<!DOCTYPE html><html lang="en"><head><title>APS report</title><link rel="icon" href="favicon.ico" type="image/x-icon"><meta name="nexus-supported-models" content="mps_plugin.Result"><link rel="nexus-home" href="?nexus={nexus}"><link rel="nexus-app" href="?nexus={nexus}&path={properties.path}"><meta name="description" content="APS viewer application"><style type="text/css">body,html{height:100%}body{margin:0;font-family:'Segoe UI',Tahoma,Geneva,Verdana,sans-serif;font-size:15px;color:#000;background-color:#f7f7f7;overflow:hidden}@media (min-width:1320px){#content{width:1190px;font-size:16.5px}}@media (min-width:1540px){#content{width:1410px;font-size:18px}}@media (min-width:1760px){#content{width:1630px;font-size:19.5px}}.ablate{display:none}#leftMenuOpenedState,#topMenuOpenedState{background-color:#00aeef;padding:.5em;color:#fff}#leftMenuOpenedState>h4,#topMenuOpenedState>h4{font-weight:700;margin-bottom:.8em}#leftMenuOpenedState>div,#leftMenuOpenedState>span,#topMenuOpenedState>div,#topMenuOpenedState>span{opacity:.75;padding-bottom:.8em;font-style:italic}#leftMenuOpenedState>a,#leftMenuOpenedState>a:visited,#topMenuOpenedState>a,#topMenuOpenedState>a:visited{font-weight:400;color:#fff;opacity:.9}#leftMenuOpenedState>a:hover,#topMenuOpenedState>a:hover{opacity:1}#topMenuOpenedState>a{padding-top:.6em;padding-left:4em}#leftMenuClosedState,#topMenuClosedState{background-color:#0071c5}#leftMenuClosedState:hover,#topMenuClosedState:hover{cursor:pointer;background-color:#00aeef}#topMenuClosedState{font-size:22.5px;line-height:60px;color:#fff;text-overflow:ellipsis;white-space:nowrap;overflow:hidden}#topMenuClosedState sup{display:none;font-size:.4em;font-weight:400;line-height:.8em}#topMenuClosedState span{padding-left:13px;padding-right:13px}#topMenuClosedState span.icon.menu:after{vertical-align:middle}#topMenuClosedState span.icon.intel{line-height:0;float:right;font-size:2em;padding-top:13px}@media (min-width:625px){#topMenuClosedState sup{display:inline}}#leftMenuClosedState div.icon{font-size:2.5em;width:1em;margin:15px auto}#leftMenuClosedState div.icon.intel{font-size:4em;position:absolute;margin:0;bottom:0;left:15px}#leftMenuClosedState,#leftMenuOpenedState{z-index:5;position:fixed}#topMenuClosedState,#topMenuOpenedState{z-index:5}#leftMenuClosedState,#leftMenuOpenedState,.feedback-panel.closed a.button{display:none}#topMenuOpenedState{display:block;position:fixed;left:0;right:0;top:-200px;bottom:0;width:100%;height:200px}#topMenuClosedState{display:block;position:absolute;left:0;right:0;top:0;height:60px}#contentWrap{position:fixed;overflow-y:auto;overflow-x:hidden;left:0;right:0;top:0;bottom:0;padding:20px;margin-top:60px}.menuOpened #topMenuOpenedState{top:0}.menuOpened #topMenuClosedState{top:-60px}.menuOpened #contentWrap{opacity:.35;margin-top:200px}@media (min-width:850px){#topMenuClosedState,#topMenuOpenedState{display:none}#leftMenuClosedState,#leftMenuOpenedState{display:block;left:-180px;top:0;bottom:0;width:180px}#leftMenuClosedState{left:0;width:90px}#contentWrap{left:0;top:0;padding-left:110px;margin-top:0}.menuOpened #leftMenuOpenedState{left:0}.menuOpened #leftMenuClosedState{left:-90px}.menuOpened #contentWrap{left:135px;margin-top:0}}.anim{-webkit-transition:.5s all cubic-bezier(.8,0,.2,1);-moz-transition:.5s all cubic-bezier(.8,0,.2,1);-o-transition:.5s all cubic-bezier(.8,0,.2,1);transition:.5s all cubic-bezier(.8,0,.2,1)}@media print{#leftMenuClosedState,#leftMenuOpenedState,#topMenuClosedState,#topMenuOpenedState{display:none}#contentWrap{position:relative}}#content{margin:0 auto;width:100%;max-height:0;overflow:hidden}#content:after{content:"Screen is too small";position:absolute;top:133.33333333px;left:0;width:100%;margin:auto;text-align:center;opacity:.5}@media (min-width:400px){#content{max-height:inherit}#content:after{content:""}}#bricks{clear:both}#bricks>article{background-color:#fff}.not-supported-browser{color:#000;background-color:#fff;padding:30px;height:100%}.feedback-panel{position:absolute;color:#fff;background-color:#0071c5;font-weight:400;padding:10px;width:200px;height:70px;transition:left ease-in-out .5s 2s,top ease-in-out .5s 2s}.feedback-panel:after{content:"";position:absolute;width:0;height:0}.feedback-panel.left{left:0;bottom:30px}.feedback-panel.left:after{left:220px;top:0;border:45px solid transparent;border-left-width:20px;border-left-color:#0071c5;border-right:0}.feedback-panel.top{right:30px;top:140px}.feedback-panel.top:after{left:100px;top:90px;border:60px solid transparent;border-top-width:20px;border-top-color:#0071c5;border-bottom-width:0}.feedback-panel.closed.left{left:-240px}.feedback-panel.closed.top{top:-110px}.feedback-panel .question{margin-bottom:.8em}.feedback-panel a.button{display:inline-block;color:#fff;padding:5px 