Difference between revisions of "Performance"

From Gw-qcd-wiki
Jump to: navigation, search
Line 1: Line 1:
 +
'''Ninja (K20)'''
 +
 +
* Tester: Andrei Alexandru
 +
* Test date: 19 Feb 2013
 +
* Commit: 4c3956dd3075f49e7bd493a813e8b922a4377877
 +
* Hardware: K20
 +
* CUDA version 5.0
 +
 +
{| class="wikitable"
 +
!Kernel 
 +
!Configuration
 +
!Bandwidth
 +
!FLOPs
 +
|-
 +
|rowspan="2"|Dslash_cuda
 +
|Dslash (24^4)         
 +
|NA GB/s 
 +
|NA GFLOP/s
 +
|-
 +
|hopping (24^4)       
 +
|170.6 GB/s
 +
|78.2 GFLOP/s
 +
|-
 +
|rowspan="3"|Dslash_multi_gpu (double)
 +
|1 node, 24^4 Dslash 
 +
|170.8 GB/s 
 +
|76.1 GFLOP/s
 +
|-
 +
|2 nodes, 24^4 Dslash
 +
|NA GB/s
 +
|NA GFLOP/s
 +
|-
 +
|4 nodes, 24^4 Dslash 
 +
|NA GB/s
 +
|NA GFLOP/s
 +
|-
 +
|rowspan="3"|Dslash_multi_gpu (single)
 +
|1 node, 24^4 Dslash 
 +
|163.5 GB/s 
 +
|145.6 GFLOP/s
 +
|-
 +
|2 nodes, 24^4 Dslash
 +
|NA GB/s
 +
|NA GFLOP/s
 +
|-
 +
|4 nodes, 24^4 Dslash 
 +
|NA GB/s
 +
|NA GFLOP/s
 +
|-
 +
|rowspan="3"|Vector utilities
 +
|Addition     
 +
|162.78 GB/s 
 +
|6.78 GFLOP/s
 +
|-
 +
|Dot product   
 +
|108.89 GB/s 
 +
|27.22
 +
|-
 +
|Copy   
 +
|125.27 GB/s
 +
|N/A
 +
|-
 +
|}
 +
 +
 
'''Lehman (Fermi)'''
 
'''Lehman (Fermi)'''
  

Revision as of 23:49, 19 February 2013

Ninja (K20)

  • Tester: Andrei Alexandru
  • Test date: 19 Feb 2013
  • Commit: 4c3956dd3075f49e7bd493a813e8b922a4377877
  • Hardware: K20
  • CUDA version 5.0
Kernel Configuration Bandwidth FLOPs
Dslash_cuda Dslash (24^4) NA GB/s NA GFLOP/s
hopping (24^4) 170.6 GB/s 78.2 GFLOP/s
Dslash_multi_gpu (double) 1 node, 24^4 Dslash 170.8 GB/s 76.1 GFLOP/s
2 nodes, 24^4 Dslash NA GB/s NA GFLOP/s
4 nodes, 24^4 Dslash NA GB/s NA GFLOP/s
Dslash_multi_gpu (single) 1 node, 24^4 Dslash 163.5 GB/s 145.6 GFLOP/s
2 nodes, 24^4 Dslash NA GB/s NA GFLOP/s
4 nodes, 24^4 Dslash NA GB/s NA GFLOP/s
Vector utilities Addition 162.78 GB/s 6.78 GFLOP/s
Dot product 108.89 GB/s 27.22
Copy 125.27 GB/s N/A


Lehman (Fermi)

  • Tester: Andrei Alexandru
  • Test date: 20 Aug 2011
  • Commit: b54a3437eeeebf773271d7a5424b41949f9283ad
  • Hardware: gtx580
  • CUDA version 4.0
Kernel Configuration Bandwidth FLOPs
Dslash_cuda Dslash (24^4) 100 GB/s 45 GFLOP/s
hopping (24^4) 114 GB/s 53 GFLOP/s
Dslash_multi_gpu (double) 1 node, 24^4 Dslash 128 GB/s 57 GFLOP/s
2 nodes, 24^4 Dslash 240 GB/s 107 GFLOP/s
4 nodes, 24^4 Dslash 391 GB/s 174 GFLOP/s
Dslash_multi_gpu (single) 1 node, 24^4 Dslash 165 GB/s 147 GFLOP/s
2 nodes, 24^4 Dslash 302 GB/s 269 GFLOP/s
4 nodes, 24^4 Dslash 402 GB/s 356 GFLOP/s
Vector utilities Addition 159 GB/s 6.6 GFLOP/s
Dot product 129 GB/s N/A
Copy 155 GB/s N/A

Carver (Fermi with ECC)

