20171107, 17:57  #1 
Sep 2003
3^{2}·7·41 Posts 
c5 instances are now available
https://aws.amazon.com/blogs/aws/now...oramazonec2/
They run on 3.0 GHz Intel Xeon Platinum 8000series, which has AVX512. Amazon claims 25% price/performance improvement over c4. Many technical details will be provided at AWS re:Invent at the end of this month. They are not available yet in useast2 (Ohio), which usually has the cheapest spot prices. 
20171107, 18:00  #2 
"Serge"
Mar 2008
Phi(4,2^7658614+1)/2
22165_{8} Posts 
Cool! Just 1 year has passed since they announced that they will be deploying these "soon"  and there! The naysayers were shamed.

20171107, 18:29  #3 
"/X\(‘‘)/X\"
Jan 2013
29·101 Posts 
I'm benchmarking a c5.large, c5.xlarge, c5.2xlarge, and a c5.18xlarge now.
The mprime 29.4 isn't setting affinities properly on the c5.18xlarge. 
20171107, 19:08  #4 
"/X\(‘‘)/X\"
Jan 2013
29×101 Posts 
c5.large:
Seems to prefer hyperthreading [Work thread Nov 7 18:13] Timing 2048K allcomplex FFT, 1 core, 1 worker. Average times: 10.26 ms. Total throughput: 97.46 iter/sec. [Work thread Nov 7 18:13] Timing 2048K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 10.04 ms. Total throughput: 99.62 iter/sec. [Work thread Nov 7 18:13] Timing 2048K allcomplex FFT, 1 core, 1 worker. Average times: 9.13 ms. Total throughput: 109.47 iter/sec. [Work thread Nov 7 18:14] Timing 2048K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 9.44 ms. Total throughput: 105.98 iter/sec. [Work thread Nov 7 18:14] Timing 2048K allcomplex FFT, 1 core, 1 worker. Average times: 9.25 ms. Total throughput: 108.07 iter/sec. [Work thread Nov 7 18:14] Timing 2048K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 9.82 ms. Total throughput: 101.88 iter/sec. [Work thread Nov 7 18:14] Timing 2048K allcomplex FFT, 1 core, 1 worker. Average times: 9.19 ms. Total throughput: 108.84 iter/sec. [Work thread Nov 7 18:15] Timing 2048K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 8.93 ms. Total throughput: 111.97 iter/sec. [Work thread Nov 7 18:15] Timing 2048K allcomplex FFT, 1 core, 1 worker. Average times: 8.99 ms. Total throughput: 111.27 iter/sec. [Work thread Nov 7 18:15] Timing 2048K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 8.50 ms. Total throughput: 117.58 iter/sec. [Work thread Nov 7 18:16] Timing 2048K allcomplex FFT, 1 core, 1 worker. Average times: 8.94 ms. Total throughput: 111.81 iter/sec. [Work thread Nov 7 18:16] Timing 2048K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 9.05 ms. Total throughput: 110.49 iter/sec. [Work thread Nov 7 18:16] Timing 2048K allcomplex FFT, 1 core, 1 worker. Average times: 9.09 ms. Total throughput: 109.98 iter/sec. [Work thread Nov 7 18:16] Timing 2048K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 9.28 ms. Total throughput: 107.71 iter/sec. [Work thread Nov 7 18:17] Timing 2048K allcomplex FFT, 1 core, 1 worker. Average times: 9.14 ms. Total throughput: 109.36 iter/sec. [Work thread Nov 7 18:17] Timing 2048K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 8.64 ms. Total throughput: 115.69 iter/sec. [Work thread Nov 7 18:17] Timing 2048K allcomplex FFT, 1 core, 1 worker. Average times: 9.30 ms. Total throughput: 107.58 iter/sec. [Work thread Nov 7 18:17] Timing 2048K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 8.87 ms. Total throughput: 112.70 iter/sec. [Work thread Nov 7 18:18] Timing 2048K allcomplex FFT, 1 core, 1 worker. Average times: 9.15 ms. Total throughput: 109.33 iter/sec. [Work thread Nov 7 18:18] Timing 2048K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 9.26 ms. Total throughput: 107.97 iter/sec. [Work thread Nov 7 18:18] Timing 2048K allcomplex FFT, 1 core, 1 worker. Average times: 9.01 ms. Total throughput: 111.00 iter/sec. [Work thread Nov 7 18:18] Timing 2048K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 9.11 ms. Total throughput: 109.78 iter/sec. [Work thread Nov 7 18:19] Timing 2048K allcomplex FFT, 1 core, 1 worker. Average times: 9.06 ms. Total throughput: 110.42 iter/sec. [Work thread Nov 7 18:19] Timing 2048K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 8.81 ms. Total throughput: 113.55 iter/sec. [Work thread Nov 7 18:19] Timing 2048K allcomplex FFT, 1 core, 1 worker. Average times: 9.61 ms. Total throughput: 104.05 iter/sec. [Work thread Nov 7 18:19] Timing 2048K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 10.78 ms. Total throughput: 92.74 iter/sec. [Work thread Nov 7 18:20] Timing 2048K allcomplex FFT, 1 core, 1 worker. Average times: 9.26 ms. Total throughput: 107.98 iter/sec. [Work thread Nov 7 18:20] Timing 2048K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 9.24 ms. Total throughput: 108.26 iter/sec. [Work thread Nov 7 18:20] Timing 2048K allcomplex FFT, 1 core, 1 worker. Average times: 9.54 ms. Total throughput: 104.85 iter/sec. [Work thread Nov 7 18:20] Timing 2048K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 9.21 ms. Total throughput: 108.60 iter/sec. [Work thread Nov 7 18:31] Timing 4096K allcomplex FFT, 1 core, 1 worker. Average times: 19.81 ms. Total throughput: 50.47 iter/sec. [Work thread Nov 7 18:31] Timing 4096K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 21.05 ms. Total throughput: 47.51 iter/sec. [Work thread Nov 7 18:31] Timing 4096K allcomplex FFT, 1 core, 1 worker. Average times: 19.59 ms. Total throughput: 51.05 iter/sec. [Work thread Nov 7 18:32] Timing 4096K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 19.70 ms. Total throughput: 50.77 iter/sec. [Work thread Nov 7 18:32] Timing 4096K allcomplex FFT, 1 core, 1 worker. Average times: 19.45 ms. Total throughput: 51.41 iter/sec. [Work thread Nov 7 18:32] Timing 4096K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 20.62 ms. Total throughput: 48.51 iter/sec. [Work thread Nov 7 18:33] Timing 4096K allcomplex FFT, 1 core, 1 worker. Average times: 18.46 ms. Total throughput: 54.17 iter/sec. [Work thread Nov 7 18:33] Timing 4096K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 19.62 ms. Total throughput: 50.97 iter/sec. [Work thread Nov 7 18:33] Timing 4096K allcomplex FFT, 1 core, 1 worker. Average times: 18.48 ms. Total throughput: 54.10 iter/sec. [Work thread Nov 7 18:33] Timing 4096K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 18.18 ms. Total throughput: 55.02 iter/sec. [Work thread Nov 7 18:34] Timing 4096K allcomplex FFT, 1 core, 1 worker. Average times: 18.75 ms. Total throughput: 53.34 iter/sec. [Work thread Nov 7 18:34] Timing 4096K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 18.89 ms. Total throughput: 52.93 iter/sec. [Work thread Nov 7 18:34] Timing 4096K allcomplex FFT, 1 core, 1 worker. Average times: 19.52 ms. Total throughput: 51.23 iter/sec. [Work thread Nov 7 18:34] Timing 4096K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 19.81 ms. Total throughput: 50.49 iter/sec. [Work