrawal from marketing for an OS about 2 years prior to that date. This means that there is at least 2 years of AIX 5.3 support, not that I would recommend anybody installing Power7 to use AIX 5.3 at this time.

The instruction features are more likely to be conditioned by the compilers (which, incidentally, are currently OS version agnostic, the Fortran and C compilers are the same packages for AIX 5.3 and AIX 6.1). What's even more impressive is that you can compile Power6 code on, say, a Power5 system, drop it onto a Power6 box, and expect it to run as well as if it had been compiled on a Power6 system.

All of this does not sound like the scenario you paint. Maybe you ought to work in a big AIX shop sometime, and see what binary compatibility is really like.

BTW. I cannot say where I am working, but we have more than one cluster in the first 100 of the November 2009 Top 500 Supercomputer list.

1
0
Coffee/keyboard

Get a grip Matt (if this really is Matt talking here)

Matt, you are to intelligent for this broadside. Let's try to debunk this well.. lack of knowledge. I always like to see the best in people.

InOrder versus OOO.

Is POWER6 an inorder core. Yes. It sure is. Do I think it was totally bonkers to make a inorder (Mostly Inorder as FP instructions could actually be .. well now it gets tricky, cause there is both an issue and a shedule on POWER.) out of order, I sure did.

Will this have an effect on code compiled for specific for POWER6. Yes. How big is this ? I don't know.. I just know that I was involved in a tuning project last year, where we moved the same binaries, and mksysb and all from POWER4->POWER5->POWER6 and our performance numbers hit exactly what we expected +/- a few percent on rPerf. So.. well.. Also as several people have mentioned here then you get new versions of the critical libraries with new patches. But if I were to handcode a piece of code, link it statically for POWER6 and on POWER6 and then run it on POWER7 would I then get a performance hit you can see. Sure..

"P7 is a return to little more than lots of P5 cores shrunk onto the same die"

That is just wrong and you know it. POWER7 is a whole different beast as you can see here:

http://eecatalog.com/power/2009/10/09/power7-power-pc-processor-cores-2/

The differences are so many, that this link should be enough for even the hardest sceptic.

IMHO the reason why POWER6 is different is cause it was originally intended to slot in the mainframes also.

This is of cause just pure speculation on my side, but perhaps someone here knows better than me.

"Which means any apps written for P5 had to have a rewrite to run best on P6 (otherwise they could even perform WORSE on P6 than they did on P5!), and now P6/6+ apps will need a rewrite to be optimised for P7. He also knows that the few new features in P7 won't be unlocked until customers get the next gen of AIX, version 7.x (by the way, when is that likely to be?). All the P6 apps won't work to their best until they're rewritten for P7 and AIX 7.x (and all the iSeries apps too will need updating by the look of this article). And considering that only a fool would put a brand-new version of an OS into production, that means at least six months (probably a year) after the Power7 servers actually become available to customers they will still be stumbling along with AIX 6.1. Who knows when the combination of new CPU, new chipsets, new OS and rewritten apps will be ready for primetime? 2011?"

What the f word are you talking about ? Rewrite the applications? Why would you need to rewrite the applications to take advantage of a new generations of processors ? It makes absolutely no sense. And you talk about chipset and new OS. Basically

dy knows that you won't get the same performance out of POWER7 on AIX 5.3, it doesn't support SMT-4.

nce report that describes the rPerf numbers you will get on AIX 5.2.

ftp://public.dhe.ibm.com/common/ssi/pm/rg/n/poo03017usen/POO03017USEN.PDF

For example you only get 73.81 rPerf on a 16 way 2.2Ghz POWER5+ p570 machine on AIX 5.2 compared to 95.96 on AIX 5.3.

IT IS NOT A SECRET.

And you don't know much about how iSeries work.

// Jesper

0
0
Troll

Supposed to be wrong

Matt Bryant (HP/UX apologist wrote):

"Insert faux surprise here that the IBMers aren't keen to admit an upgrade to Power7 is pointless unless you revalidate/upgrade your whole software stack. Just like it as from Power5 to Power6."

Naughty Matt, don't know how many customers you've had to deal with, but I can't think of many - sorry any - of mine that have had to validate the software stack moving from P5 -> P5+ -> P6. Yes, there has been an APAR (patch) or two to support the new processor better - but that's it. It's pretty painless! On the other hand, am I the only one wondering if IBM will make a better job of AIX7 than they did with AIX6 - so many faults, with so many patches! I bet there were a few banged tables and red faces there...