10px;margin-bottom:5px;text-decoration:none;background-color:rgba(255,255,255,.2);border-radius:5px}.feedback-panel a.button:hover{background-color:rgba(255,255,255,.3)}.icon:after{display:inline-block;width:1em;height:1em;line-height:1em;vertical-align:bottom}.icon.menu:after{content:url(data:image/svg+xml,%3Csvg%20xmlns%3D%27http%3A%2F%2Fwww.w3.org%2F2000%2Fsvg%27%20viewBox%3D%270%200%2020%2020%27%20fill%3D%27%23FFF%27%3E%3Crect%20x%3D%273%27%20y%3D%273%27%20width%3D%2714%27%20height%3D%273%27%2F%3E%3Crect%20x%3D%273%27%20y%3D%2713%27%20width%3D%2714%27%20height%3D%273%27%2F%3E%3Crect%20x%3D%273%27%20y%3D%278%27%20width%3D%2714%27%20height%3D%273%27%2F%3E%3C%2Fsvg%3E)}.icon.email:after{content:url("data:image/svg+xml,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 20 20'%3E%3Cpath fill='white' d='M10,0L2,6v9.5h16V6L10,0z M10,1.5L16,6l-6,3.2L4,6L10,1.5z M16.5,14.2h-13v-7l6.5,3.3L16.5,7V14.2z'/%3E%3C/svg%3E")}.icon.intel:after{content:url("data:image/svg+xml,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 293 293' fill='white'%3E%3Cpath d='M 291.28737,55.058565 C 277.50683,-12.11587 147.525,-16.369963 63.737895,34.808098 L 63.737895,40.460487 C 147.41497,-2.7316788 266.14859,-2.4562338 276.95315,59.422033 C 280.59385,79.920851 269.12707,101.24048 248.56679,113.50757 L 248.56679,129.56114 C 273.31695,120.47998 298.61629,91.08816 291.28737,55.058565 M 138.92617,172.67053 C 81.102616,178.02238 20.853265,169.59848 12.418485,124.23687 C 8.2296052,101.90025 18.426965,78.192925 31.877897,63.483967 L 31.877897,55.608451 C 7.6234054,76.957185 -5.5514356,103.96031 2.0555709,135.84419 C 11.756767,176.75954 63.462304,199.91949 142.39931,192.20804 C 173.65361,189.19069 214.55609,179.09205 242.94244,163.42581 L 242.94244,141.16845 C 217.14544,156.61695 174.47938,169.38024 138.92617,172.67053 z'/%3E%3Cpath d='M 238.31142,45.347552 L 223.15342,45.347552 L 223.15342,113.16 C 223.15342,121.1244 226.95767,128.05143 238.31142,129.14959'/%3E%3Cpath d='M 57.729915,70.130433 L 42.57092,70.130433 L 42.57092,114.42193 C 42.57092,122.38934 46.375174,129.31386 57.729915,130.41152'/%3E%3Cpath d='M 188.42548,81.588742 C 183.30029,81.588742 179.33248,84.253255 177.67794,87.85316 C 176.68611,90.022849 176.3545,91.672094 176.19095,94.338616 L 199.39759,94.338616 C 199.06648,87.82553 196.14302,81.588742 188.42548,81.588742 M 176.19095,104.61387 C 176.19095,112.33513 181.03904,118.01982 189.52834,118.01982 C 196.19954,118.01982 199.50663,116.15155 203.3654,112.33563 L 212.62645,121.26456 C 206.67299,127.14267 200.44444,130.71544 189.41981,130.71544 C 175.03357,130.71544 161.25152,122.82991 161.25152,99.85904 C 161.25152,80.21479 173.26798,69.115166 189.0887,69.115166 C 205.12948,69.115166 214.33351,82.110693 214.33351,99.173822 L 214.33351,104.61437 L 176.19095,104.61437'/%3E%3Cpath d='M 98.576374,82.329722 C 102.98533,82.329722 104.80493,84.500918 104.80493,88.045061 L 104.80493,129.78055 L 119.85388,129.78055 L 119.85388,87.990304 C 119.85388,79.499933 115.33339,70.129931 102.15955,70.129931 L 71.125329,70.129931 L 71.125329,129.78005 L 86.118769,129.78005 L 86.118769,82.329219'/%3E%3Cpath d='M 148.57135,129.69867 C 136.2803,129.69867 131.09859,121.12491 131.09859,112.66317 L 131.09859,53.837923 L 146.09203,53.837923 L 146.09203,70.130433 L 157.44627,70.130433 L 157.44627,82.329722 L 146.09203,82.329722 L 146.09203,111.75641 C 146.09203,115.21716 147.74557,117.11508 151.32725,117.11508 L 157.44627,117.11508 L 157.44627,129.69867 L 148.57135,129.69867'/%3E%3Crect x='42.570919' y='47.40823' width='15.158995' height='14.425744'/%3E%3Ctext x='244.26051' y='55.85825'%3ER%3C/text%3E%3C/svg%3E")}.icon.flag:after{content:url("data:image/svg+xml,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 20 20'%3E%3Cpolygon fill='%23BE3B26' points='4.5,0 11,3 10,6 13,11 7,9 7,7.5 4.5,7 4.5,15.5 3,15.5 3,0'/%3E%3C/svg%3E")}.icon.cross:after{content:url("data:image/svg+xml,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 20 20'%3E%3Cpolygon fill='%230071C5' points='1,17 3,19 10,12 17,19 19,17 12,10 19,3 17,1 10,8 3,1 1,3 8,10'/%3E%3C/svg%3E")}.icon.arrow-s-w:after{content:url("data:image/svg+xml,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 20 20'%3E%3Cpolygon fill='grey' points='19,3 17,1 5.5,12.5 1,8 1,19 12,19 7.5,14.5'/%3E%3C/svg%3E")}.tooltipped[data-tip-id]{cursor:default}.tooltipped[data-tip-id]:not(.icon){position:relative;cursor:help}.tooltipped[data-tip-id]:not(.icon):before{content:"";position:absolute;width:100%;bottom:.1em;border-bottom-width:1px;border-bottom-style:dashed;border-bottom-color:inherit}h3.tooltipped[data-tip-id],h4.tooltipped[data-tip-id]{display:table}.tip>div[data-id]{position:fixed;padding:.75em;display:none;max-width:35%;-moz-box-sizing:border-box;-webkit-box-sizing:border-box;box-sizing:border-box;background-color:rgba(255,255,255,.9);border:0 