vApusMark II Response Time

Each tile in vApusMark II demands 18 virtual CPUs: four for the Oracle OLTP test, eight for the MS SQL Server OLAP test, and six for the three web application VMs (two CPUs each). Therefore, a four tile test will require 72 virtual CPUs. A quad Xeon E7-4870 contains 40 cores and 80 threads with Hyper-Threading enabled. With a test that puts 72 virtual CPUs to work, you cannot measure the total throughput of the quad Xeon E7. In fact, some of those 72 virtual CPUs are not working at 100% all of the time. For example, the CPU load caused by the web VMs shows a lot of spikes. Thus, we can not interprete the throughput numbers without a look at the response times.

vApus Mark II Response time

Back to our benchmark or throughput scores. Ideally, we should measure throughput at exactly the same response times. But with our current stress testing software, trying to keep response time the same would be an extremely time consuming process.

vApus Mark II score revisited

Since the quad Opteron shows a 40% increase in response time from 4 to 5 tiles (or from 20 to 25 VMs), we believe that the four tile score (149) is more representative of the "real performance". The extra throughput that the five tile test delivers comes at a response time price that is too high.

The response time of the Quad Xeon 7560 increases 9% when we try to load it with five extra VMs. In this case, the "real and fair" throughput score is a little bit harder to determine. It is somewhere between the score of 4 tiles and 5 tiles, probably around 180 or so.

In case of the Quad Xeon E7, however, things are crystal clear. Running 20 or 25 VMs does not make any difference: the response times stay in the same league. In this case we take the highest score to be the real one.

So if we take response times into account, the quad E7-4870 is about 35% faster than its predecessor (243 vs 180) and about 63% faster than the AMD system in our test (243 vs 149). AMD's fastest processor is the 2.5GHz 6180SE now. This CPU is clocked around 13% higher and should thus be able to reach a score of around 168. That means the Xeon E7-4870 should still have a 44% (or more) advantage over its nearest but much cheaper competitor in this particular workload.

Virtual Performance on vSphere 4 Power Extremes


View All Comments

  • Fallen Kell - Thursday, May 19, 2011 - link

    As the subject says. Would love to see how these deal with something like Linpack or similar. Reply
  • erple2 - Thursday, May 19, 2011 - link

    I'd be more interested at seeing how they perform in slightly more "generic" and non-GPU optimizeable workloads. If I'm running Linpack or other FPU operations, particularly those that parallelize exceptionally well, I'd rather invest time and money into developing algorithms that run on a GPU than a fast CPU. The returns for that work are generally astounding.

    Now, that's not to say that General Purpose problems work well on a GPU (and I understand that). However, I'm not sure that measuring the "speed" of a single processor (or even a massively parallelized load) would tell you much, other than "it's pretty fast, but if you can massively parallelize a computational workload, figure out how to do it on a commodity GPU, and blow through it at orders of magnitude faster than any CPU can do it".

    However, I can't see running any virtualization work on a GPU anytime soon!
  • stephenbrooks - Thursday, May 19, 2011 - link

    Yeah, well, in an ideal world...

    But sometimes (actually, every single time in my experience) the "expensive software" that's been bought to run on these servers lacks a GPU option. I'm thinking of electromagnetic or finite element analysis code.

    Finite element engines are the sort of thing that companies make a lot of money selling. They are complicated. The commercial ones probably have >10 programmer-years of work in them, and even if they weren't fiercely-protected closed source, porting and re-optimising for a GPU would be additional years work requiring programmers again at a high level and with a lot of mathematical expertise.

    (There might be some decent open-source alternatives around, but they lack the front ends and GUI that most engineers are comfortable using.)

    If you think fixing the above issues are "easy", go ahead. You'll make millions.
  • L. - Thursday, May 19, 2011 - link


    if you code .. i don't want to read your code
  • carnachion - Friday, May 20, 2011 - link

    I agree with you. In my experience GPU computing for scientific applications are still in it's infancy, and in some cases the performance gains are not so high.

    There's still a big performance penalty by using double precision for the calculations. In my lab we are porting some programs to GPU, we started using a matrix multiplication library that uses GPU in a GTX590. Using one of the 590's GPU it was 2x faster than a Phenon X6 1100T, and using both GPUs it was 3.5x faster. So not that huge gain, using a Magny-Cours processor we could reach the performance of a single GPU, but of course at a higher price.

    Usually scientific applications can use hundreds of cores, and they are tunned to get a good scaling. But I don't know how GPU calculations scales with the number of GPUs, from 1 to 2 GPUs we got this 75% boost, but how it will perform using inter-node communication, even with a Infiniband connection I don't know if there'll be a bottleneck for real world applications. So that's why people still invest in thousands of cores computers, GPU still need a lot of work to be a real competitor.
  • DanNeely - Saturday, May 21, 2011 - link

    single vs double precision isn't the only limiting factor for GPU computing. The amount of data you can have in cache per thread is far smaller than on a traditional CPU. If your working set is too big to fit into the tiny amount of cache available performance is going to nose dive. This is farther aggravated by the fact that GPU memory systems are heavily optimized for streaming access and that random IO (like cache misses) suffers in performance.

    The result is that some applications which can be written to fit the GPU model very well will see enormous performance increases vs CPU equivalents. Others will get essentially nothing.

    Einstein @ Home's gravitational wave search app is an example of the latter. The calculations are inherently very random in memory access (to the extent that it benefits by about 10% from triple channel memory on intel quads; Intel's said that for quads there shouldn't be any real world app benefit from the 3rd channel). A few years ago when they launched cuda, nVidia worked with several large projects on the BOINC platform to try and port their apps to CUDA. The E@H cuda app ended up no faster than the CPU app and didn't scale at all with more cuda cores since all they did was to increase the number of threads stalled on memory IO.
  • Marburg U - Thursday, May 19, 2011 - link

    Finally something juicy, Reply
  • JarredWalton - Thursday, May 19, 2011 - link

    So, just curious: is this spam (but no links to a separate site), or some commentary that didn't really say anything? All I've got is this, "On the nature of things":

    Maybe I missed what he's getting at, or maybe he's just saying "Westmere-EX rocks!"
  • bobbozzo - Monday, May 23, 2011 - link

    Jarred, my guess is that it is spam, and that there was a link or some HTML posted which was filtered out by the comments system.

  • lol123 - Thursday, May 19, 2011 - link

    Why is there a 2 socket only line of E7 (E7-28xx), but at least as far as I can tell, not any 2-socket motherboards or servers? Are those simply not available yet? Reply

Log in

Don't have an account? Sign up now