diff --git a/plotting/roofline_data.json b/plotting/roofline_data.json index a7bd0de474416b7a93395e7607314fb949384b70..24aa9880e93cb5ec574a206c6434530b07a446ff 100644 --- a/plotting/roofline_data.json +++ b/plotting/roofline_data.json @@ -1,62 +1,164 @@ { "skylakesp2": { "hostname": "skylakesp2", - "date": "Wed 05 Apr 2023 01:52:27", + "date": "Wed 06 Dec 2023 04:01:16", "perf": { - "peakflops": 175468.42, - "peakflops_avx": 700520.72, - "peakflops_avx512": 1387513.58, - "peakflops_avx512_fma": 2775042.98, - "peakflops_avx_fma": 1400352.81, - "peakflops_sse": 350839.04 + "peakflops": 191576.11, + "peakflops_avx": 763833.61, + "peakflops_avx512": 1400126.51, + "peakflops_avx512_fma": 2788529.02, + "peakflops_avx_fma": 1530811.16, + "peakflops_sse": 383112.88 }, "bw": { - "clload": 223674.53, - "load": 222974.73, - "load_avx": 222807.28, - "load_avx512": 223727.57, - "load_mem": 222455.68, - "load_sse": 222343.68 + "clload": 224237.22, + "load": 224121.85, + "load_avx": 223421.16, + "load_avx512": 224137.22, + "load_mem": 223113.84, + "load_sse": 223041.82, + "clcopy": 133911.59, + "copy": 133004.75, + "copy_avx": 133527.2, + "copy_avx512": 135695.95, + "copy_mem": 172224.93, + "copy_mem_avx": 182510.65, + "copy_mem_avx512": 195685.28, + "copy_mem_sse": 173442.02, + "copy_sse": 133260.19, + "stream": 149736.99, + "stream_avx": 150111.84, + "stream_avx512": 151654.29, + "stream_avx512_fma": 151779.8, + "stream_avx_fma": 150781.49, + "stream_mem_avx": 185286.22, + "stream_mem_avx512": 194939.75, + "stream_mem_avx_fma": 186048.01, + "stream_mem_sse": 178267.37, + "stream_mem_sse_fma": 175031.84, + "stream_sp": 149653.78, + "stream_sp_avx": 150356.24, + "stream_sp_avx512": 151167.69, + "stream_sp_avx512_fma": 152009.84, + "stream_sp_avx_fma": 150967.54, + "stream_sp_mem_avx": 186155.2, + "stream_sp_mem_avx512": 194857.15, + "stream_sp_mem_avx512_fma": 195048.62, + "stream_sp_mem_avx_fma": 186185.8, + "stream_sp_mem_sse": 178561.14, + "stream_sp_mem_sse_fma": 177882.85, + "stream_sp_sse": 151165.42, + "stream_sp_sse_fma": 150875.04, + "stream_sse": 150824.25, + "stream_sse_fma": 151115.48 } }, "icx36": { "hostname": "icx36", - "date": "Wed 05 Apr 2023 01:52:20", + "date": "Wed 06 Dec 2023 03:57:23", "perf": { - "peakflops": 356545.52, - "peakflops_avx": 1757982.29, - "peakflops_avx512": 2926326.05, - "peakflops_avx512_fma": 5554621.48, - "peakflops_avx_fma": 3520359.9, - "peakflops_sse": 883549.78 + "peakflops": 302985.33, + "peakflops_avx": 1452830.85, + "peakflops_avx512": 2721965.07, + "peakflops_avx512_fma": 5610446.63, + "peakflops_avx_fma": 3484664.82, + "peakflops_sse": 824665.88 }, "bw": { - "clload": 359284.52, - "load": 360422.78, - "load_avx": 361228.42, - "load_avx512": 359790.13, - "load_mem": 360727.01, - "load_sse": 360198.79 + "clload": 371310.03, + "load": 364667.12, + "load_avx": 372186.78, + "load_avx512": 370037.47, + "load_mem": 373005.21, + "load_sse": 