solid #000;box-shadow:rgba(0,0,0,.25) 0 5px 8px;z-index:10}.tip>div[data-id]:after{position:absolute;content:"";height:0;border:5px solid transparent}.tip>div[data-id].north{border-bottom-width:5px}.tip>div[data-id].north:after{bottom:-16px;border-top:6px solid #000}.tip>div[data-id].south{border-top-width:5px}.tip>div[data-id].south:after{top:-16px;border-bottom:6px solid #000}.tip>div[data-id].west:after{left:15px}.tip>div[data-id].east:after{right:1px}.tip>div[data-id].visible{display:block}.tip>div[data-id$=_ISSUE]{border-color:#bf3600}.tip>div[data-id$=_ISSUE].north:after{border-top-color:#bf3600}.tip>div[data-id$=_ISSUE].south:after{border-bottom-color:#bf3600}@media (min-width:650px){.tip>div{max-width:30%}}@media (min-width:850px){.tip>div{max-width:25%}}@media (min-width:1100px){.tip>div{max-width:20%}}body .icon.flag{color:inherit;transition:color .5s ease .5s}body .icon.flag:after{width:0;transition:width .5s ease .5s}body.withSuggestions .icon.flag{color:#bf3600}body.withSuggestions .icon.flag:after{width:1em}body article#suggestion{position:relative;padding:1.5em;margin-bottom:2.5em;margin-left:1.5em;border:2px solid gray;border-radius:4px;width:0%;max-height:0;opacity:.5;float:right;background-color:#fff;transition:max-height .75s ease 0s,width .5s ease .25s}body article#suggestion::after{content:"";position:absolute;width:0;height:0;left:10px;bottom:-10px;border-bottom:10px solid transparent;border-right:25px solid gray;transition:left .5s ease 0s,bottom .5s ease 0s,border-bottom-width .5s ease 0s,border-right-width .5s ease 0s,border-color .5s ease 0s}body article#suggestion section.aspect_map,body article#suggestion section[data-id=SUGGESTION_TEXT]{opacity:0;overflow:hidden;transition:opacity .5s ease 0s}body.withSuggestions article#suggestion{opacity:1;width:85%;max-height:700px;border-color:#0071c5;transition:max-height .5s ease .25s,width .5s ease 0s,border-color .5s ease 0s}body.withSuggestions article#suggestion::after{left:20px;bottom:-20px;border-bottom-width:20px;border-right-width:80px;border-right-color:#0071c5}body.withSuggestions article#suggestion section.aspect_map,body.withSuggestions article#suggestion section[data-id=SUGGESTION_TEXT]{opacity:1;transition:opacity .25s ease .5s}@media (min-width:625px){body.withSuggestions article#suggestion{width:50%}}a.button.suggestion{position:absolute;right:.9em;top:.75em}section[data-id=SUGGESTION_TEXT]{margin-bottom:1.5em}section[data-id=SUGGESTION_TEXT] span#first_line{font-size:150%;color:#0071c5}section.aspect_map{width:100%;display:table}section.aspect_map>section.aspect{display:table-row;height:1.6em}section.aspect_map>section.aspect>span{display:table-cell;vertical-align:middle;white-space:nowrap;padding-left:.5em}section.aspect_map>section.aspect>span:first-child{display:inline-block;padding-left:0}section.aspect_map>section.aspect>span:nth-child(2){text-align:right}section.aspect_map>section.aspect>span:not(.icon):nth-child(2){padding-right:1em}section.aspect_map>section.aspect>span:nth-child(3){padding-right:1em}section.aspect_map>section.aspect>span:last-child{position:relative;width:60%}section.aspect_map>section.aspect>span:last-child>span:not(.tooltipped){position:absolute;left:0;right:100%;top:.5em;height:.6em;background-color:#bf3600}section.aspect_map>section.aspect:first-child{color:gray;text-align:center;font-size:.75em}article[data-id=INFO]{color:gray;margin-bottom:20px;padding-right:2em;font-size:80%}article[data-id=INFO] span[data-id]:last-child{font-style:italic;word-wrap:break-word;word-break:break-all;padding-left:.5em;color:#000}h1{font-size:2.5em;color:rgba(128,128,128,.5);margin:0 0 20px;display:none}h1 sup{display:block;font-size:.4em;font-weight:400;line-height:.8em}@media (min-width:850px){h1{display:block}}article[data-id=CPI_RATE],article[data-id=GFLOPS],article[data-id=WALLCLOCK]{float:left;font-size:1em;font-weight:400;margin-bottom:1.5em;padding-right:10%}article[data-id=CPI_RATE]>span[data-id]:first-child,article[data-id=GFLOPS]>span[data-id]:first-child,article[data-id=WALLCLOCK]>span[data-id]:first-child{display:block;padding-top:.75em;font-size:2.2em;font-weight:500;color:#00aeef}div#content article[data-id=WALLCLOCK]>span[data-id]:first-child{font-size:3.4em;padding-top:0}div#bricks article{color:gray}div#bricks article h3{margin:0;font-size:1.5em;font-weight:400;color:#00aeef}div#bricks article h4{margin:.8em 0 0;padding:0;font-size:1em;font-weight:400;color:#000}div#bricks article .indent{margin-left:1.75em}div#bricks article table{margin-top:.8em;width:100%;border-collapse:collapse}div#bricks article table tr{border-bottom:1px solid rgba(128,128,128,.25)}div#bricks article table tr td,div#bricks article table tr th{text-align:right}div#bricks article table tr