With the launch of Kaveri, some people have been wondering if the platform is suitable for HPC applications.  Floating point peak performance of the CPU and GPU  on both fp32 and fp64 datatypes is one of the considerations. At launch time, we were not clear on the fp64 performance of Kaveri's GPU but now we have official confirmation from AMD that it is 1/16th the rate of fp32 (similar to most GCN based GPUs except the flagships) and we have verified this on our 7850K by running FlopsCL.  

I am taking this opportunity to summarize the info about Kaveri, Trinity, Llano and Intel's competing platforms Haswell and Ivy Bridge on both the CPU and GPU side. We provide a per-cycle estimate for the chips as well as peak calculated in gflops. The estimates are chip-wide, i.e. already take into account the number of cores or modules. Due to turbo boost, it was difficult to decide what frequency to use for peak calculations. For CPUs, we are using the base frequency and for GPUs we are using the boost frequency because in multithreaded and/or heterogeneous scenarios the CPU is less likely to turbo. In any case, we believe our readers are smart enough to calculate peaks at any frequency they want, given that we already supply per-cycle peaks :)

The peak CPU performance will depend on the SIMD ISA that your code was written and compiled for. We consider three cases: SSE, AVX (without FMA) and AVX with FMA (either FMA3 or FMA4).

 

CPU floating-point peak performance
Platform Kaveri Trinity Llano Haswell Ivy Bridge
Chip 7850K 5800K 3870K 4770K 3770K
CPU frequency 3.7 GHz 3.8 GHz 3.0GHz 3.5GHz 3.5GHz
SSE fp32 (/cycle) 16 16 32 32 32
SSE fp64 (/cycle) 8 8 16 16 16
AVX fp32 (/cycle) 16 16 - 64 64
AVX fp64 (/cycle) 8 8 - 32 32
AVX FMA fp32 (/cycle) 32 32 - 128 -
AVX FMA fp64 (/cycle) 16 16 - 64 -
SSE fp32 (gflops) 59.2 60.8 96 112 112
SSE fp64 (gflops) 29.6 30.4 48 56 56
AVX fp32 (gflops) 59.2 60.8 - 224 224
AVX fp64 (gflops) 29.6 30.4 - 112 112
AVX FMA fp32 (gflops) 118.4 121.6 - 448 -
AVX FMA fp64 (gflops) 59.2 60.8 - 224 -

It is no secret that AMD's Bulldozer family cores (Steamroller in Kaveri and Piledriver in Trinity) are no match for recent Intel cores in FP performance due to the shared FP unit in each module. As a comparison point, one core in Haswell has the same floating point performance per cycle as two modules (or four cores) in Steamroller.

Now onto GPU peaks. Here, for Haswell, we chose to include both GT2 and GT3e variants.

Platform Kaveri Trinity Llano Haswell GT3e Haswell GT2 Ivy Bridge
GPU floating-point peak performance
Chip 7850K 5800K 3870K 4770R 4770K 3770K
GPU frequency 720 MHz 800 MHz 600 MHz 1.3 GHz 1.25 GHz 1.15 GHz
fp32/cycle 1024 768 800 640 320 256

fp64/cycle (OpenCL)

64 48** 0 0 0 0

fp64/cycle (Direct3D)

64 0? 0 160 80 64
fp32 gflops 737.3 614 480 832 400 294.4

fp64 gflops (OpenCL)

46.1 38.4** 0 0 0 0

fp64 gflops (Direct3D)

46.1 0? 0 208 100 73.6

The fp64 support situation is a bit of a mess because some GPUs only support fp64 under some APIs.  The fp64 rate of Intel's GPUs does not appear to be published but David Kanter provides an estimate of 1/4 speed compared to fp32. However Intel only enables fp64 under DirectCompute but does not enable fp64 under OpenCL for any of its GPUs.

Situation on AMD's Trinity/Richland is even more complicated. fp64 support under OpenCL is not standards-compliant and depends upon using a proprietary extension (cl_amd_fp64). Trinity/Richland do not appear to support fp64 under DirectCompute (and MS C++ AMP implementation) from what I can tell. From an API standapoint, Kaveri's GCN GPUs should work fine on for fp64 under all APIs.