thread Nov 7 18:35] Timing 4096K allcomplex FFT, 1 core, 1 worker. Average times: 19.31 ms. Total throughput: 51.78 iter/sec. [Work thread Nov 7 18:35] Timing 4096K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 18.53 ms. Total throughput: 53.97 iter/sec. [Work thread Nov 7 18:35] Timing 4096K allcomplex FFT, 1 core, 1 worker. Average times: 19.36 ms. Total throughput: 51.66 iter/sec. [Work thread Nov 7 18:35] Timing 4096K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 17.90 ms. Total throughput: 55.85 iter/sec. [Work thread Nov 7 18:39] Timing 8192K allcomplex FFT, 1 core, 1 worker. Average times: 39.01 ms. Total throughput: 25.64 iter/sec. [Work thread Nov 7 18:40] Timing 8192K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 43.20 ms. Total throughput: 23.15 iter/sec. [Work thread Nov 7 18:40] Timing 8192K allcomplex FFT, 1 core, 1 worker. Average times: 38.95 ms. Total throughput: 25.68 iter/sec. [Work thread Nov 7 18:40] Timing 8192K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 40.22 ms. Total throughput: 24.87 iter/sec. [Work thread Nov 7 18:41] Timing 8192K allcomplex FFT, 1 core, 1 worker. Average times: 39.43 ms. Total throughput: 25.36 iter/sec. [Work thread Nov 7 18:41] Timing 8192K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 41.40 ms. Total throughput: 24.16 iter/sec. [Work thread Nov 7 18:41] Timing 8192K allcomplex FFT, 1 core, 1 worker. Average times: 39.07 ms. Total throughput: 25.60 iter/sec. [Work thread Nov 7 18:41] Timing 8192K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 40.58 ms. Total throughput: 24.64 iter/sec. [Work thread Nov 7 18:42] Timing 8192K allcomplex FFT, 1 core, 1 worker. Average times: 38.44 ms. Total throughput: 26.01 iter/sec. [Work thread Nov 7 18:42] Timing 8192K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 38.43 ms. Total throughput: 26.02 iter/sec. [Work thread Nov 7 18:42] Timing 8192K allcomplex FFT, 1 core, 1 worker. Average times: 38.49 ms. Total throughput: 25.98 iter/sec. [Work thread Nov 7 18:42] Timing 8192K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 36.85 ms. Total throughput: 27.14 iter/sec. [Work thread Nov 7 18:43] Timing 8192K allcomplex FFT, 1 core, 1 worker. Average times: 41.35 ms. Total throughput: 24.19 iter/sec. [Work thread Nov 7 18:43] Timing 8192K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 44.96 ms. Total throughput: 22.24 iter/sec. [Work thread Nov 7 18:43] Timing 8192K allcomplex FFT, 1 core, 1 worker. Average times: 40.09 ms. Total throughput: 24.95 iter/sec. [Work thread Nov 7 18:44] Timing 8192K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 39.27 ms. Total throughput: 25.47 iter/sec. [Work thread Nov 7 18:44] Timing 8192K allcomplex FFT, 1 core, 1 worker. Average times: 40.61 ms. Total throughput: 24.62 iter/sec. [Work thread Nov 7 18:44] Timing 8192K allcomplex FFT, 1 core hyperthreaded, 1 worker. Average times: 38.18 ms. Total throughput: 26.19 iter/sec. 
20171107, 19:10  #5 
"/X\(‘‘)/X\"
Jan 2013
101101110001_{2} Posts 
c5.xlarge
Prefers hyperthreading with 2 workers. [Work thread Nov 7 18:17] Timing 2048K allcomplex FFT, 2 cores, 1 worker. Average times: 4.55 ms. Total throughput: 219.90 iter/sec. [Work thread Nov 7 18:17] Timing 2048K allcomplex FFT, 2 cores, 2 workers. Average times: 8.41, 8.41 ms. Total throughput: 237.75 iter/sec. [Work thread Nov 7 18:18] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 4.03 ms. Total throughput: 248.40 iter/sec. [Work thread Nov 7 18:18] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 8.08, 8.06 ms. Total throughput: 247.78 iter/sec. [Work thread Nov 7 18:18] Timing 2048K allcomplex FFT, 2 cores, 1 worker. Average times: 4.70 ms. Total throughput: 212.56 iter/sec. [Work thread Nov 7 18:18] Timing 2048K allcomplex FFT, 2 cores, 2 workers. Average times: 8.31, 8.31 ms. Total throughput: 240.76 iter/sec. [Work thread Nov 7 18:19] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 4.04 ms. Total throughput: 247.38 iter/sec. [Work thread Nov 7 18:19] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 8.11, 8.11 ms. Total throughput: 246.65 iter/sec. [Work thread Nov 7 18:19] Timing 2048K allcomplex FFT, 2 cores, 1 worker. Average times: 4.93 ms. Total throughput: 203.02 iter/sec. [Work thread Nov 7 18:20] Timing 2048K allcomplex FFT, 2 cores, 2 workers. Average times: 8.42, 8.42 ms. Total throughput: 237.45 iter/sec. [Work thread Nov 7 18:20] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 4.21 ms. Total throughput: 237.74 iter/sec. [Work thread Nov 7 18:20] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 8.38, 8.24 ms. Total throughput: 240.65 iter/sec. [Work thread Nov 7 18:20] Timing 2048K allcomplex FFT, 2 cores, 1 worker. Average times: 4.39 ms. Total throughput: 227.99 iter/sec. [Work thread Nov 7 18:21] Timing 2048K allcomplex FFT, 2 cores, 2 workers. Average times: 8.26, 8.26 ms. Total throughput: 242.15 iter/sec. [Work thread Nov 7 18:21] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 3.83 ms. Total throughput: 261.00 iter/sec. [Work thread Nov 7 18:21] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 7.75, 7.75 ms. Total throughput: 258.15 iter/sec. [Work thread Nov 7 18:21] Timing 2048K allcomplex FFT, 2 cores, 1 worker. Average times: 4.48 ms. Total throughput: 223.15 iter/sec. [Work thread Nov 7 18:22] Timing 2048K allcomplex FFT, 2 cores, 2 workers. Average times: 8.18, 8.18 ms. Total throughput: 244.52 iter/sec. [Work thread Nov 7 18:22] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 3.88 ms. Total throughput: 257.93 iter/sec. [Work thread Nov 7 18:22] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 7.68, 7.67 ms. Total throughput: 260.62 iter/sec. [Work thread Nov 7 18:22] Timing 2048K allcomplex FFT, 2 cores, 1 worker. Average times: 4.59 ms. Total throughput: 217.68 iter/sec. [Work thread Nov 7 18:23] Timing 2048K allcomplex FFT, 2 cores, 2 workers. Average times: 8.16, 8.16 ms. Total throughput: 245.24 iter/sec. [Work thread Nov 7 18:23] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 3.99 ms. Total throughput: 250.89 iter/sec. [Work thread Nov 7 18:23] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 7.72, 7.72 ms. Total throughput: 259.04 iter/sec. [Work thread Nov 7 18:23] Timing 2048K allcomplex FFT, 2 cores, 1 worker. Average times: 4.48 ms. Total throughput: 223.16 iter/sec. [Work thread Nov 7 18:24] Timing 2048K allcomplex FFT, 2 cores, 2 workers. Average times: 8.66, 8.63 ms. Total throughput: 231.41 iter/sec. [Work thread Nov 7 18:24] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 4.01 ms. Total throughput: 249.52 iter/sec. [Work thread Nov 7 18:24] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 8.04, 8.03 ms. Total throughput: 248.90 iter/sec. [Work thread Nov 7 18:24] Timing 2048K allcomplex FFT, 2 cores, 1 worker. Average times: 4.62 