td:first-child,div#bricks article table tr th:first-child{text-align:left;white-space:normal}div#bricks article table tr td{padding:.2em;position:relative;overflow:hidden;white-space:nowrap}div#bricks article table tr th{font-weight:400;padding:.2em 0;border-bottom:2px solid rgba(128,128,128,.25)}div#bricks article table tr th:first-child{color:#000}div#bricks article table tr .progress{position:absolute;width:0%;height:100%;background:#00aeef;opacity:.15;left:0;top:0;-webkit-animation:filler 4s ease-in-out;-moz-animation:filler 4s ease-in-out;animation:filler 4s ease-in-out}div[data-id=MEMORY_INFO]>div,div[data-id=RSS_INFO]>div{margin-left:1.5em}div[data-id=MEMORY_INFO]>span,div[data-id=RSS_INFO]>span{margin-left:1em}</style><script>var json={"INFO":{"APPLICATION_NAME":"ex_modify","NUM_RANKS":"64","RANKS_PER_NODE":"64","STAT_FILES":"/blues/gpfs/home/sajid/packages/xwp_petsc/2d/matter_repeat/aps_result_20190227","CREATION_DATETIME":"2019-02-27 15:54:29","NUM_LOGIC_CORE":"272","HW_PLATFORM":"Intel(R) Processor code named Knights Landing","COLLECTOR_TYPE":"Driverless Perf system-wide counting"},"FPU_UTILIZATION":null,"FLOPS_PER_CYCLE":null,"FLOPS_OUT_OF":null,"VECTOR_CAPACITY":null,"FP_PACKED":null,"FP_128_BIT":null,"FP_256_BIT":null,"FP_512_BIT":null,"SCALAR_FP":null,"FP_ARITH_MEM_RD":null,"FP_ARITH_MEM_WR":null,"CACHE_BOUND":null,"DRAM_BOUND":null,"NUMA_USE":null,"DRAM_BANDWIDTH":{"DRAM_BANDWIDTH_VAL":"23.60","DRAM_BANDWIDTH_UNIT":" GB/s","DRAM_BANDWIDTH_DESCR":"Average amount of data transferred through DRAM memory controller per second."},"MCDRAM_BANDWIDTH":{"MCDRAM_BANDWIDTH_VAL":"0.03","MCDRAM_BANDWIDTH_UNIT":" GB/s","MCDRAM_BANDWIDTH_DESCR":"Average amount of data transferred through MCDRAM memory controller per second."},"CPU_USAGE":null,"AVERAGE_CPU_USE":null,"CPU_OUT_OF":null,"PHYSICAL_CORE_USAGE":null,"PH_CORE_USE":null,"PH_CORE_OUT_OF":null,"BACK_END_BOUND":{"BACK_END_BOUND_PERC":"60.70%","BACK_END_BOUND_DESCR":"Superscalar processors can be conceptually divided into the `front-end`, where instructions are fetched and decoded into the operations that constitute them and the `back-end`, where the required computation is performed. During each cycle, the front-end generates up to two of these operations, places them into pipeline slots and moves them through the back-end. The actual number of retired pipeline slots containing useful work rarely equals this maximum. This can be because the back-end was not prepared to accept more operations of a certain kind (`Back-end bound` execution). Back-end bound execution may be due to long-latency operations or other contention for execution resources like too many operations being directed to a single execution port.","BACK_END_BOUND_ISSUE":"A significant proportion of pipeline slots remain empty. When operations take too long in the back-end, they introduce bubbles in the pipeline that ultimately cause fewer pipeline slots containing useful work to be retired per cycle than the machine is capable of supporting. This opportunity cost results in slower execution. Long-latency operations like division and memory operations can cause this, as can too many operations being directed to a single execution port (for example, more multiplication operations arriving in the back-end per cycle than the execution unit can support). Explore second level metrics or use <a href='https://software.intel.com/en-us/intel-vtune-amplifier-xe' target='_blank'>Intel® VTune™ Amplifier</a> <a href='https://software.intel.com/en-us/vtune-amplifier-help-memory-access-analysis' target='_blank'>Memory Access analysis</a> to learn more."},"L2_HIT_BOUND":{"L2_HIT_BOUND_PERC":"2.10%","L2_HIT_BOUND_DESCR":"Percentage of CPU cycles are being spent on data fetches that miss the L1 but hit the L2. The cycles spent handling L2 hits are calculated as L2 CACHE HIT COST * L2 CACHE HIT COUNT where L2 CACHE HIT COST is a constant measured as typical L2 access latency in cycles. This metric includes coherence penalties for shared data."},"L2_MISS_BOUND":{"L2_MISS_BOUND_PERC":"3.80%","L2_MISS_BOUND_DESCR":"Percentage of CPU cycles are being spent waiting for L2 load misses to be serviced. Any memory requests missing here must be serviced by local or remote DRAM or MCDRAM, with significant latency. The cycles spent handling L2 misses are calculated as L2 CACHE MISS COST * L2 CACHE MISS COUNT where L2 CACHE MISS COST is a constant measured as typical DRAM access latency in cycles."},"SIMD_PER_CYCLE":{"SIMD_PER_CYCLE_VAL":"0.03","SIMD_PER_CYCLE_DESCR":"This metric represents how intensively your program uses the FPU. Fused multiply-add (FMA) is counted as two instructions.","SIMD_PER_CYCLE_ISSUE":"The metric value indicates that FPU might be underutilized. This can be a result of significant fraction of non-floating point instructions, inefficient vectorization because of legacy vector instruction set or memory access pattern issues, or different kinds of stalls in the code execution. Explore second level metrics to identify the next steps in FPU usage improvements."