  • Tester: Ben Gamari
  • Test date: 14 Jul 2010
  • Commit: e3e4ffafd158abd004c483694a27f4f6bc7d2185
  • Hardware: Tesla C2050 (ECC on)
  • CUDA version 3.0
Kernel Configuration Bandwidth FLOPs
Dslash_cuda Dslash (24^4) 73 GB/s 32 GFLOP/s
hopping (24^4) 74 GB/s 34 GFLOP/s
Dslash_multi_gpu (double) 1 node, 24^4 Dslash 79 GB/s 35 GFLOP/s
2 nodes, 24^4 Dslash 145 GB/s 64 GFLOP/s
4 nodes, 24^4 Dslash 256 GB/s 114 GFLOP/s
Dslash_multi_gpu (single) 1 node, 24^4 Dslash 79 GB/s 76 GFLOP/s
2 nodes, 24^4 Dslash 156 GB/s 140 GFLOP/s
4 nodes, 24^4 Dslash 283 GB/s 252 GFLOP/s
Vector utilities Addition 82 GB/s 3.4 GFLOP/s
Dot product 88 GB/s N/A
Copy 84 GB/s N/A


Carver (Fermi without ECC)

  • Tester: Ben Gamari
  • Test date: 14 Jul 2010
  • Commit: e3e4ffafd158abd004c483694a27f4f6bc7d2185
  • Hardware: Tesla C2050 (ECC off)
  • CUDA version 3.0
Kernel Configuration Bandwidth FLOPs
Dslash_cuda Dslash (24^4) 85 GB/s 38 GFLOP/s
hopping (24^4) 86 GB/s 39 GFLOP/s
Dslash_multi_gpu (double) 1 node, 24^4 Dslash 96 GB/s 43 GFLOP/s
2 nodes, 24^4 Dslash 179 GB/s 82 GFLOP/s
4 nodes, 24^4 Dslash 304 GB/s 139 GFLOP/s
Dslash_multi_gpu (single) 1 node, 24^4 Dslash 114 GB/s 101 GFLOP/s
2 nodes, 24^4 Dslash 210 GB/s 187 GFLOP/s
4 nodes, 24^4 Dslash 279 GB/s 256 GFLOP/s
Vector utilities Addition 110 GB/s 4.6 GFLOP/s
Dot product 119 GB/s N/A
Copy 114 GB/s N/A


Carver (Tesla)

  • Tester: Ben Gamari
  • Test date: 14 Jul 2010
  • Commit: e3e4ffafd158abd004c483694a27f4f6bc7d2185
  • Hardware: Tesla C1060
  • CUDA version 3.0
Kernel Configuration Bandwidth FLOPs
Dslash_cuda Dslash (24^4) 69 GB/s 31 GFLOP/s
hopping (24^4) 68 GB/s 31 GFLOP/s
Dslash_multi_gpu (double) 1 node, 24^4 Dslash 71 GB/s 32 GFLOP/s
2 nodes, 24^4 Dslash 134 GB/s 60 GFLOP/s
4 nodes, 24^4 Dslash 240 GB/s 107 GFLOP/s
Dslash_multi_gpu (single) 1 node, 24^4 Dslash 65 GB/s 58 GFLOP/s
2 nodes, 24^4 Dslash 122 GB/s 112 GFLOP/s
4 nodes, 24^4 Dslash 224 GB/s 199 GFLOP/s
Vector utilities Addition 83 GB/s 3.5 GFLOP/s
Dot product 81 GB/s N/A
Copy 83 GB/s N/A


jlab (GF100)

  • Tester: Ben Gamari
  • Test date: 14 Jul 2010
  • Commit: e3e4ffafd158abd004c483694a27f4f6bc7d2185
  • Hardware: GeForce GTX 480
  • CUDA version 3.1
Kernel Configuration Bandwidth FLOPs
Dslash_cuda Dslash (24^4) 102 GB/s 46 GFLOP/s
hopping (24^4) 105 GB/s 48 GFLOP/s
Dslash_multi_gpu (double) 1 node, 24^4 Dslash 113 GB/s 51 GFLOP/s
Dslash_multi_gpu (single) 1 node, 24^4 Dslash 140 GB/s 125 GFLOP/s
Vector utilities Addition 136 GB/s 5.65 GFLOP/s
Dot product 136 GB/s N/A
Copy 139 GB/s N/A

Samurai (GT200)

  • Tester: Ben Gamari
  • Test date: 14 Jul 2010
  • Commit: e4b3a766ec873f57cc7ef31b4167bc2f032d418e
  • Hardware: GeForce GTX 280
  • CUDA version 3.1
Kernel Configuration Bandwidth FLOPs
Dslash_cuda Dslash (24^4) 76 GB/s 34 GFLOP/s
hopping (24^4) 75 GB/s 34 GFLOP/s
Dslash_multi_gpu (double) 1 node, 24^4 Dslash 77 GB/s 34 GFLOP/s
Dslash_multi_gpu (single) 1 node, 24^4 Dslash 91 GB/s 81 GFLOP/s
Vector utilities Addition 113 GB/s 4.7 GFLOP/s
Dot product 77 GB/s N/A
Copy 121 GB/s N/A
Retrieved from "http://samurai.phys.gwu.edu/wiki/index.php?title=Performance&oldid=559"