ms. Total throughput: 216.66 iter/sec. [Work thread Nov 7 18:25] Timing 2048K allcomplex FFT, 2 cores, 2 workers. Average times: 8.74, 8.73 ms. Total throughput: 229.01 iter/sec. [Work thread Nov 7 18:25] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 4.01 ms. Total throughput: 249.60 iter/sec. [Work thread Nov 7 18:25] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 7.84, 7.85 ms. Total throughput: 254.96 iter/sec. [Work thread Nov 7 18:26] Timing 2048K allcomplex FFT, 2 cores, 1 worker. Average times: 4.86 ms. Total throughput: 205.82 iter/sec. [Work thread Nov 7 18:26] Timing 2048K allcomplex FFT, 2 cores, 2 workers. Average times: 9.08, 9.07 ms. Total throughput: 220.36 iter/sec. [Work thread Nov 7 18:26] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 4.20 ms. Total throughput: 238.33 iter/sec. [Work thread Nov 7 18:26] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 8.10, 8.10 ms. Total throughput: 246.89 iter/sec. [Work thread Nov 7 18:27] Timing 2048K allcomplex FFT, 2 cores, 1 worker. Average times: 4.60 ms. Total throughput: 217.30 iter/sec. [Work thread Nov 7 18:27] Timing 2048K allcomplex FFT, 2 cores, 2 workers. Average times: 8.60, 8.60 ms. Total throughput: 232.68 iter/sec. [Work thread Nov 7 18:27] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 4.27 ms. Total throughput: 234.22 iter/sec. [Work thread Nov 7 18:27] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 8.60, 8.61 ms. Total throughput: 232.46 iter/sec. [Work thread Nov 7 18:28] Timing 2048K allcomplex FFT, 2 cores, 1 worker. Average times: 4.65 ms. Total throughput: 214.88 iter/sec. [Work thread Nov 7 18:28] Timing 2048K allcomplex FFT, 2 cores, 2 workers. Average times: 8.45, 8.45 ms. Total throughput: 236.82 iter/sec. [Work thread Nov 7 18:28] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 4.09 ms. Total throughput: 244.46 iter/sec. [Work thread Nov 7 18:28] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 8.07, 8.07 ms. Total throughput: 247.86 iter/sec. [Work thread Nov 7 18:29] Timing 2048K allcomplex FFT, 2 cores, 1 worker. Average times: 4.76 ms. Total throughput: 209.89 iter/sec. [Work thread Nov 7 18:29] Timing 2048K allcomplex FFT, 2 cores, 2 workers. Average times: 8.70, 8.69 ms. Total throughput: 230.01 iter/sec. [Work thread Nov 7 18:29] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 4.07 ms. Total throughput: 245.77 iter/sec. [Work thread Nov 7 18:29] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 8.07, 8.05 ms. Total throughput: 248.09 iter/sec. [Work thread Nov 7 18:30] Timing 2048K allcomplex FFT, 2 cores, 1 worker. Average times: 4.82 ms. Total throughput: 207.52 iter/sec. [Work thread Nov 7 18:30] Timing 2048K allcomplex FFT, 2 cores, 2 workers. Average times: 8.89, 8.89 ms. Total throughput: 224.99 iter/sec. [Work thread Nov 7 18:30] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 4.77 ms. Total throughput: 209.49 iter/sec. [Work thread Nov 7 18:30] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 10.39, 10.38 ms. Total throughput: 192.53 iter/sec. [Work thread Nov 7 18:31] Timing 2048K allcomplex FFT, 2 cores, 1 worker. Average times: 4.78 ms. Total throughput: 209.19 iter/sec. [Work thread Nov 7 18:31] Timing 2048K allcomplex FFT, 2 cores, 2 workers. Average times: 8.55, 8.55 ms. Total throughput: 233.82 iter/sec. [Work thread Nov 7 18:31] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 4.26 ms. Total throughput: 234.83 iter/sec. [Work thread Nov 7 18:31] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 8.49, 8.45 ms. Total throughput: 236.21 iter/sec. [Work thread Nov 7 18:32] Timing 2048K allcomplex FFT, 2 cores, 1 worker. Average times: 4.82 ms. Total throughput: 207.42 iter/sec. [Work thread Nov 7 18:32] Timing 2048K allcomplex FFT, 2 cores, 2 workers. Average times: 8.78, 8.77 ms. Total throughput: 227.97 iter/sec. [Work thread Nov 7 18:32] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 4.23 ms. Total throughput: 236.66 iter/sec. [Work thread Nov 7 18:33] Timing 2048K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 8.24, 8.20 ms. Total throughput: 243.41 iter/sec. [Work thread Nov 7 18:34] Timing 4096K allcomplex FFT, 2 cores, 1 worker. Average times: 10.28 ms. Total throughput: 97.27 iter/sec. [Work thread Nov 7 18:35] Timing 4096K allcomplex FFT, 2 cores, 2 workers. Average times: 19.20, 19.08 ms. Total throughput: 104.50 iter/sec. [Work thread Nov 7 18:35] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 9.54 ms. Total throughput: 104.80 iter/sec. [Work thread Nov 7 18:35] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 19.61, 19.61 ms. Total throughput: 102.00 iter/sec. [Work thread Nov 7 18:35] Timing 4096K allcomplex FFT, 2 cores, 1 worker. Average times: 10.33 ms. Total throughput: 96.85 iter/sec. [Work thread Nov 7 18:36] Timing 4096K allcomplex FFT, 2 cores, 2 workers. Average times: 18.75, 18.75 ms. Total throughput: 106.65 iter/sec. [Work thread Nov 7 18:36] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 9.44 ms. Total throughput: 105.99 iter/sec. [Work thread Nov 7 18:36] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 18.88, 18.66 ms. Total throughput: 106.56 iter/sec. [Work thread Nov 7 18:37] Timing 4096K allcomplex FFT, 2 cores, 1 worker. Average times: 10.45 ms. Total throughput: 95.68 iter/sec. [Work thread Nov 7 18:37] Timing 4096K allcomplex FFT, 2 cores, 2 workers. Average times: 18.89, 18.65 ms. Total throughput: 106.56 iter/sec. [Work thread Nov 7 18:37] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 9.64 ms. Total throughput: 103.73 iter/sec. [Work thread Nov 7 18:37] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 19.18, 18.83 ms. Total throughput: 105.23 iter/sec. [Work thread Nov 7 18:38] Timing 4096K allcomplex FFT, 2 cores, 1 worker. Average times: 9.14 ms. Total throughput: 109.42 iter/sec. [Work thread Nov 7 18:38] Timing 4096K allcomplex FFT, 2 cores, 2 workers. Average times: 17.87, 17.87 ms. Total throughput: 111.94 iter/sec. [Work thread Nov 7 18:38] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 8.81 ms. Total throughput: 113.51 iter/sec. [Work thread Nov 7 18:38] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 17.96, 18.07 ms. Total throughput: 111.01 iter/sec. [Work thread Nov 7 18:39] Timing 4096K allcomplex FFT, 2 cores, 1 worker. Average times: 9.63 ms. Total throughput: 103.86 iter/sec. [Work thread Nov 7 18:39] Timing 4096K allcomplex FFT, 2 cores, 2 workers. Average times: 19.07, 18.29 ms. Total throughput: 107.12 iter/sec. [Work thread Nov 7 18:39] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 8.82 ms. Total throughput: 113.39 iter/sec. [Work thread Nov 7 18:39] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 17.91, 17.58 ms. Total throughput: 112.71 iter/sec. [Work thread Nov 7 18:40] Timing 4096K allcomplex FFT, 2 cores, 1 worker. Average