},"PACKED_SIMD":{"PACKED_SIMD_PERC":"99.20%","PACKED_SIMD_DESCR":"This metric represents the percentage of all packed floating point instructions."},"SCALAR_SIMD":{"SCALAR_SIMD_PERC":"0.80%","SCALAR_SIMD_DESCR":"The metric represents the percentage of scalar SIMD instructions."},"CPI_RATE":{"CPI_RATE_SEVERITY":"severityCritical","CPI_RATE_VAL":"5.88","CPI_RATE_DESCR":"Cycles per Instruction Retired, or CPI, is a fundamental performance metric indicating approximately how much time each executed instruction took, in units of cycles. Modern superscalar processors issue up to four instructions per cycle, suggesting a theoretical best CPI of 0.25. But various effects (long-latency memory, floating-point, or SIMD operations; non-retired instructions due to branch mispredictions; instruction starvation in the front-end) tend to pull the observed CPI up. A CPI of 1 is generally considered acceptable for HPC applications, but different application domains will have very different expected values.","CPI_RATE_ISSUE":"The CPI value may be <i>too high</i>.<br> This could be caused by such issues as memory stalls, instruction starvation, branch misprediction, or long latency instructions.<br> Use <a href='https://software.intel.com/en-us/intel-vtune-amplifier-xe' target='_blank'>Intel® VTune™ Amplifier</a> General Exploration analysis to specify particular reasons of high CPI.","CPI_RATE_MIN_VAL":"5.88","CPI_RATE_MIN_NODE":"apsxrmd-0001","CPI_RATE_MAX_VAL":"5.88","CPI_RATE_MAX_NODE":"apsxrmd-0001"},"GFLOPS":null,"WALLCLOCK":{"WALLCLOCK_SEVERITY":"severityUnknown","WALLCLOCK_VAL":"451.84s","WALLCLOCK_PERC":"0.00%","WALLCLOCK_DESCR":"Total application wall-time."},"MPI_TIME":{"MPI_TIME_SEVERITY":"severityCritical","MPI_TIME_VAL":"287.31s","MPI_TIME_PERC":"63.59%","MPI_TIME_DESCR":"Time spent inside the MPI library. Values more than 10% might need additional exploration on MPI communication efficiency. This might be caused by high wait times inside the library, active communications, non-optimal settings of the MPI library. See MPI Imbalance metric to see if the application has load balancing problem. Use <a href='https://software.intel.com/en-us/intel-trace-analyzer' target='_blank'>Intel® Trace Analyzer and Collector</a> to explore communication efficiency.","MPI_TIME_ISSUE":"Your application is MPI bound. This may be caused by high busy wait time inside the library (imbalance), non-optimal communication schema or MPI library settings. Explore the MPI Imbalance metric if it is available or use <a href='https://software.intel.com/en-us/get-started-with-itac' target='_blank'>MPI profiling tools</a> like <a href='https://software.intel.com/en-us/intel-trace-analyzer' target='_blank'>Intel® Trace Analyzer and Collector</a> to explore possible performance bottlenecks."},"MPI_IMBALANCE":{"MPI_IMBALANCE_SEVERITY":"severityUnknown","MPI_IMBALANCE_VAL":"142.00s","MPI_IMBALANCE_PERC":"31.43%","MPI_IMBALANCE_DESCR":"Mean unproductive wait time per process spent in the MPI library calls when a process is waiting for data.","MPI_IMBALANCE_ISSUE":"The application workload is <i>not well balanced</i> between MPI ranks.For more details about the MPI communication scheme use <a href='https://software.intel.com/en-us/intel-trace-analyzer' target='_blank'>Intel® Trace Analyzer and Collector</a> available as part of <a href='https://software.intel.com/en-us/intel-parallel-studio-xe' target='_blank'>Intel® Parallel Studio Cluster Edition</a>."},"NON_MPI_TIME":null,"OMP_REGIONS_TIME":null,"OMP_IMBALANCE_TIME":null,"OMP_SERIAL_TIME":null,"MEMORY":{"PEAK_MEM_RANK":"Rank: 1","PEAK_MEM_VAL":"1448.33 MB","MEAN_MEM_VAL":"1448.22 MB","MEM_DESCR":"Per-process memory usage affects the application scalability."},"RSS":{"PEAK_RSS_RANK":"Rank: 47","PEAK_RSS_VAL":"706.03 MB","MEAN_RSS_VAL":"700.27 MB","RSS_DESCR":"RSS usage affects the application scalability."},"MEMORY_PER_NODE":{"PEAK_MEM_NODE":"Node: apsxrmd-0001","PEAK_MEM_PER_NODE_VAL":"92686.37 MB","MEAN_MEM_PER_NODE_VAL":"92686.37 MB","MEM_PER_NODE_DESCR":"Per-node memory usage affects the application scalability."},"RSS_PER_NODE":{"PEAK_RSS_NODE":"Node: apsxrmd-0001","PEAK_RSS_PER_NODE_VAL":"44817.16 MB","MEAN_RSS_PER_NODE_VAL":"44817.16 MB","RSS_PER_NODE_DESCR":"Per-node RSS usage affects the application scalability."