Anyway MB, if you trot out this kind of marketing droid spiel (the quoted bit) then you end up looking like a clueless marketing droid yourself, (see icon for details). It certainly doesn't help persuade folks to move from AIX-on-Power to Linux-on-Xeon, (notice that I didn't say HP/UX-on-Itanium because I still remain to be convinced of the overall technical benefits of EITHER of those - but then again I came originally from OpenVMS, so you can maybe understand my bias against the young pretender HPUX*).

(*and HP/UX's virtualization is pretty poor v's AIX's - but then again IBM can get a hardware assist from the processor, HP/UX has to rely on what Intel deigns to provide)

PS Not all HPers have "HP/UX" and "Itanium" tattooed on each knuckle - some of us are actually quite sensible! :D

0
0
Anonymous Coward

Itanium is dead

The largest software vendor MS has pulled the plug. Titanic has no future. I hear Intel manegement is going to kill Titanic. Only POWER7 has a bright future.

1
1
Gold badge

Itanium & Power7

"The largest software vendor MS has pulled the plug."

MS is not the largest software vendor for Itanium, they are in fact quite minor (I've heard 5%)... from all I've heard, the port of Windows to Itanium was quite second-class and pretty much a joke. I don't think the Itanium market is particularly healthy but wouldn't take MS pulling out as a sign of anything.

@Matt Bryant: You really seem to be making a mountain out of a molehill. A newer OS will get some extra speed boosts on the POWER7. This is fairly normal -- Windows doesn't tend to benefit so much, but Linux, MacOS, various UNIXes, they tend to have cache handling improvements, scheduler improvements, and little tweaks for newer chips as they come out to allow for somewhat faster & more efficient operation... even just running on x86/x86-64. Doesn't mean (necessarily) that an older version won't run.

0
0
Silver badge
Flame

Slowaris had this too

>All of this does not sound like the scenario you paint. Maybe you ought to work in a big AIX shop sometime, and see what binary compatibility is really like.

In my experience Solaris was awesome at this too. You could run SunOs 4 binaries on the latest Solaris 10 box no problems. Not breaking driver apis every two weeks ala linux were nice also. Too bad Solaris was unbearably slow (expecially on Sparc), expensive and will soon be a memory we can tell our kids about ala DEC, Wang, or Osborne computers.

0
0
Thumb Down

Trollage

asdf wrote: "Not breaking driver apis every two weeks ala linux were nice also"

Nope - been using my Ubuntu 8.04LTS for about two years now quite happily and haven't had to replace drivers, other than for the normal bug fixes and feature upgrades (same as Windows).

asdf wrote: "Too bad Solaris was unbearably slow (expecially on Sparc), expensive and will soon be a memory"

Partially right. Solaris is expensive and I'm pretty sure that Horracle will make damned sure that "proper" Solaris (on SPARC) dies a long and painful death. Then again, OpenSolaris seems to be running a treat, and everyone I've heard has a good feeling about Solaris10 on Intel processors. As to the speed, I'm not sure what systems you've been using (maybe one's that you've setup - badly!) but Sol boxes I've always found to be acceptably speedy, and I've been Solaris-ing from 2.4 through 2.10, sorry "10" on both SPARC and Intel. Okay, I'll admit that the startup and shutdown times aren't exactly rapid, but then again the uptimes on our Solaris boxes are such that this isn't really an issue.

0
0
Silver badge
Flame

I will take the flamebait

>The largest software vendor MS has pulled the plug. Titanic has no future. I hear Intel manegement is going to kill Titanic. Only POWER7 has a bright future.

Watch the POWER line will only last as long as IBM is able to get game console contracts with it. That is the only way they can get the volume to justify ever increasing costs for each new generation.

0
0
Troll

Flame on dude

asdf wrote: "Watch the POWER line will only last as long as IBM is able to get game console contracts with it. That is the only way they can get the volume to justify ever increasing costs for each new generation."

To misquote Monty Python - Life of Brian:

What has Power ever done for us - apart from servers?

Games consoles.

Okay, apart from servers and games consoles, what has it ever done for us?

Printer controllers

Okay, servers, games consoles and printer controllers ...

And NAS boxes

Okay servers, consoles, printer controllers, NAS boxes ...

Dont forget robotics

Okay, servers, consoles, printer controllers, NAS boxes, robotics

Etc, etc, etc

Seriously, if you think Power will die off just because XBox720, PS4, Wii+ don't use it then you're fooling yourself. Apart from the small matter that IBM sell a _LOT_ of servers with Power, you've also got all the embedded processor spin-offs.