times: 10.07 ms. Total throughput: 99.28 iter/sec. [Work thread Nov 7 18:40] Timing 4096K allcomplex FFT, 2 cores, 2 workers. Average times: 19.69, 18.59 ms. Total throughput: 104.58 iter/sec. [Work thread Nov 7 18:40] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 9.05 ms. Total throughput: 110.48 iter/sec. [Work thread Nov 7 18:41] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 18.11, 18.08 ms. Total throughput: 110.53 iter/sec. [Work thread Nov 7 18:41] Timing 4096K allcomplex FFT, 2 cores, 1 worker. Average times: 10.13 ms. Total throughput: 98.71 iter/sec. [Work thread Nov 7 18:41] Timing 4096K allcomplex FFT, 2 cores, 2 workers. Average times: 20.10, 19.32 ms. Total throughput: 101.53 iter/sec. [Work thread Nov 7 18:41] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 9.66 ms. Total throughput: 103.50 iter/sec. [Work thread Nov 7 18:42] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 19.46, 19.29 ms. Total throughput: 103.22 iter/sec. [Work thread Nov 7 18:42] Timing 4096K allcomplex FFT, 2 cores, 1 worker. Average times: 10.13 ms. Total throughput: 98.75 iter/sec. [Work thread Nov 7 18:42] Timing 4096K allcomplex FFT, 2 cores, 2 workers. Average times: 19.89, 19.14 ms. Total throughput: 102.53 iter/sec. [Work thread Nov 7 18:42] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 8.96 ms. Total throughput: 111.63 iter/sec. [Work thread Nov 7 18:43] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 18.11, 17.74 ms. Total throughput: 111.57 iter/sec. [Work thread Nov 7 18:43] Timing 4096K allcomplex FFT, 2 cores, 1 worker. Average times: 10.22 ms. Total throughput: 97.81 iter/sec. [Work thread Nov 7 18:43] Timing 4096K allcomplex FFT, 2 cores, 2 workers. Average times: 19.86, 19.15 ms. Total throughput: 102.56 iter/sec. [Work thread Nov 7 18:44] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 8.85 ms. Total throughput: 113.05 iter/sec. [Work thread Nov 7 18:44] Timing 4096K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 17.70, 17.50 ms. Total throughput: 113.65 iter/sec. [Work thread Nov 7 18:46] Timing 8192K allcomplex FFT, 2 cores, 1 worker. Average times: 20.82 ms. Total throughput: 48.03 iter/sec. [Work thread Nov 7 18:46] Timing 8192K allcomplex FFT, 2 cores, 2 workers. Average times: 41.04, 38.91 ms. Total throughput: 50.07 iter/sec. [Work thread Nov 7 18:46] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 21.14 ms. Total throughput: 47.31 iter/sec. [Work thread Nov 7 18:47] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 42.45, 41.62 ms. Total throughput: 47.59 iter/sec. [Work thread Nov 7 18:47] Timing 8192K allcomplex FFT, 2 cores, 1 worker. Average times: 20.79 ms. Total throughput: 48.10 iter/sec. [Work thread Nov 7 18:47] Timing 8192K allcomplex FFT, 2 cores, 2 workers. Average times: 41.12, 38.82 ms. Total throughput: 50.08 iter/sec. [Work thread Nov 7 18:47] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 20.54 ms. Total throughput: 48.70 iter/sec. [Work thread Nov 7 18:48] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 41.65, 39.76 ms. Total throughput: 49.16 iter/sec. [Work thread Nov 7 18:48] Timing 8192K allcomplex FFT, 2 cores, 1 worker. Average times: 21.55 ms. Total throughput: 46.41 iter/sec. [Work thread Nov 7 18:48] Timing 8192K allcomplex FFT, 2 cores, 2 workers. Average times: 42.03, 39.56 ms. Total throughput: 49.07 iter/sec. [Work thread Nov 7 18:49] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 21.03 ms. Total throughput: 47.55 iter/sec. [Work thread Nov 7 18:49] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 41.89, 41.12 ms. Total throughput: 48.19 iter/sec. [Work thread Nov 7 18:49] Timing 8192K allcomplex FFT, 2 cores, 1 worker. Average times: 20.57 ms. Total throughput: 48.61 iter/sec. [Work thread Nov 7 18:50] Timing 8192K allcomplex FFT, 2 cores, 2 workers. Average times: 41.53, 39.00 ms. Total throughput: 49.72 iter/sec. [Work thread Nov 7 18:50] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 21.75 ms. Total throughput: 45.97 iter/sec. [Work thread Nov 7 18:50] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 43.27, 40.33 ms. Total throughput: 47.90 iter/sec. [Work thread Nov 7 18:50] Timing 8192K allcomplex FFT, 2 cores, 1 worker. Average times: 20.63 ms. Total throughput: 48.48 iter/sec. [Work thread Nov 7 18:51] Timing 8192K allcomplex FFT, 2 cores, 2 workers. Average times: 41.13, 38.54 ms. Total throughput: 50.26 iter/sec. [Work thread Nov 7 18:51] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 19.09 ms. Total throughput: 52.39 iter/sec. [Work thread Nov 7 18:51] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 37.92, 37.10 ms. Total throughput: 53.32 iter/sec. [Work thread Nov 7 18:52] Timing 8192K allcomplex FFT, 2 cores, 1 worker. Average times: 20.66 ms. Total throughput: 48.40 iter/sec. [Work thread Nov 7 18:52] Timing 8192K allcomplex FFT, 2 cores, 2 workers. Average times: 40.30, 38.52 ms. Total throughput: 50.78 iter/sec. [Work thread Nov 7 18:52] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 18.94 ms. Total throughput: 52.80 iter/sec. [Work thread Nov 7 18:52] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 37.50, 37.02 ms. Total throughput: 53.68 iter/sec. [Work thread Nov 7 18:53] Timing 8192K allcomplex FFT, 2 cores, 1 worker. Average times: 22.69 ms. Total throughput: 44.07 iter/sec. [Work thread Nov 7 18:53] Timing 8192K allcomplex FFT, 2 cores, 2 workers. Average times: 43.79, 41.92 ms. Total throughput: 46.69 iter/sec. [Work thread Nov 7 18:53] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 23.85 ms. Total throughput: 41.93 iter/sec. [Work thread Nov 7 18:54] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 47.42, 46.00 ms. Total throughput: 42.83 iter/sec. [Work thread Nov 7 18:54] Timing 8192K allcomplex FFT, 2 cores, 1 worker. Average times: 21.59 ms. Total throughput: 46.31 iter/sec. [Work thread Nov 7 18:54] Timing 8192K allcomplex FFT, 2 cores, 2 workers. Average times: 42.04, 40.28 ms. Total throughput: 48.61 iter/sec. [Work thread Nov 7 18:54] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 20.47 ms. Total throughput: 48.86 iter/sec. [Work thread Nov 7 18:55] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 40.56, 39.13 ms. Total throughput: 50.21 iter/sec. [Work thread Nov 7 18:55] Timing 8192K allcomplex FFT, 2 cores, 1 worker. Average times: 21.80 ms. Total throughput: 45.88 iter/sec. [Work thread Nov 7 18:55] Timing 8192K allcomplex FFT, 2 cores, 2 workers. Average times: 42.42, 40.58 ms. Total throughput: 48.22 iter/sec. [Work thread Nov 7 18:56] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 1 worker. Average times: 19.58 ms. Total throughput: 51.06 iter/sec. [Work thread Nov 7 18:56] Timing 8192K allcomplex FFT, 2 cores hyperthreaded, 2 workers. Average times: 38.75, 37.78 ms. Total throughput: 52.27 iter/sec. 