},"TOP_5":{"Allreduce":"52.31","File_set_view":"5.97","Iprobe":"1.31","Test":"1.27","Waitall":"1.00"},"MEMORY_BOUND":null,"IO":{"READ_BYTES_SEVERITY":"severityUnknown","READ_BYTES_DESCR":"Amount of bytes read by the application. High volumes of I/O operations may lead to unproductive wait.","WRITTEN_BYTES_SEVERITY":"severityUnknown","WRITTEN_BYTES_DESCR":"Amount of bytes written by the application. High volumes of I/O operations may lead to unproductive wait.","IOWAIT_USEC_SEVERITY":"severityLow","IOWAIT_USEC_VAL":"0.04s","IOWAIT_USEC_PERC":"0.01%","IOWAIT_USEC_DESCR":"This is the time the application spends waiting for an I/O operation to complete.","READ_BYTES_AVG_VAL":" 8.6 MB","READ_BYTES_MAX_VAL":" 261.5 MB","READ_BYTES_MAX_RANK":"Rank: 0","WRITTEN_BYTES_AVG_VAL":" 8.5 MB","WRITTEN_BYTES_MAX_VAL":" 262.0 MB","WRITTEN_BYTES_MAX_RANK":"Rank: 0","IOWAIT_USEC_AVG_VAL":"0.04","IOWAIT_USEC_AVG_PERC":"0.01%","IOWAIT_USEC_MAX_VAL":"0.06","IOWAIT_USEC_MAX_PERC":"0.01%","IOWAIT_USEC_MAX_RANK":"Rank: 18"},"SUGGESTION":{"TEXT":"<span id='first_line'>Your application is MPI bound.</span><br> This may be caused by high busy wait time inside the library (imbalance), non-optimal communication schema or MPI library settings. Use <a href='https://software.intel.com/en-us/get-started-with-itac' target='_blank'>MPI profiling tools</a> like <a href='https://software.intel.com/en-us/intel-trace-analyzer' target='_blank'>Intel® Trace Analyzer and Collector</a> to explore performance bottlenecks.","STYLE":"bad"}};</script><script>function feedbackInit(){document.querySelectorAll(".feedback-panel").forEach(function(e){e.classList.remove("closed"),e.querySelector(".question").innerHTML="Like this report?"})}function feedback(e){var t=document.querySelectorAll(".feedback-panel");NodeList.prototype.forEach=Array.prototype.forEach,t.forEach(function(e){e.querySelector(".question").innerHTML="Thank you",e.classList.add("closed")}),sendAnalytics(e)}function isBrowserSupported(){var e,t,o={"Trident/":{version:11,offset:3},"Edge/":{version:12,offset:5},Firefox:{version:17,offset:8},Chrome:{version:40,offset:7},Safari:{version:8,offset:7}},r=(navigator.appVersion,navigator.userAgent),s=(navigator.appName,""+parseFloat(navigator.appVersion)),n=parseInt(navigator.appVersion,10);for(var i in o)if(-1!=(e=r.indexOf(i))&&("Trident/"==i&&(e=r.indexOf("rv:")),s=r.substring(e+o[i].offset),-1!=(t=s.indexOf(";"))&&(s=s.substring(0,t)),-1!=(t=s.indexOf(" "))&&(s=s.substring(0,t)),-1!=(t=s.indexOf(")"))&&(s=s.substring(0,t)),n=parseInt(""+s,10),isNaN(n)&&(s=""+parseFloat(navigator.appVersion),n=parseInt(navigator.appVersion,10)),n>=o[i].version))return!0;return!1}function renderForNotSupportedBrowsers(){var e=document.querySelectorAll("body > div:not(.not-supported-browser)");for(var t in e)e[t].parentNode&&e[t].parentNode.removeChild(e[t])}function renderForSupportedBrowsers(){var e=document.querySelector("body > div.not-supported-browser");e.parentNode&&e.parentNode.removeChild(e)}function masonryLayout(e){function t(){var e=function(){var e=document.body.clientWidth,t=s.columns;for(var o in s.breakAt)if(e<o){t=s.breakAt[o];break}return t}();if(1===e)r.style.height="auto",o(r.children,function(e,t){t.removeAttribute("style"),t.style.padding=s.padding+"px",t.style.marginBottom=s.gap+"px"});else{var t=function(e){return"calc("+100/e+"% - "+(2*s.padding+(e-1)*s.gap/e)+"px)"}(e),n=function(e,t){for(var o=[],r=0;r<e;r++)o.push(0);return o}(e);o(r.children,function(o,r){r.style.position="absolute",r.style.width=t,r.style.left=function(e,t){if(0===e)return"0";return"calc("+100/t*e+"% + "+s.gap*e/t+"px)"}(o%e,e),r.style.top=n[o%e]+"px",r.style.padding=s.padding+"px",n[o%e]+=function(e){var t=parseInt(window.getComputedStyle(e,null).getPropertyValue("height"),10);return isNaN(t)?0:t+s.gap+2*s.padding}(r)}),r.style.height=Math.max.apply(null,n)+"px"}}function o(e,t){for(var o=0,r=0,s=e.length;o<s;o++)"none"!==window.getComputedStyle(e[o],null).getPropertyValue("display")&&(t(r,e[o]),r++)}var r=e,s={columns:4,gap:15,padding:10,breakAt:{1100:3,850:2,625:1}};window.addEventListener("resize",t),r.style.position="relative",t()}function IsDataReady(){return"undefined"!=typeof json&&"{JSON}"!==json}function getDataFromNexus(){for(var e="",t="",o=document.location.search.slice(1).split("&"),r=0;r<o.length;r++){var s=o[r].split("=");"nexus"===s[0]&&s[1]&&(e="http://"+s[1]),"path"===s[0]&&(t=s[1])}var n=e+"/repository"+t+"/data/obj.json",i=new XMLHttpRequest;i.open("GET",n,!0),i.onload=function(e){4===i.readyState&&200==i.status&&(json=JSON.parse(i.responseText),setUpAndRender())},i.send(null)}function setUpAndRender(){isBrowserSupported()?