372330.65, + "clcopy": 206674.12, + "copy": 205276.01, + "copy_avx": 211103.88, + "copy_avx512": 209326.52, + "copy_mem": 336722.44, + "copy_mem_avx": 332971.05, + "copy_mem_avx512": 333131.43, + "copy_mem_sse": 338922.21, + "copy_sse": 210318.6, + "stream": 237068.79, + "stream_avx": 242574.99, + "stream_avx512": 243102.71, + "stream_avx512_fma": 244294.47, + "stream_avx_fma": 242808.27, + "stream_mem_avx": 334227.77, + "stream_mem_avx512": 336268.38, + "stream_mem_avx_fma": 333878.5, + "stream_mem_sse": 337718.84, + "stream_mem_sse_fma": 336922.86, + "stream_sp": 236001.67, + "stream_sp_avx": 242223.72, + "stream_sp_avx512": 239843.86, + "stream_sp_avx512_fma": 243636.0, + "stream_sp_avx_fma": 242763.11, + "stream_sp_mem_avx": 335298.17, + "stream_sp_mem_avx512": 336247.2, + "stream_sp_mem_avx512_fma": 334333.07, + "stream_sp_mem_avx_fma": 334012.93, + "stream_sp_mem_sse": 337377.02, + "stream_sp_mem_sse_fma": 337325.34, + "stream_sp_sse": 242465.76, + "stream_sp_sse_fma": 241854.64, + "stream_sse": 241103.83, + "stream_sse_fma": 242330.87 } }, "icx32": { "hostname": "icx32", - "date": "Thu 03 Aug 2023 09:31:24", + "date": "Wed 06 Dec 2023 04:42:47", "perf": { - "peakflops": 341026.79, - "peakflops_avx": 1665387.7, - "peakflops_avx512": 2929571.2, - "peakflops_avx512_fma": 5863109.89, - "peakflops_avx_fma": 3331900.48, - "peakflops_sse": 835759.6 + "peakflops": 347316.31, + "peakflops_avx": 1666961.37, + "peakflops_avx512": 2936992.07, + "peakflops_avx512_fma": 5874509.57, + "peakflops_avx_fma": 3330175.26, + "peakflops_sse": 836426.5 }, "bw": { - "clload": 352128.79, - "load": 351975.45, - "load_avx": 355997.88, - "load_avx512": 350567.06, - "load_mem": 355333.83, - "load_sse": 355441.79 + "clload": 352623.68, + "load": 351671.38, + "load_avx": 355451.26, + "load_avx512": 351943.93, + "load_mem": 355361.49, + "load_sse": 355606.88, + "clcopy": 209340.42, + "copy": 300090.39, + "copy_avx": 301682.63, + "copy_avx512": 300385.84, + "copy_mem": 319597.63, + "copy_mem_avx": 317390.23, + "copy_mem_avx512": 316579.65, + "copy_mem_sse": 319115.04, + "copy_sse": 302063.84, + "stream": 309766.36, + "stream_avx": 309380.7, + "stream_avx512": 307797.77, + "stream_avx512_fma": 307255.27, + "stream_avx_fma": 309337.9, + "stream_mem_avx": 321319.09, + "stream_mem_avx512": 320969.56, + "stream_mem_avx_fma": 321284.5, + "stream_mem_sse": 322563.27, + "stream_mem_sse_fma": 322935.75, + "stream_sp": 308123.47, + "stream_sp_avx": 309431.7, + "stream_sp_avx512": 302386.78, + "stream_sp_avx512_fma": 307270.5, + "stream_sp_avx_fma": 309519.91, + "stream_sp_mem_avx": 321146.25, + "stream_sp_mem_avx512": 320779.22, + "stream_sp_mem_avx512_fma": 319507.23, + "stream_sp_mem_avx_fma": 320953.4, + "stream_sp_mem_sse": 322721.35, + "stream_sp_mem_sse_fma": 322624.44, + "stream_sp_sse": 309980.62, + "stream_sp_sse_fma": 310705.96, + "stream_sse": 310301.7, + "stream_sse_fma": 310048.07 } }, "rome1": { @@ -78,22 +180,56 @@ }, "genoa1": { "hostname": "genoa1", - "date": "Fri 04 Aug 2023 02:12:45", + "date": "Wed 06 Dec 2023 04:40:44", "perf": { - "peakflops": 1368628.47, - "peakflops_avx": 5029454.95, - "peakflops_avx512": 5438098.11, - "peakflops_avx512_fma": 5813369.46, - "peakflops_avx_fma": 5823783.55, - "peakflops_sse": 2351787.31 + "peakflops": 1741537.21, + "peakflops_avx": 5647161.82, + "peakflops_avx512": 6863817.23, + "peakflops_avx512_fma": 7324962.91, + "peakflops_avx_fma": 7327090.92, + "peakflops_sse": 2958173.97 }, "bw": { - "clload": 1134599.06, - "load": 1131179.34, - "load_avx": 1132381.32, - "load_avx512": 1132872.9, - "load_mem": 1129060.9, - "load_sse": 1132834.75 + "clload": 1136692.99, + "load": 1132123.61, + "load_avx": 1128481.09, + "load_avx512": 1140680.35, + "load_mem": 1131031.14, + "load_sse": 1132199.16, + "clcopy": 2237797.38, + "copy": 1551694.52, + "copy_avx": 1944285.99, + "copy_avx512": 2118524.86, + "copy_mem": 1041020.07, + "copy_mem_avx": 1075602.4, + "copy_mem_avx512": 1072237.74, + "copy_mem_sse": 1059201.61, + "copy_sse": 1777570.45, + "stream": 1890985.23, + "stream_avx": 2190902.45, + "stream_avx512": 2277729.75, + "stream_avx512_fma": 2264519.67, + "stream_avx_fma": 2400759.77, + "stream_mem_avx": 1082966.23, + "stream_mem_avx512": 1097757.5, + "stream_mem_avx_fma": 1083338.4, + "stream_mem_sse": 1073931.87, + "stream_mem_sse_fma": 1074093.74, + "stream_sp": 1361574.05, + "stream_sp_avx": 2266666.64, + "stream_sp_avx512": 2164379.43, + "stream_sp_avx512_fma": 2232869.19, + "stream_sp_avx_fma": 2411565.29, + "stream_sp_mem_avx": 1083383.43, + "stream_sp_mem_avx512": 1102042.11, + "stream_sp_mem_avx512_fma": 1101998.51, + "stream_sp_mem_avx_fma": 1079838.89, + "stream_sp_mem_sse": 1074224.27, + "stream_sp_mem_sse_fma": 1071852.65, + "stream_sp_sse": 2079301.95, + "stream_sp_sse_fma": 1675522.32, + "stream_sse": 2068784.3, + "stream_sse_fma": 2223342.22 } }, "medusa": { @@ -115,5 +251,251 @@ "load_mem": 88748.42, "load_sse": 88742.26 } + }, + "casclakesp2": { + "hostname": "casclakesp2", + "date": "Wed 06 Dec 2023 04:00:16", + "perf": { + "peakflops": 198931.63, + "peakflops_avx": 794256.16, + "peakflops_avx512": 1581720.1, + "peakflops_avx512_fma": 3157764.05, + "peakflops_avx_fma": 1589809.96, + "peakflops_sse": 397601.78 + }, + "bw": { + "clload": 221693.56, + "load": 218666.18, + "load_avx": 220362.53, + "load_avx512": 221443.71, + "load_mem": 218325.63, + "load_sse": 218287.55, + "clcopy": 138852.46, + "copy": 137936.51, + "copy_avx": 138749.05, + "copy_avx512": 140057.61, + "copy_mem": 167216.72, + "copy_mem_avx": 178348.56, + "copy_mem_avx512": 193942.34, + "copy_mem_sse": 168700.77, + "copy_sse": 138569.27, + "stream": 153743.85, + "stream_avx": 154775.97, + "stream_avx512": 154774.15, + "stream_avx512_fma": 154957.28, + "stream_avx_fma": 154727.06, + "stream_mem_avx": 183463.17, + "stream_mem_avx512": 