On the other hand we've got Microsoft going very cool on Itantium, RedHat saying that RHEL6 won't support it, and others playing a "wait and see" on it. You've also got Xeon's looking better and better performance-wise, and since they _do_ have the software support, you've got to wonder what the heck Itanium is good for? That being the case I really see Intel keeping the Itanium fabs going just to feed HP with processors for HP/UX and OpenVMS ... NOT!

0
0
FAIL

@Matt

It is pretty funny to notice that MB is running out of ammo.

But, the most funny thing is that HP uses for a long time a pretty crap PA-RISC emulator called Aries to run these codes on Itanium.

Perfornance degratation ranges from 20% to 80% if those apps are not coded to run on Itanium.

You waste your money using tech craps like Itanium, not POWER nor AIX.

The market share speaks for itself.

Adios, and hit the iceberg with Itanic, die slowly.

Cheers!

0
0
Pint

@Matt

Is the local HP Rep in your town of residence also the local dope pusher ? You write tons of crap because you need your next shot so badly ?

Power has been for a long time simply the fastest CPU *of all*. I once tested a client/server application of mine on a power system with four CPUs and on normal PCs. I got ca 1000 requests/sec on the x86 and about 12000 requests/sec on the power machines.

So the power CPUs were probably three times (!) faster than the x86 CPUs.

spec.org says basically the same to the present day. And Itanic does not even match the current x86 processors.

I suspect the Itanic transport trays are made out of coke and an ingenious scheme of the Cali Cartel to import stuff. That's why Matt likes Itanic.

Matt, switch to beer and give Power a test drive.

0
0
Silver badge
Happy

..... and now we reel them in!

Dear IBM trolls,

I can't help laughing at how readily you piled into my trap. I hardly had to hold the door open and you all rushed in like headless chickens on acid! Especially the ever-frothing jlocke - what a picture of informed logic he is not! Is the P6 to P7 upgrade a big issue? Not really, not unless you really want to tune the heck out of an application. Will P6 applications run fine on P7? Probably, for 99% of applications. Did you all get irate over my little piece of FUD? 100% definately!

So, the next time you get a hankering to post some FUD about Linux, or hp-ux, or Windows, or Itanium, or Xeon, or just about anything not IBM, please think back to this little lesson and hopefully come to the conclusion that maybe FUD is not big or clever, and our industry would be a lot better off if vendors stopped pushing it.

/SP&L, espcially @ jlocke!

0
1

Translation:

I may translate:

It was all a joke, I wasn´t that thick.

Really.

told you to get a coffee first...

0
0
Silver badge

What a climbdown

Shock news....

Matt Bryant, using an amazing piece of ass-covering, claimed that his original post was just troll bait, rather than a real comment.

Unfortunately, the more sensible members of the Register commenting community were able to see through this with apparent ease, identifying Matt as one of the trolls that he claims to be targeting.

0
0
Silver badge
Happy

RE: What a climbdown

"....Matt Bryant, using an amazing piece of ass-covering...." No, Peter, it was just I knew of the forthcoming Integrity blades announcement and anticipated the frothing of the IBM trolls, so I though I might point out the folly of their ways to you. As expected, you lot didn't fail and performed to cue like the mindless, clockwork monkeys you are! I will enjoy reading your "thoughts" after the hp announcement, in which I am sure you will condemn the hp kit without having even seen it, let alone tried it, and no doubt for your own, completely unbiased reasons. Enjoy your frothing! Try not to get too much spittle on your keyboard when venting.

".....Unfortunately, the more sensible members of the Register commenting community...." Don't worry, the "sensible members" have long since marked out the trolls, and we can spot you a mile off!

/SP&L

0
1
Anonymous Coward

Matt - Serial FUD

Matt,

You are the mindless clockwork monkey - posting the same non-sense in another thread:

http://forums.theregister.co.uk/forum/1/2010/04/16/ioug_oracle_survey/

On the 20th April - 7 days after admitting it was FUD here - you say in that thread .....

"and set aside more money for when you have to upgrade and recertify your stack with AIX7.x if you want the best performance on Power7."

So now we can see you are serially posting material which you admit yourself is FUD.

No doubt you'll spout more excuses - but it doesn't much matter - we all see you for what you really are ......

0
0
This topic is closed for new posts.

Forums