Problems with AIX

Quicksearch

Disclaimer

The individual owning this blog works for Oracle in Germany. The opinions expressed here are his own, are not necessarily reviewed in advance by anyone but the individual author, and neither Oracle nor any other party necessarily agrees with them.

Navigation

< Problems with AIX | Homeoffice >

Sunday, May 13. 2007

The Register writes to the the delay of AIX 5.4:

Most notably, IBM has shifted from an out of order design with Power5's integer units to an in order design with Power6. Such a drastic change will require first AIX 5.4 and then a full software recompile should customers want to see maximum performance. IBM has yet to say anything at all about the efforts underway to deal with this issue.

This would be really interesting for ISV. It seems, you need a pre-Power6 version and a Power6-Version of your code to get maximum performance. I assume, this will lead to nice benchmarketing tricks with performance levels unreachable for users with standard software from ISV.

Posted by Joerg Moellenkamp in English, The IT Business at 06:58 | Comments (5) | Trackbacks (0)

Defined tags for this entry: aix

, ibm

, power6

, The IT Business

Related entries by tags:

View as PDF: This entry | This month | Full blog

Trackbacks

Trackback specific URI for this entry

No Trackbacks

Comments

Display comments as (Linear | Threaded)

I don't think in-order versus out of order execution is that big of a deal. Certainly in the HPC space, customers compile their code for maximum performance. But as long as the ISA is the same, things like in-order vs. out of order execution, prefetch, cache architecture, etc. should not matter.

I would note UltraSPARC has always been in-order execution, while SPARC64 has always been out of order execution. The same Solaris applications ran fine on both.

#1 Mark on 2007-05-13 19:36 (Reply)

As soon as you use optimizing compilers this is not the case. I saw already programs, that were heavly optimized towards the cache hierarchy of SPARC IV+.

I rememeber, that for example different AIX released made a big performance difference. One versions had optimizations for a new Power proc, the other hadn´t them.

#1.1 Joerg (Homepage) on 2007-05-14 15:15 (Reply)

I just had IBM visiting us today. It was really a lot of benchmarketting. Sun was also mentioned a lot, despite it was a presentation about IBM Products

Even when there is a performance hit for non-optimized applications on Power6, IBM will downplay this (like my question about the hypervisor/vio server overhead...)

But then again... Power6 really looks like a fast CPU, the question is, how much optimization is relevant....

PS: IBM really tries to sell only HW, they did not mention AIX that much, maybe they are ashamed of it

PS2: It was also funny to listen to sales-engineers who know Sun/Solaris only from benchmarks and gartner/idc studies....

#2 Mika on 2007-05-14 22:16 (Reply)

Thanks god the Java Community does not have to care about those optimizations at all...

#3 Bernd Eckenfels (Homepage) on 2007-05-17 18:58 (Reply)

Not exactly: This is a thing JVM developers have to take into consideration. I think they are part of the Java community as well.

And in the end: When the JVM doesn´t optimize to a new processor and the groundlaying operating system isn´t optimized toward the processor, the java developer looses performance. Furthermore: They can´t do anything about it.

#3.1 Joerg (Homepage) on 2007-05-17 20:16 (Reply)

Add Comment

Name
Email
Homepage
In reply to
Comment	E-Mail addresses will not be displayed and will only be used for E-Mail notifications. To prevent automated Bots from commentspamming, please enter the string you see in the image below in the appropriate input box. Your comment will only be submitted if the strings match. Please ensure that your browser supports and accepts cookies, or your comment cannot be verified correctly. Enter the string from the spam-prevention image above: Standard emoticons like :-) and ;-) are converted to images. Enclosing asterisks marks text as bold (word), underscore are made via _word_.
	Remember Information? Subscribe to this entry