192199.77, + "stream_mem_avx_fma": 184733.28, + "stream_mem_sse": 175947.06, + "stream_mem_sse_fma": 175084.87, + "stream_sp": 152859.85, + "stream_sp_avx": 154731.72, + "stream_sp_avx512": 143452.81, + "stream_sp_avx512_fma": 154792.15, + "stream_sp_avx_fma": 154735.71, + "stream_sp_mem_avx": 184867.62, + "stream_sp_mem_avx512": 191367.62, + "stream_sp_mem_avx512_fma": 193790.86, + "stream_sp_mem_avx_fma": 184793.34, + "stream_sp_mem_sse": 175912.18, + "stream_sp_mem_sse_fma": 175141.26, + "stream_sp_sse": 154429.04, + "stream_sp_sse_fma": 154457.87, + "stream_sse": 154473.85, + "stream_sse_fma": 154476.72 + } + }, + "genoa2": { + "hostname": "genoa2", + "date": "Wed 06 Dec 2023 03:32:07", + "perf": { + "peakflops": 782913.26, + "peakflops_avx": 2706146.46, + "peakflops_avx512": 3102270.14, + "peakflops_avx512_fma": 3309503.37, + "peakflops_avx_fma": 3308892.26, + "peakflops_sse": 1375423.06 + }, + "bw": { + "clload": 495627.41, + "load": 493751.76, + "load_avx": 492690.82, + "load_avx512": 486975.07, + "load_mem": 493650.5, + "load_sse": 493936.73, + "clcopy": 290622.04, + "copy": 286341.73, + "copy_avx": 288232.4, + "copy_avx512": 287975.29, + "copy_mem": 409707.12, + "copy_mem_avx": 408579.17, + "copy_mem_avx512": 419509.4, + "copy_mem_sse": 409253.25, + "copy_sse": 285598.21, + "stream": 327132.87, + "stream_avx": 334618.27, + "stream_avx512": 334032.47, + "stream_avx512_fma": 331139.83, + "stream_avx_fma": 338330.73, + "stream_mem_avx": 448155.86, + "stream_mem_avx512": 453523.56, + "stream_mem_avx_fma": 470345.9, + "stream_mem_sse": 431041.16, + "stream_mem_sse_fma": 429452.82, + "stream_sp": 326454.25, + "stream_sp_avx": 334369.77, + "stream_sp_avx512": 328632.18, + "stream_sp_avx512_fma": 330990.71, + "stream_sp_avx_fma": 338729.68, + "stream_sp_mem_avx": 448045.84, + "stream_sp_mem_avx512": 468444.46, + "stream_sp_mem_avx512_fma": 453837.61, + "stream_sp_mem_avx_fma": 452128.9, + "stream_sp_mem_sse": 437658.64, + "stream_sp_mem_sse_fma": 427841.3, + "stream_sp_sse": 326304.31, + "stream_sp_sse_fma": 326724.04, + "stream_sse": 326685.87, + "stream_sse_fma": 326701.01 + } + }, + "naples1": { + "hostname": "naples1", + "date": "Wed 06 Dec 2023 04:38:48", + "perf": { + "peakflops": 384649.79, + "peakflops_avx": 809279.65, + "peakflops_avx_fma": 880098.45, + "peakflops_sse": 773973.09 + }, + "bw": { + "clload": 303436.9, + "load": 299789.87, + "load_avx": 298704.9, + "load_mem": 300478.91, + "load_sse": 298155.54, + "clcopy": 165727.15, + "copy": 164280.55, + "copy_avx": 164388.62, + "copy_mem": 261852.44, + "copy_mem_avx": 262189.71, + "copy_mem_sse": 261862.75, + "copy_sse": 164415.93, + "stream": 181839.35, + "stream_avx": 181301.45, + "stream_avx_fma": 181551.87, + "stream_mem_avx": 258993.25, + "stream_mem_avx_fma": 259411.01, + "stream_mem_sse": 259472.54, + "stream_mem_sse_fma": 259491.12, + "stream_sp": 181575.4, + "stream_sp_avx": 