[Split] PowerSpectrum Unit Test

[PatrickV2]

Cheers & No worries Patrick,
     Just wasn't sure extending the test was going to be needed.  Naked eye judgement is plenty for the purposes of testing scientific repeatability here, and running multiple times in the same exe would make it one large test rather than several small ones for comparison (if that makes any sense).  I'm happy that the 8800 seems to have some headroom left, and the 'Min' numbers indicate the sloest kernels have received a niice boost. 

Win7(WDDM) & XP(XPDM) driver model performance difference is 'gone' 

Thanks for the extended explanation; my remark was merely given in by curiosity (and probably a large lack in understanding the underlying higher goals), but I feel more enlightened now.

Regards, Patrick.

[Miep]

Device: Quadro FX 570M, 950 MHz clock, 242 MB memory.
Compute capability 1.1
Compiled with CUDA 3020.
                PowerSpectrum+summax Unit test #10 (FFT pipeline throughput)
Stock:
  Processing... Done!
  Compute Thoughput GFlops Avg(    9.58) Peak(   13.91) Min(    2.48) [OK]
   Memory thoughput GB/s   Avg(    5.70) Peak(    9.09) Min(    3.53)


Opt1 (worst case): 64 thrds/block, 2 x 524288 element streams
  revert to single stream from size 128
  Processing... Done!
  Compute thoughput [GFlops] -
      Avg(   11.23, 1.17x) Peak(   15.13, 1.09x) Min(    4.27, 1.72x) [OK]
   Memory thoughput [GB/s]   -
      Avg(    6.99, 1.23x) Peak(    9.88, 1.09x) Min(    5.01, 1.42x)

values roughly +- .3 on stock and +- .1 on opt1

[edit]compute speedup 1.56x - 1.76x memory speedup 1.22x -1.47x

[Jason G]

Thanks for the extended explanation; my remark was merely given in by curiosity (and probably a large lack in understanding the underlying higher goals), but I feel more enlightened now.

Yeah, a bit more info along those lines, the actual kernels under test run in timing loops set to roughly half a second, which is enough for ~thousands to millions of runs, so I was expecting 'fair' stability in the Avg, Peak & Min values, so we are alright for discrete kernel performance measurements. 

I have however picked up an interesting thing on a friends i7-860 w/GTX480 in comparing against mine ( 45nM core2 w/GTX480)
-  His Peaks & Averages are ~same as mine for the same clockrate ... BUT ... the 'Min (slowest kernels) are several times faster ... Better CPU & RAM does have significant impact on the running of the toughest parts of code, it seems

Jason

[Jason G]

values roughly +- .3 on stock and +- .1 on opt1

Hey that's decent! ... and there you were going to start a riot when initial mods yielded about 5% slowdown on yours ... tsk tsk tsk 

[Miep]

Hey that's decent! ... and there you were going to start a riot when initial mods yielded about 5% slowdown on yours ... tsk tsk tsk 

Oh I just learned how to complain when not suffering did the trick didn't it?

[Claggy]

My 9800GTX+ on Win 7 x64:

Device: GeForce 9800 GTX/9800 GTX+, 1900 MHz clock, 496 MB memory.
Compute capability 1.1
Compiled with CUDA 3020.
                PowerSpectrum+summax Unit test #10 (FFT pipeline throughput)
Stock:
  Processing... Done!
  Compute Thoughput GFlops Avg(   49.81) Peak(   71.73) Min(    8.11) [OK]
   Memory thoughput GB/s   Avg(   29.08) Peak(   44.80) Min(   14.31)


Opt1 (worst case): 64 thrds/block, 2 x 524288 element streams
  revert to single stream from size 128
  Processing... Done!
  Compute thoughput [GFlops] -
      Avg(   57.66, 1.16x) Peak(   80.19, 1.12x) Min(   18.07, 2.23x) [OK]
   Memory thoughput [GB/s]   -
      Avg(   35.80, 1.23x) Peak(   50.46, 1.13x) Min(   24.47, 1.71x)

Claggy

[glennaxl]

Device: GeForce GTX 260, 1441 MHz clock, 869 MB memory.
Compute capability 1.3
Compiled with CUDA 3020.
                PowerSpectrum+summax Unit test #10 (FFT pipeline throughput)
Stock:
  Processing... Done!
  Compute Thoughput GFlops Avg(   47.55) Peak(   65.20) Min(   10.16) [OK]
   Memory thoughput GB/s   Avg(   28.09) Peak(   37.12) Min(   17.92)


Opt1 (worst case): 128 thrds/block, 2 x 524288 element streams
  revert to single stream from size 256
  Processing... Done!
  Compute thoughput [GFlops] -
      Avg(   84.83, 1.78x) Peak(  111.50, 1.71x) Min(   31.57, 3.11x) [OK]
   Memory thoughput [GB/s]   -
      Avg(   52.63, 1.87x) Peak(   67.26, 1.81x) Min(   36.24, 2.02x)

[Jason G]

Thanks both!

@glenaxl: that's some impressive speedup on GTX 260, I'll have to look at that here carefully on mine when I get a chance to do so.

@Claggy, average at 3/4 of peak seems pretty good, but I think we can get some more maybe.

@ALL, Thanks! I'm closing this test for now.  It's been an extremely valuable contribution from you all that has had a huge impact on the pace & quality of our progress (mine in particular). 

FYI: Some urgent issues may have come to light from Raistmer's OpenCL development when combined with the refinements here.  Those will need some fairly close attention for a short while, to get some information back to Berkeley, but stay tuned as there are more tests to come 

[Locking thread, Please stay tuned for further Unit Tests!]
Jason

[Jason G]

@All:
   Just a note that the concerns that arose, and distracted me from testing & development along this line, have now been at least partially resolved, and don't require any immediate action on our part.   I'm back to ruggedising &  integrating what we've accomplished here into the X-builds, and plan to start devising tests for PoT (Power over Time) processing refinement soon, in similar fashion to this thread.  PoT processing covers Gaussian searches, Triplet & Pulse finding, for which all Cuda releases have known issues to address, so there'll be plenty of tests to devise & collect data for yet.

Cheers once again! 
Jason