20171107, 19:11  #6 
"/X\(‘‘)/X\"
Jan 2013
2929_{10} Posts 
c5.2xlarge
Prefers hyperthreads [Work thread Nov 7 18:21] Timing 2048K allcomplex FFT, 4 cores, 1 worker. Average times: 2.72 ms. Total throughput: 367.71 iter/sec. [Work thread Nov 7 18:21] Timing 2048K allcomplex FFT, 4 cores, 4 workers. Average times: 9.07, 9.08, 9.07, 9.07 ms. Total throughput: 440.88 iter/sec. [Work thread Nov 7 18:22] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 2.42 ms. Total throughput: 413.89 iter/sec. [Work thread Nov 7 18:22] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 9.29, 9.36, 9.30, 9.29 ms. Total throughput: 429.63 iter/sec. [Work thread Nov 7 18:22] Timing 2048K allcomplex FFT, 4 cores, 1 worker. Average times: 2.75 ms. Total throughput: 364.04 iter/sec. [Work thread Nov 7 18:22] Timing 2048K allcomplex FFT, 4 cores, 4 workers. Average times: 9.12, 9.13, 9.12, 9.12 ms. Total throughput: 438.40 iter/sec. [Work thread Nov 7 18:23] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 2.49 ms. Total throughput: 401.66 iter/sec. [Work thread Nov 7 18:23] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 9.71, 9.72, 9.82, 9.67 ms. Total throughput: 411.11 iter/sec. [Work thread Nov 7 18:23] Timing 2048K allcomplex FFT, 4 cores, 1 worker. Average times: 3.27 ms. Total throughput: 305.46 iter/sec. [Work thread Nov 7 18:24] Timing 2048K allcomplex FFT, 4 cores, 4 workers. Average times: 9.16, 9.16, 9.16, 9.16 ms. Total throughput: 436.46 iter/sec. [Work thread Nov 7 18:24] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 2.83 ms. Total throughput: 353.76 iter/sec. [Work thread Nov 7 18:24] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 9.93, 9.92, 9.93, 9.91 ms. Total throughput: 403.10 iter/sec. [Work thread Nov 7 18:24] Timing 2048K allcomplex FFT, 4 cores, 1 worker. Average times: 2.46 ms. Total throughput: 406.25 iter/sec. [Work thread Nov 7 18:25] Timing 2048K allcomplex FFT, 4 cores, 4 workers. Average times: 9.13, 9.20, 9.20, 9.11 ms. Total throughput: 436.73 iter/sec. [Work thread Nov 7 18:25] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 2.27 ms. Total throughput: 441.35 iter/sec. [Work thread Nov 7 18:25] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 8.97, 8.80, 9.08, 8.99 ms. Total throughput: 446.51 iter/sec. [Work thread Nov 7 18:26] Timing 2048K allcomplex FFT, 4 cores, 1 worker. Average times: 2.52 ms. Total throughput: 396.51 iter/sec. [Work thread Nov 7 18:26] Timing 2048K allcomplex FFT, 4 cores, 4 workers. Average times: 8.94, 8.98, 8.95, 8.92 ms. Total throughput: 447.00 iter/sec. [Work thread Nov 7 18:26] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 2.22 ms. Total throughput: 451.31 iter/sec. [Work thread Nov 7 18:26] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 8.58, 8.49, 8.49, 8.44 ms. Total throughput: 470.64 iter/sec. [Work thread Nov 7 18:27] Timing 2048K allcomplex FFT, 4 cores, 1 worker. Average times: 2.61 ms. Total throughput: 382.97 iter/sec. [Work thread Nov 7 18:27] Timing 2048K allcomplex FFT, 4 cores, 4 workers. Average times: 8.92, 8.92, 8.92, 8.92 ms. Total throughput: 448.63 iter/sec. [Work thread Nov 7 18:27] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 2.36 ms. Total throughput: 422.92 iter/sec. [Work thread Nov 7 18:27] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 8.88, 8.92, 8.87, 8.81 ms. Total throughput: 451.00 iter/sec. [Work thread Nov 7 18:28] Timing 2048K allcomplex FFT, 4 cores, 1 worker. Average times: 2.47 ms. Total throughput: 404.53 iter/sec. [Work thread Nov 7 18:28] Timing 2048K allcomplex FFT, 4 cores, 4 workers. Average times: 9.05, 9.05, 9.05, 9.05 ms. Total throughput: 442.14 iter/sec. [Work thread Nov 7 18:28] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 2.27 ms. Total throughput: 440.17 iter/sec. [Work thread Nov 7 18:28] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 9.02, 9.06, 9.01, 8.93 ms. Total throughput: 444.30 iter/sec. [Work thread Nov 7 18:29] Timing 2048K allcomplex FFT, 4 cores, 1 worker. Average times: 2.47 ms. Total throughput: 405.53 iter/sec. [Work thread Nov 7 18:29] Timing 2048K allcomplex FFT, 4 cores, 4 workers. Average times: 9.08, 9.14, 9.07, 9.07 ms. Total throughput: 440.02 iter/sec. [Work thread Nov 7 18:29] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 2.24 ms. Total throughput: 445.67 iter/sec. [Work thread Nov 7 18:30] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 8.38, 8.40, 8.75, 8.36 ms. Total throughput: 472.17 iter/sec. [Work thread Nov 7 18:30] Timing 2048K allcomplex FFT, 4 cores, 1 worker. Average times: 2.52 ms. Total throughput: 396.58 iter/sec. [Work thread Nov 7 18:30] Timing 2048K allcomplex FFT, 4 cores, 4 workers. Average times: 9.29, 9.29, 9.29, 9.28 ms. Total throughput: 430.74 iter/sec. [Work thread Nov 7 18:30] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 2.32 ms. Total throughput: 430.35 iter/sec. [Work thread Nov 7 18:31] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 8.77, 8.77, 8.76, 8.69 ms. Total throughput: 457.31 iter/sec. [Work thread Nov 7 18:31] Timing 2048K allcomplex FFT, 4 cores, 1 worker. Average times: 2.45 ms. Total throughput: 407.72 iter/sec. [Work thread Nov 7 18:31] Timing 2048K allcomplex FFT, 4 cores, 4 workers. Average times: 9.14, 9.13, 9.13, 9.13 ms. Total throughput: 437.94 iter/sec. [Work thread Nov 7 18:31] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 2.40 ms. Total throughput: 417.22 iter/sec. [Work thread Nov 7 18:32] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 9.67, 9.83, 9.69, 9.65 ms. Total throughput: 411.93 iter/sec. [Work thread Nov 7 18:32] Timing 2048K allcomplex FFT, 4 cores, 1 worker. Average times: 2.54 ms. Total throughput: 392.94 iter/sec. [Work thread Nov 7 18:32] Timing 2048K allcomplex FFT, 4 cores, 4 workers. Average times: 8.96, 8.96, 8.96, 8.96 ms. Total throughput: 446.33 iter/sec. [Work thread Nov 7 18:33] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 2.27 ms. Total throughput: 440.94 iter/sec. [Work thread Nov 7 18:33] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 8.74, 8.81, 8.74, 8.80 ms. Total throughput: 455.98 iter/sec. [Work thread