Some of you might be wondering whether Kaveri is good for HPC applications. Compared to discrete GPUs, applications that are already ported and work well on discrete GPUs will continue to be best run on discrete GPUs.  However, Kaveri and HSA will enable many more applications  to be GPU accelerated. 

Now we compare Kaveri against Haswell. In applications depending upon fp64 performance, conditions are not generally favorable to Kaveri. Kaveri's fp64 peak including both the CPU and GPU is only about 110 gflops.  You will generally be better off first optimizing your code for AVX and FMA instructions and running on Haswell's CPU cores. If you are using Windows 8,  you might also want to explore using Iris Pro through C++ AMP in conjunction with the CPU. Overall I doubt we will see Kaveri being used for fp64 workloads.

For heterogeneous fp32 applications, Kaveri should outperform Haswell GT2 and Ivy Bridge.  Haswell GT3e will again be a strong contender on Windows given the extremely capable Haswell CPU cores and Iris Pro graphics.  Intel's GPUs  do not currently support OpenCL under Linux, but a driver is being worked on.  Thus, on Linux, Kaveri will simply win out on fp32 heterogeneous applications. However, even on Windows Haswell GT3e will get strong competiton from Kaveri.  While AMD has advantages such as excellent GCN architecture and HSA software stack (when ready) enabling many more applications to take advantage of GPU, Iris Pro will have the eDRAM to potentially provide much improved bandwidth and the backing of strong CPU cores.

I hope I have provided a fair overview of the FP capabilities of each platform. Application performance will of course depend on many more factors. Your questions and comments are welcome.

Comments Locked

101 Comments

View All Comments

  • sanaris - Saturday, March 1, 2014 - link

    I have card with "unbound" DP performance. It is complete brick. It says it should get 400 gigaflops, but in reality It does Prime95 about 24 msec/iter. When Opteron 110W chip does it twice faster - about 12 msec.

    All AMD GPU efforts are turned into bricks cause they fail to test their designs with real software.

    Very bad AMD does not move into 32 core opteron chips cause that is what I need now.
  • HalloweenJack - Wednesday, January 22, 2014 - link

    Iris Pro - FCBGA ONLY and OEM.... so why are you testing it anyway???
  • MrSpadge - Wednesday, January 22, 2014 - link

    Because it exists. Don't shout at AT for listing it but rather at Intel for not giving it to us.
  • HalloweenJack - Wednesday, January 22, 2014 - link

    no , its pointless from AT for testing it - might as well get a 16 core G34 opteron for multithread as an `oooh shiny`...
  • DigitalFreak - Wednesday, January 22, 2014 - link

    Why are you AMD fanboys always so butthurt?
  • wumpus - Saturday, February 8, 2014 - link

    To show how badly 8 floating point cores running at half speed will do? I think we know that already.
  • BMNify - Wednesday, January 22, 2014 - link

    you bitching because you cant afford it so dont want it tested or what ?

    as already said it exists https://www.system76.com/laptops/model/galu1
    its been benched for both windows and linux os http://www.phoronix.com/scan.php?page=article&...

    by all accounts its ok, i wish Intel would OC put it's followup on their mainstream mid/high i somethings and also improve its data throughput compared to that above linked test, we shall see when it arrives or not...
  • michael2k - Wednesday, January 22, 2014 - link

    Can you compare Kaveri to the Bay Trail parts? The J2850 is a quad core part, though only 2.41GHz. It appears that the BT parts might be more congruent, if weaker GPU wise, in terms of CPU perf:
    http://hothardware.com/Reviews/Betting-On-Bay-Trai...
  • rahulgarg - Wednesday, January 22, 2014 - link

    That review has major errors. The AMD APU they are testing (A4-5000) is not Kaveri at all even they keep calling it Kaveri. A4-5000 is actually the low-end Kabini. Kaveri is MUCH faster than Bay Trail.
  • BMNify - Wednesday, January 22, 2014 - link

    oc Bay Trail even the quad has been crippled as it does NOT have AVX/AVX2 SIMD only at best SSE4 / SSE4.1 + SSE4.2 / Streaming SIMD Extensions 4

Log in

Don't have an account? Sign up now