(renderForSupportedBrowsers(),sendAnalytics("pageOpen"),menu(),render(json),masonryLayout(document.querySelector("#bricks"))):renderForNotSupportedBrowsers()}function menu(){function e(){o.classList.add("menuOpened"),feedbackInit()}var t,o=document.body,r=document.getElementById("topMenuClosedState"),s=document.getElementById("leftMenuClosedState");r.onmouseenter=s.onmouseenter=function(){t=setTimeout(e,800)},r.onmouseleave=s.onmouseleave=function(){clearTimeout(t)},r.onclick=s.onclick=e,document.getElementById("content").onmouseenter=function(){o.classList.remove("menuOpened")}}function drawProgress(){for(var e=document.querySelectorAll(".progress"),t=0;t<e.length;t++)e[t].style.width=e[t].dataset.progress+"%"}function drawThresholdBars(e){var t={BACK_END_BOUND:{threshold:20,lessIsBetter:!0,perc:!0},CACHE_BOUND:{threshold:20,lessIsBetter:!0,perc:!0},CPI_RATE:{threshold:1,lessIsBetter:!0,perc:!1},CPU_USAGE:{threshold:90,lessIsBetter:!1,perc:!0},PHYSICAL_CORE_USAGE:{threshold:80,lessIsBetter:!1,perc:!0},DRAM_BOUND:{threshold:20,lessIsBetter:!0,perc:!0},FP_128_BIT:{threshold:5,lessIsBetter:!0,perc:!0},FP_256_BIT:{threshold:5,lessIsBetter:!0,perc:!0},FP_ARITH_MEM_RD:{threshold:.5,lessIsBetter:!1,perc:!1},FP_ARITH_MEM_WR:{threshold:.5,lessIsBetter:!1,perc:!1},FPU_UTILIZATION:{threshold:50,lessIsBetter:!1,perc:!0},IO:{threshold:10,lessIsBetter:!0,perc:!0},L2_HIT_BOUND:{threshold:10,lessIsBetter:!0,perc:!0},L2_MISS_BOUND:{threshold:15,lessIsBetter:!0,perc:!0},MEMORY_BOUND:{threshold:20,lessIsBetter:!0,perc:!0},MPI_IMBALANCE:{threshold:30,lessIsBetter:!0,perc:!1},MPI_TIME:{threshold:10,lessIsBetter:!0,perc:!0},NUMA_USE:{threshold:15,lessIsBetter:!0,perc:!0},OMP_IMBALANCE_TIME:{threshold:10,lessIsBetter:!0,perc:!0},OMP_SERIAL_TIME:{threshold:15,lessIsBetter:!0,perc:!0},SCALAR_FP:{threshold:15,lessIsBetter:!0,perc:!0},SCALAR_SIMD:{threshold:15,lessIsBetter:!0,perc:!0},SIMD_PER_CYCLE:{threshold:1,lessIsBetter:!1,perc:!1},VECTOR_CAPACITY:{threshold:50,lessIsBetter:!1,perc:!0}};null!=e.BACK_END_BOUND&&(t.CPI_RATE.threshold=2);var o,r,s,n=["BACK_END_BOUND","CPU_USAGE","PHYSICAL_CORE_USAGE","FPU_UTILIZATION","IO","MEMORY_BOUND","MPI_TIME","OMP_IMBALANCE_TIME","OMP_SERIAL_TIME","SIMD_PER_CYCLE"];for(var i in t)if(null!=e[i]){var l=n.indexOf(i)>-1;if(s=t[i].perc?"_PERC":"_VAL",o=parseFloat(e[i][i+s]),"MPI_IMBALANCE"==i){var a=parseFloat(e.MPI_TIME.MPI_TIME_VAL);o=a>0?100*o/a:0,s="_PERC"}if(null!=(r=l?document.querySelector("[data-id="+i+"_BAR]"):document.querySelector("[data-id="+i+s+"]"))&&(t[i].lessIsBetter&&o>=t[i].threshold||!t[i].lessIsBetter&&o<=t[i].threshold)){if("FP_256_BIT"==i&&null==e.FP_512_BIT)continue;if(l&&(r.style.right=95-95/90*Math.abs(o-t[i].threshold)+"%"),null!=(r=document.querySelectorAll("[data-id="+i+s+"]")))for(var d=0;d<r.length;d++)r[d].classList.add("icon"),r[d].classList.add("flag"),null!=e[i][i+"_ISSUE"]&&r[d].classList.add("tooltipped")}}}function makeInvisible(e){var t=document.querySelectorAll(e);if(null!=t)for(var o=0;o<t.length;o++)t[o].style.display="none"}function render(e){var t;if(null!=e.SUGGESTION&&(null!=(t=document.querySelector("[data-id=SUGGESTION_TEXT]"))&&(t.innerHTML=e.SUGGESTION.TEXT),null!=(t=document.getElementById("suggestion"))&&t.classList.add(e.SUGGESTION.STYLE)),null!=(t=document.querySelectorAll("[data-id=INFO] span")))for(var o=0;o<t.length;o++)null==e.INFO[t[o].dataset.id]&&makeInvisible("div[data-id="+t[o].dataset.id+"_STR]");for(var r in e){var s=e[r];if(null==s)makeInvisible("[data-id="+r+"],[data-id="+r+"_TITLE],[data-id="+r+"_VAL],[data-id="+r+"_PERC]");else if("TOP_5"==r){var n=document.querySelector("[data-id=TOP_5]");for(var i in s){var l=document.createElement("tr");l.innerHTML="<td><span class='progress' data-progress='"+s[i]+"'></span>"+i+"</td><td>"+s[i]+"</td>",n.appendChild(l)}drawProgress()}else for(var a in s)if(null!=(t=document.querySelectorAll("[data-id="+a+"]")))for(o=0;o<t.length;o++)t[o].innerHTML=s[a]}drawThresholdBars(e);for(var d=["DRAM_BANDWIDTH","MCDRAM_BANDWIDTH"],u=0;u<d.length;u++)if(null!=e[d[u]])for(t=document.querySelectorAll("[data-id="+d[u]+"_VAL]"),o=0;o<t.length;o++)t[o].classList.remove("tooltipped"),t[o].dataset.tipId=d[u]+"_ISSUE";else if("DRAM_BANDWIDTH"==d[u]||null!=e.INFO.HW_PLATFORM&&e.INFO.HW_PLATFORM.indexOf("Knight")>-1)for(t=document.querySelectorAll("[data-id="+d[u]+"_TITLE],[data-id="+d[u]+"_VAL]"),o=0;o<t.length;o++)t[o].style.display="";if(null!=(t=document.querySelectorAll(".tip > div[data-id$=_ISSUE]")))for(o=0;o<t.length;o++)if(""==t[o].innerHTML||void 0==t[o].innerHTML){var c=document.querySelectorAll("[data-tip-id="+t[o].dataset.id+"].tooltipped");if(null!=c)for(var p=0;p<c.length;p++)c[p].classList.remove("tooltipped")}else{var f=document.querySelectorAll("[data-tip-id="+t[o].dataset.id+"]:not(.tooltipped)");if(null!=f)for(p=0;p<f.length;p++)f[p].classList.add("icon"),f[p].classList.add("flag"),f[p].classList.add("tooltipped")}makeInvisible(null==e.MPI_TIME?"[data-id$=_INFO],[data-id=NUM_RANKS_STR],[data-id=RANK_STR],[data-id=MEMORY_DATA_BLOCK]>table":"[data-id=MEMORY_APP],[data-id=RSS_APP]"),makeInvisible(null!=e.PCONTROL&&1==e.PCONTROL?"