Nov 7 18:33] Timing 2048K allcomplex FFT, 4 cores, 1 worker. Average times: 2.52 ms. Total throughput: 396.97 iter/sec. [Work thread Nov 7 18:33] Timing 2048K allcomplex FFT, 4 cores, 4 workers. Average times: 9.02, 9.02, 9.02, 9.02 ms. Total throughput: 443.37 iter/sec. [Work thread Nov 7 18:34] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 2.28 ms. Total throughput: 438.83 iter/sec. [Work thread Nov 7 18:34] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 8.74, 8.82, 8.70, 8.77 ms. Total throughput: 456.79 iter/sec. [Work thread Nov 7 18:34] Timing 2048K allcomplex FFT, 4 cores, 1 worker. Average times: 2.68 ms. Total throughput: 373.75 iter/sec. [Work thread Nov 7 18:34] Timing 2048K allcomplex FFT, 4 cores, 4 workers. Average times: 9.89, 9.90, 9.89, 9.77 ms. Total throughput: 405.52 iter/sec. [Work thread Nov 7 18:35] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 2.69 ms. Total throughput: 371.91 iter/sec. [Work thread Nov 7 18:35] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 11.54, 11.73, 11.65, 11.52 ms. Total throughput: 344.51 iter/sec. [Work thread Nov 7 18:35] Timing 2048K allcomplex FFT, 4 cores, 1 worker. Average times: 2.63 ms. Total throughput: 380.29 iter/sec. [Work thread Nov 7 18:36] Timing 2048K allcomplex FFT, 4 cores, 4 workers. Average times: 9.22, 9.22, 9.22, 9.22 ms. Total throughput: 433.66 iter/sec. [Work thread Nov 7 18:36] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 2.41 ms. Total throughput: 414.12 iter/sec. [Work thread Nov 7 18:36] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 9.58, 9.53, 9.60, 9.24 ms. Total throughput: 421.75 iter/sec. [Work thread Nov 7 18:36] Timing 2048K allcomplex FFT, 4 cores, 1 worker. Average times: 2.65 ms. Total throughput: 377.03 iter/sec. [Work thread Nov 7 18:37] Timing 2048K allcomplex FFT, 4 cores, 4 workers. Average times: 9.47, 9.47, 9.48, 9.47 ms. Total throughput: 422.29 iter/sec. [Work thread Nov 7 18:37] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 2.40 ms. Total throughput: 416.27 iter/sec. [Work thread Nov 7 18:37] Timing 2048K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 8.99, 8.99, 8.94, 8.82 ms. Total throughput: 447.74 iter/sec. [Work thread Nov 7 18:41] Timing 4096K allcomplex FFT, 4 cores, 1 worker. Average times: 5.50 ms. Total throughput: 181.98 iter/sec. [Work thread Nov 7 18:41] Timing 4096K allcomplex FFT, 4 cores, 4 workers. Average times: 19.83, 19.83, 19.83, 19.80 ms. Total throughput: 201.78 iter/sec. [Work thread Nov 7 18:41] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 5.20 ms. Total throughput: 192.12 iter/sec. [Work thread Nov 7 18:42] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 20.97, 20.97, 20.97, 21.24 ms. Total throughput: 190.14 iter/sec. [Work thread Nov 7 18:42] Timing 4096K allcomplex FFT, 4 cores, 1 worker. Average times: 5.50 ms. Total throughput: 181.92 iter/sec. [Work thread Nov 7 18:42] Timing 4096K allcomplex FFT, 4 cores, 4 workers. Average times: 19.42, 19.43, 19.42, 19.42 ms. Total throughput: 205.92 iter/sec. [Work thread Nov 7 18:43] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 5.19 ms. Total throughput: 192.83 iter/sec. [Work thread Nov 7 18:43] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 20.02, 20.61, 19.95, 19.90 ms. Total throughput: 198.84 iter/sec. [Work thread Nov 7 18:43] Timing 4096K allcomplex FFT, 4 cores, 1 worker. Average times: 5.72 ms. Total throughput: 174.89 iter/sec. [Work thread Nov 7 18:44] Timing 4096K allcomplex FFT, 4 cores, 4 workers. Average times: 19.76, 19.76, 19.76, 19.76 ms. Total throughput: 202.39 iter/sec. [Work thread Nov 7 18:44] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 5.33 ms. Total throughput: 187.46 iter/sec. [Work thread Nov 7 18:44] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 20.89, 20.88, 20.88, 20.88 ms. Total throughput: 191.54 iter/sec. [Work thread Nov 7 18:44] Timing 4096K allcomplex FFT, 4 cores, 1 worker. Average times: 4.87 ms. Total throughput: 205.28 iter/sec. [Work thread Nov 7 18:45] Timing 4096K allcomplex FFT, 4 cores, 4 workers. Average times: 18.45, 18.51, 18.45, 18.42 ms. Total throughput: 216.68 iter/sec. [Work thread Nov 7 18:45] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 4.69 ms. Total throughput: 213.42 iter/sec. [Work thread Nov 7 18:45] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 19.38, 19.43, 19.60, 19.62 ms. Total throughput: 205.07 iter/sec. [Work thread Nov 7 18:46] Timing 4096K allcomplex FFT, 4 cores, 1 worker. Average times: 4.73 ms. Total throughput: 211.35 iter/sec. [Work thread Nov 7 18:46] Timing 4096K allcomplex FFT, 4 cores, 4 workers. Average times: 18.33, 18.38, 18.34, 18.33 ms. Total throughput: 218.04 iter/sec. [Work thread Nov 7 18:46] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 4.54 ms. Total throughput: 220.46 iter/sec. [Work thread Nov 7 18:46] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 17.69, 17.69, 17.69, 17.71 ms. Total throughput: 226.09 iter/sec. [Work thread Nov 7 18:47] Timing 4096K allcomplex FFT, 4 cores, 1 worker. Average times: 4.96 ms. Total throughput: 201.47 iter/sec. [Work thread Nov 7 18:47] Timing 4096K allcomplex FFT, 4 cores, 4 workers. Average times: 18.66, 18.66, 18.62, 18.66 ms. Total throughput: 214.47 iter/sec. [Work thread Nov 7 18:47] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 4.69 ms. Total throughput: 213.43 iter/sec. [Work thread Nov 7 18:48] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 18.46, 18.57, 18.55, 18.41 ms. Total throughput: 216.27 iter/sec. [Work thread Nov 7 18:48] Timing 4096K allcomplex FFT, 4 cores, 1 worker. Average times: 4.98 ms. Total throughput: 200.71 iter/sec. [Work thread Nov 7 18:48] Timing 4096K allcomplex FFT, 4 cores, 4 workers. Average times: 19.42, 19.44, 19.42, 19.39 ms. Total throughput: 206.02 iter/sec. [Work thread Nov 7 18:49] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 4.95 ms. Total throughput: 201.82 iter/sec. [Work thread Nov 7 18:49] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 20.32, 20.07, 20.15, 20.11 ms. Total throughput: 198.41 iter/sec. [Work thread Nov 7 18:49] Timing 4096K