181283.48, + "stream_sp_avx_fma": 181453.02, + "stream_sp_mem_avx": 259393.9, + "stream_sp_mem_avx_fma": 259342.37, + "stream_sp_mem_sse": 259542.24, + "stream_sp_mem_sse_fma": 259220.59, + "stream_sp_sse": 181503.09, + "stream_sp_sse_fma": 181775.95, + "stream_sse": 181458.49, + "stream_sse_fma": 181288.78 + } + }, + "optane1": { + "hostname": "optane1", + "date": "Wed 06 Dec 2023 04:38:59", + "perf": { + "peakflops": 401117.09, + "peakflops_avx": 1725257.94, + "peakflops_avx512": 2844994.52, + "peakflops_avx512_fma": 5696138.55, + "peakflops_avx_fma": 3446874.67, + "peakflops_sse": 889982.76 + }, + "bw": { + "clload": 178634.72, + "load": 178683.99, + "load_avx": 179153.09, + "load_avx512": 178670.03, + "load_mem": 178993.22, + "load_sse": 179143.45, + "clcopy": 104944.03, + "copy": 153312.52, + "copy_avx": 153435.98, + "copy_avx512": 153368.56, + "copy_mem": 161977.06, + "copy_mem_avx": 160740.4, + "copy_mem_avx512": 161584.51, + "copy_mem_sse": 161403.81, + "copy_sse": 153441.19, + "stream": 156810.2, + "stream_avx": 156580.83, + "stream_avx512": 156656.8, + "stream_avx512_fma": 156639.09, + "stream_avx_fma": 156288.09, + "stream_mem_avx": 161179.59, + "stream_mem_avx512": 162151.0, + "stream_mem_avx_fma": 161244.82, + "stream_mem_sse": 161692.68, + "stream_mem_sse_fma": 161694.29, + "stream_sp": 156736.24, + "stream_sp_avx": 156654.29, + "stream_sp_avx512": 155852.93, + "stream_sp_avx512_fma": 156578.2, + "stream_sp_avx_fma": 156401.27, + "stream_sp_mem_avx": 161443.02, + "stream_sp_mem_avx512": 162269.97, + "stream_sp_mem_avx512_fma": 162182.91, + "stream_sp_mem_avx_fma": 161122.99, + "stream_sp_mem_sse": 161643.96, + "stream_sp_mem_sse_fma": 161691.11, + "stream_sp_sse": 156622.08, + "stream_sp_sse_fma": 156654.5, + "stream_sse": 156722.99, + "stream_sse_fma": 156724.27 + } + }, + "rome2": { + "hostname": "rome2", + "date": "Wed 06 Dec 2023 04:40:22", + "perf": { + "peakflops": 387029.08, + "peakflops_avx": 1615438.45, + "peakflops_avx_fma": 1753096.9, + "peakflops_sse": 773242.24 + }, + "bw": { + "clload": 168146.96, + "load": 168258.38, + "load_avx": 168347.42, + "load_mem": 169747.86, + "load_sse": 167974.21, + "clcopy": 100615.29, + "copy": 100313.89, + "copy_avx": 100363.43, + "copy_mem": 162865.9, + "copy_mem_avx": 163191.29, + "copy_mem_sse": 163047.28, + "copy_sse": 100149.02, + "stream": 112515.32, + "stream_avx": 112467.94, + "stream_avx_fma": 112480.26, + "stream_mem_avx": 157486.73, + "stream_mem_avx_fma": 157852.7, + "stream_mem_sse": 157792.52, + "stream_mem_sse_fma": 158740.4, + "stream_sp": 112605.41, + "stream_sp_avx": 112348.2, + "stream_sp_avx_fma": 112381.14, + "stream_sp_mem_avx": 158379.2, + "stream_sp_mem_avx_fma": 157662.96, + "stream_sp_mem_sse": 158568.55, + "stream_sp_mem_sse_fma": 158355.93, + "stream_sp_sse": 112688.97, + "stream_sp_sse_fma": 112490.14, + "stream_sse": 112685.3, + "stream_sse_fma": 112676.21 + } } } \ No newline at end of file