tag:blogger.com,1999:blog-15440695.post5424170103824817092..comments2024-02-29T10:39:57.857-05:00Comments on Alex Fatkulin's Blog: Oracle SuperCluster M7 SLOB LIO Tests vs Intel Xeon E5-2699 V4Alex Fatkulinhttp://www.blogger.com/profile/06361288475877100451noreply@blogger.comBlogger7125tag:blogger.com,1999:blog-15440695.post-37427290674398657552017-05-31T10:11:17.594-04:002017-05-31T10:11:17.594-04:00Thanks for your comment. I think comparing M7 thre...Thanks for your comment. I think comparing M7 threads to E5 cores is still a good comparison because:<br /><br />1. It shows that M7 still have a single threaded performance deficit. There is a class of applications/tasks that care about it.<br /><br />2. If I have a query running with a certain DOP on Intel and run the same query with the same DOP on M7 I shouldn't be surprised if I get much slower performance (assuming CPU-bound case). Hopefully this blog post will prevent some other people from having a "bad surprise". DOPs will likely have to be increased for such queries (potentially bringing more concurrency issues?).<br /><br />3. Would you rather have M7 with 4 threads per core where each thread is twice as fast?<br /><br />4. Why can't we compare 22 E5 cores vs 22 M7 cores where both are running 22 threads? Well that's because such a case renders M7 useless ;-) All the while it's a valid use case for Intel -- Azure runs with HT disabled, for instance. Food for thought.Alex Fatkulinhttps://www.blogger.com/profile/06361288475877100451noreply@blogger.comtag:blogger.com,1999:blog-15440695.post-67524917894268393562017-05-31T08:54:58.348-04:002017-05-31T08:54:58.348-04:00Hi Alex,
I can confirm that single-threaded perfor...Hi Alex,<br />I can confirm that single-threaded performance of SPARC M7 core is worse than Intel x86-64 core<br />it was bad surprise that hash joins perform longer after moving from x86-64 to SuperCluster....<br />BTW:<br />good comparison would be to compare core-by-core performance - actually price based performance if we consider core factor table.<br />so <br />22 cores(44 threads) of E5-2699V4<br />vs<br />22 cores(176 threads) of SPARC M7<br />?odenysenkohttps://www.blogger.com/profile/13962845912456085125noreply@blogger.comtag:blogger.com,1999:blog-15440695.post-91615094612164639752017-05-25T10:36:07.500-04:002017-05-25T10:36:07.500-04:00This was a LIO test so the entire test was done in...This was a LIO test so the entire test was done in-memory (buffer cache).Alex Fatkulinhttps://www.blogger.com/profile/06361288475877100451noreply@blogger.comtag:blogger.com,1999:blog-15440695.post-62661316867950448522017-05-25T09:36:46.213-04:002017-05-25T09:36:46.213-04:00Hello, Alex !
What the storage equipment did you ...Hello, Alex !<br /><br />What the storage equipment did you use for the tests ? Thank youpanzer.hierhttps://www.blogger.com/profile/16150395032367802483noreply@blogger.comtag:blogger.com,1999:blog-15440695.post-91113272829468362592017-05-24T09:42:42.081-04:002017-05-24T09:42:42.081-04:00Well, I guess I didn't understand what you'...Well, I guess I didn't understand what you're trying to prove here? If you are trying to compare SPARC M7 to Intel's E5-2699 V4, I would think you'd want to compare both architectures at its limits? And of course, licensing, licensing costs most likely will be the biggest consideration in comparing these two and as licensing is mostly per core based, comparing performance core vs core most likely what most readers would be interested in. Im certainly interested ;-)Anonymoushttps://www.blogger.com/profile/01548704524448863996noreply@blogger.comtag:blogger.com,1999:blog-15440695.post-31759179698511725882017-05-24T09:36:50.125-04:002017-05-24T09:36:50.125-04:00Hello Phil,
thank you for your comment. The way y...Hello Phil,<br /><br />thank you for your comment. The way you posed the question seems to indicate that you find this comparison unfair. Which by the way I do too :) Assuming we agree on this I also think that comparing a chip with 256 threads to a chip with 44 threads in a way that benefits highly threaded architecture is unfair as well.Alex Fatkulinhttps://www.blogger.com/profile/06361288475877100451noreply@blogger.comtag:blogger.com,1999:blog-15440695.post-86080007341366329082017-05-24T09:11:48.214-04:002017-05-24T09:11:48.214-04:00Why are you comparing SPARC M7 thread performance ...Why are you comparing SPARC M7 thread performance to Intel E5-2699 V4 core performance? A single SPARC M7 chip has 32 x cores and 256 x threads versus Intel E5-2699 V4 @ 22 x cores, 44-threads, almost 6x fewer threads! Why not compare core to core performance or atleast run 88-threads to max out the Intel 2-socket box to see how the SPARC M7 would compare?Anonymoushttps://www.blogger.com/profile/01548704524448863996noreply@blogger.com