allcomplex FFT, 4 cores, 1 worker. Average times: 5.13 ms. Total throughput: 194.84 iter/sec. [Work thread Nov 7 18:49] Timing 4096K allcomplex FFT, 4 cores, 4 workers. Average times: 19.17, 19.17, 19.17, 19.17 ms. Total throughput: 208.68 iter/sec. [Work thread Nov 7 18:50] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 4.41 ms. Total throughput: 226.78 iter/sec. [Work thread Nov 7 18:50] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 17.85, 17.87, 17.85, 17.85 ms. Total throughput: 224.00 iter/sec. [Work thread Nov 7 18:50] Timing 4096K allcomplex FFT, 4 cores, 1 worker. Average times: 4.99 ms. Total throughput: 200.21 iter/sec. [Work thread Nov 7 18:51] Timing 4096K allcomplex FFT, 4 cores, 4 workers. Average times: 19.12, 19.15, 19.14, 19.12 ms. Total throughput: 209.06 iter/sec. [Work thread Nov 7 18:51] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 4.48 ms. Total throughput: 223.42 iter/sec. [Work thread Nov 7 18:51] Timing 4096K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 18.19, 17.92, 17.81, 17.65 ms. Total throughput: 223.59 iter/sec. [Work thread Nov 7 18:54] Timing 8192K allcomplex FFT, 4 cores, 1 worker. Average times: 10.72 ms. Total throughput: 93.32 iter/sec. [Work thread Nov 7 18:55] Timing 8192K allcomplex FFT, 4 cores, 4 workers. Average times: 39.05, 39.12, 39.01, 39.00 ms. Total throughput: 102.45 iter/sec. [Work thread Nov 7 18:55] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 11.01 ms. Total throughput: 90.86 iter/sec. [Work thread Nov 7 18:55] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 43.68, 45.09, 44.34, 45.01 ms. Total throughput: 89.84 iter/sec. [Work thread Nov 7 18:56] Timing 8192K allcomplex FFT, 4 cores, 1 worker. Average times: 10.51 ms. Total throughput: 95.18 iter/sec. [Work thread Nov 7 18:56] Timing 8192K allcomplex FFT, 4 cores, 4 workers. Average times: 38.90, 38.95, 38.86, 38.87 ms. Total throughput: 102.85 iter/sec. [Work thread Nov 7 18:56] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 10.68 ms. Total throughput: 93.67 iter/sec. [Work thread Nov 7 18:57] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 40.67, 41.64, 41.53, 40.52 ms. Total throughput: 97.36 iter/sec. [Work thread Nov 7 18:57] Timing 8192K allcomplex FFT, 4 cores, 1 worker. Average times: 10.83 ms. Total throughput: 92.31 iter/sec. [Work thread Nov 7 18:57] Timing 8192K allcomplex FFT, 4 cores, 4 workers. Average times: 39.50, 39.53, 39.49, 39.50 ms. Total throughput: 101.25 iter/sec. [Work thread Nov 7 18:58] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 10.86 ms. Total throughput: 92.07 iter/sec. [Work thread Nov 7 18:58] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 42.50, 42.94, 42.89, 42.40 ms. Total throughput: 93.72 iter/sec. [Work thread Nov 7 18:58] Timing 8192K allcomplex FFT, 4 cores, 1 worker. Average times: 10.10 ms. Total throughput: 99.05 iter/sec. [Work thread Nov 7 18:59] Timing 8192K allcomplex FFT, 4 cores, 4 workers. Average times: 39.29, 39.37, 39.20, 39.18 ms. Total throughput: 101.88 iter/sec. [Work thread Nov 7 18:59] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 10.88 ms. Total throughput: 91.90 iter/sec. [Work thread Nov 7 18:59] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 41.43, 42.17, 41.26, 41.58 ms. Total throughput: 96.14 iter/sec. [Work thread Nov 7 19:00] Timing 8192K allcomplex FFT, 4 cores, 1 worker. Average times: 10.27 ms. Total throughput: 97.38 iter/sec. [Work thread Nov 7 19:00] Timing 8192K allcomplex FFT, 4 cores, 4 workers. Average times: 38.44, 38.45, 38.44, 38.34 ms. Total throughput: 104.11 iter/sec. [Work thread Nov 7 19:00] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 9.63 ms. Total throughput: 103.89 iter/sec. [Work thread Nov 7 19:01] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 37.96, 38.35, 38.01, 38.21 ms. Total throughput: 104.90 iter/sec. [Work thread Nov 7 19:01] Timing 8192K allcomplex FFT, 4 cores, 1 worker. Average times: 10.02 ms. Total throughput: 99.82 iter/sec. [Work thread Nov 7 19:01] Timing 8192K allcomplex FFT, 4 cores, 4 workers. Average times: 38.40, 38.46, 38.38, 38.37 ms. Total throughput: 104.16 iter/sec. [Work thread Nov 7 19:02] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 9.28 ms. Total throughput: 107.72 iter/sec. [Work thread Nov 7 19:02] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 36.85, 37.05, 36.85, 36.95 ms. Total throughput: 108.32 iter/sec. [Work thread Nov 7 19:02] Timing 8192K allcomplex FFT, 4 cores, 1 worker. Average times: 11.30 ms. Total throughput: 88.48 iter/sec. [Work thread Nov 7 19:03] Timing 8192K allcomplex FFT, 4 cores, 4 workers. Average times: 42.37, 42.56, 42.39, 42.24 ms. Total throughput: 94.36 iter/sec. [Work thread Nov 7 19:03] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 12.45 ms. Total throughput: 80.33 iter/sec. [Work thread Nov 7 19:03] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 46.97, 48.20, 47.49, 46.94 ms. Total throughput: 84.40 iter/sec. [Work thread Nov 7 19:04] Timing 8192K allcomplex FFT, 4 cores, 1 worker. Average times: 10.53 ms. Total throughput: 94.99 iter/sec. [Work thread Nov 7 19:04] Timing 8192K allcomplex FFT, 4 cores, 4 workers. Average times: 40.29, 40.31, 40.29, 40.29 ms. Total throughput: 99.27 iter/sec. [Work thread Nov 7 19:04] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 10.62 ms. Total throughput: 94.16 iter/sec. [Work thread Nov 7 19:05] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 40.75, 40.28, 40.40, 39.80 ms. Total throughput: 99.24 iter/sec. [Work thread Nov 7 19:05] Timing 8192K allcomplex FFT, 4 cores, 1 worker. Average times: 10.67 ms. Total throughput: 93.70 iter/sec. [Work thread Nov 7 19:05] Timing 8192K allcomplex FFT, 4 cores, 4 workers. Average times: 40.55, 40.59, 40.56, 40.52 ms. Total throughput: 98.63 iter/sec. [Work thread Nov 7 19:06] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 1 worker. Average times: 9.92 ms. Total throughput: 100.81 iter/sec. [Work thread Nov 7 19:06] Timing 8192K allcomplex FFT, 4 cores hyperthreaded, 4 workers. Average times: 38.19, 38.00, 37.79, 37.78 ms. Total throughput: 105.44 iter/sec. 
20171107, 19:12  #7 
"/X\(‘‘)/X\"
Jan 2013
29·101 Posts 
c5.18xlarge
I wouldn't trust these as affinities weren't being set properly due to a bug. I've removed the error output Because the affinities are messed, I decided not to benchmark all the FFTs to find the fastest. [Worker #1 Nov 7 18:28] Timing 2048K allcomplex FFT, 36 cores, 1 worker. Average times: 1.25 ms. Total throughput: 797.24 iter/sec. [Worker #1 Nov 7 18:28] Timing 2048K allcomplex FFT, 36 cores, 2 workers. Average times: 1.08, 1.28 ms. Total throughput: 1706.70 iter/sec. [Worker #1 Nov 7 18:29] Timing 2048K allcomplex FFT, 36 cores, 4 workers. Average times: 2.22, 2.22, 0.84, 1.41 ms. Total throughput: 2799.02 iter/sec. [Worker #1 Nov 7 18:29] Timing 2048K allcomplex FFT, 36 cores, 36 workers. Average times: 34.21, 33.66, 33.35, 33.35, 33.83, 33.93, 33.23, 33.96, 32.88, 33.05, 33.53, 33.75, 33.47, 34.09, 33.56, 33.74, 33.43, 33.99, 8.53, 8.49, 8.76, 8.58, 8.67, 8.67, 8.76, 8.67, 8.74, 19.16, 18.99, 19.29, 19.05, 18.99, 19.06, 19.10, 19.04, 19.16 ms. Total throughput: 2047.25 iter/sec. [Worker #1 Nov 7 18:43] Timing 4096K allcomplex FFT, 36 cores, 1 worker. Average times: 2.01 ms. Total throughput: 498.58 iter/sec. [Worker #1 Nov 7 18:43] Timing 4096K allcomplex FFT, 36 cores, 2 workers. Average times: 2.15, 2.15 ms. Total throughput: 930.42 iter/sec. [Worker #1 Nov 7 18:43] Timing 4096K allcomplex FFT, 36 cores, 4 workers. Average times: 6.26, 6.28, 1.45, 3.59 ms. Total throughput: 1285.26 iter/sec. [Worker #1 Nov 7 18:44] Timing 4096K allcomplex FFT, 36 cores, 36 workers. Average times: 66.77, 69.46, 66.12, 69.89, 67.33, 68.40, 66.98, 69.07, 68.37, 69.44, 68.03, 68.91, 67.69, 69.14, 66.91, 69.37, 67.80, 70.73, 18.66, 18.45, 18.44, 18.66, 18.64, 18.64, 18.74, 18.73, 18.65, 37.09, 37.40, 36.93, 37.02, 37.17, 37.54, 37.17, 37.14, 37.33 ms. Total throughput: 988.67 iter/sec. [Worker #1 Nov 7 18:46] Timing 8192K allcomplex FFT, 36 cores, 1 worker. Average times: 3.06 ms. Total throughput: 326.87 iter/sec. [Worker #1 Nov 7 18:47] Timing 8192K allcomplex FFT, 36 cores, 2 workers. Average times: 6.13, 3.93 ms. Total throughput: 417.28 iter/sec. [Worker #1 Nov 7 18:47] Timing 8192K allcomplex FFT, 36 cores, 4 workers. Average times: 15.32, 15.17, 3.53, 8.20 ms. Total throughput: 536.48 iter/sec. [Worker #1 Nov 7 18:47] Timing 8192K allcomplex FFT, 36 cores, 36 workers. Average times: 137.61, 139.49, 134.09, 136.52, 135.62, 140.54, 137.09, 141.09, 137.18, 139.40, 136.97, 139.04, 134.56, 137.79, 135.77, 138.37, 136.77, 139.72, 37.54, 37.76, 37.79, 37.67, 38.11, 37.71, 37.95, 37.74, 37.68, 74.31, 73.64, 74.08, 74.22, 74.43, 73.80, 74.92, 74.60, 74.30 ms. Total throughput: 490.27 iter/sec. 
20171108, 00:35  #8 
Sep 2003
3^{2}×7×41 Posts 
c5.large seems to be about 30% faster then c4.large, using mprime 29.4b3
I didn't do a proper benchmark, I just started two LL tests at the same time, for nearly identical exponents in the 47.09M range, both in new subdirectories. The c5.large subdirectory had HyperthreadLL=1 in local.txt, as recommended by Mark Rose; the c4.large subdirectory did not, since my own earlier tests indicated that it doesn't help. Note that mprime has not yet been modified to use AVX512 instructions, so further speed improvements may be available. Mlucas v17 does use AVX512, but there's a compile error at the moment... Last fiddled with by GP2 on 20171108 at 02:35 
20171108, 10:14  #9  
Banned
"Luigi"
Aug 2002
Team Italia
4803_{10} Posts 
Quote:
Last fiddled with by ET_ on 20171108 at 10:14 

20171108, 13:26  #10 
(loop (#_fork))
Feb 2006
Cambridge, England
6382_{10} Posts 
The new platform has six rather than four memory channels, and 1MB rather than 256kB L2 caches.

20171109, 06:36  #11 
Sep 2003
3^{2}·7·41 Posts 
Compiling code in Amazon Linux on c5 instances
First of all, if you use Amazon Linux, you should use version 2017.09 or later.
The instance launch page should propose this as one of the options, but if not, the AMI IDs are listed here for the various regions. In this table, we care mostly about the first column, because for c4 or c5 instances you can only use HVM (not PV) and EBSBacked (not Instance Store), as shown in the type matrix. By default, Amazon Linux only supplies a minimum set of packages. If you want a compiler, you have to install it. As described in the Preparing to Compile Software documentation page, you can install the compiler and associated tools with the command Code:
sudo yum groupinstall "Development Tools" However, as described in the Amazon Linux AMI 2017.09 Release Notes, gcc version 6.4 is available as a separate download: Code:
sudo yum install gcc64 Furthermore, you should invoke the compiler with the march=skylakeavx512 flag to generate code that takes advantage of Skylake. This is documented in the man gcc64 page. For example, to compile Mlucas, as described in the README page, you would: Fetch http://www.mersenneforum.org/mayer/src/C/mlucas_v17.txz and then run: Code:
tar xJf mlucas_v17.txz cd mlucas_v17/src mkdir obj cd obj gcc64 c O3 DUSE_AVX512 DUSE_THREADS march=skylakeavx512 ../*.c >& build.log grep i error build.log gcc64 o Mlucas *.o lm lpthread lrt Code:
./Mlucas s m iters 1000 cpu 0:1 >& selftest.log Then copy the Mlucas executable and the mlucas.cfg file to an empty working directory, and create a worktodo.ini file with the usual Test= or DoubleCheck= lines, which you can get from the Manual Assignment page. Or you can use the primenet.py auxiliary program, as described in the README file, to populate worktodo.ini When invoking the program in the working directory, use Code:
nohup ./Mlucas cpu 0:1 & 
Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Upgrading with multiple instances  smoffat  Software  1  20081115 05:04 
q6600 = run 4 instances of prime95?  bazza  Information & Answers  2  20070920 23:23 
2 instances  brandonriffel  Software  3  20070215 16:15 
running two instances of llrnet  drakkar67  Prime Sierpinski Project  6  20060504 04:20 
multiple instances  Mayhem316  Software  2  20030729 13:34 