[data-id$=_DATA_BLOCK],section[data-id=IO]":"[data-id$=_TEXT_BLOCK]"),initTooltips()}function sendAnalytics(e){}function suggestionShow(){document.body.classList.add("withSuggestions");var e=document.querySelector("a.button.suggestion");e.classList.remove("arrow-s-w"),e.classList.add("cross"),e.onclick=suggestionHide,initTooltips(),sendAnalytics("suggestionShow")}function suggestionHide(){document.body.classList.remove("withSuggestions");var e=document.querySelector("a.button.suggestion");e.classList.remove("cross"),e.classList.add("arrow-s-w"),e.onclick=suggestionShow,removeTooltips(),sendAnalytics("suggestionHide")}function initTooltips(){function e(e){if(void 0!==e){var o=document.querySelector(".tip > div[data-id="+e.dataset.tipId+"]");if(null!=o){o.classList.add("visible");var r=e.getBoundingClientRect(),n=o.getBoundingClientRect();window.innerHeight-r.top-n.height>0?(o.classList.add("south"),o.style.top=r.bottom+8+"px"):(o.classList.add("north"),o.style.top=r.top-n.height-8+"px"),window.innerWidth-r.left-n.width>0?(o.classList.add("west"),o.style.left=r.left+"px"):(o.classList.add("east"),o.style.right=window.innerWidth-r.right+"px"),o.onmouseenter=function(){clearTimeout(s)},o.onmouseleave=function(){t(e)}}}}function t(e){if(void 0!==e){var t=document.querySelector(".tip > div[data-id="+e.dataset.tipId+"]");null!=t&&(t.removeAttribute("style"),t.classList.remove("visible","north","south","east","west"))}}for(var o=document.querySelectorAll(".tooltipped[data-tip-id]"),r=0;r<o.length;r++){var s,n,i,l=o[r];l.onmouseenter=function(){clearTimeout(s),void 0!==i&&t(i),i=this,n=setTimeout(e,200,this)},l.onmouseleave=function(){clearTimeout(n),s=setTimeout(t,800,this)}}}function removeTooltips(){for(var e=document.querySelectorAll(".tooltipped[data-tip-id].icon.flag"),t=0;t<e.length;t++)e[t].onmouseenter=e[t].onmouseleave=null}window.onload=function(){NodeList.prototype.forEach=Array.prototype.forEach,setUpAndRender()};</script></script></head><body class="withSuggestions"><div id="leftMenuOpenedState" class="anim"><h4>Additional Performance Analysis Tools:</h4><a href="https://software.intel.com/en-us/intel-trace-analyzer">Intel® Trace Analyzer and Collector</a><div>MPI Analyzer and Profiler</div><a href="https://software.intel.com/en-us/intel-vtune-amplifier-xe">Intel® VTune™ Amplifier</a><div>Performance Profiler</div><a href="https://software.intel.com/en-us/intel-advisor-xe">Intel® Advisor</a><div>Vectorization Optimization & Thread Prototyping</div><a href="http://www.intel.com/storage-snapshot">Storage Performance Snapshot</a><div>Visualize System Storage Bottlenecks</div><section class="feedback-panel left"><div class="question"></div><a class="button like" href="#" onclick="feedback('apsLike');">Yes</a> <a class="button dislike" href="#" onclick="feedback('apsDislike');">No</a> <a class="button icon email" href="mailto:parallel.studio.support@intel.com?Subject=MPI%20Performance%20Snapshot:%20feedback" target="_top"></a></section></div><div id="leftMenuClosedState" class="anim"><div class="icon menu"></div><div class="icon intel"></div></div><div id="topMenuOpenedState" class="anim"><h4>Additional Performance Analysis Tools:</h4><a href="https://software.intel.com/en-us/intel-trace-analyzer">Intel® Trace Analyzer and Collector</a> <span>- MPI Analyzer and Profiler</span><br><a href="https://software.intel.com/en-us/intel-vtune-amplifier-xe">Intel® VTune™ Amplifier</a> <span>- Performance Profiler</span><br><a href="https://software.intel.com/en-us/intel-advisor-xe">Intel® Advisor</a> <span>- Vectorization Optimization & Thread Prototyping</span><br><a href="http://www.intel.com/storage-snapshot">Storage Performance Snapshot</a> <span>- Visualize System Storage Bottlenecks</span><section class="feedback-panel top"><div class="question"></div><a class="button like" href="#" onclick="feedback('apsLike');">Yes</a> <a class="button dislike" href="#" onclick="feedback('apsDislike');">No</a> <a class="button icon email" href="mailto:parallel.studio.support@intel.com?Subject=MPI%20Performance%20Snapshot:%20feedback" target="_top"></a></section></div><div id="topMenuClosedState" class="anim"><span class="icon intel"></span> <span class="icon menu"></span> <sup>Intel® VTune™ Amplifier</sup> Application Performance Snapshot</div><div id="contentWrap" class="anim"><div id="content"><h1><sup>Intel® VTune™ Amplifier</sup>Application Performance Snapshot</h1><article id="suggestion"><a class="button suggestion icon cross" href="#" onclick="suggestionHide();"></a><section data-id="SUGGESTION_TEXT"></section><section class="aspect_map"><section class="aspect"><span></span> <span class="aspect_value"><span class="tooltipped" data-tip-id="current_run">Current run</span></span> <span class="aspect_target"><span class="tooltipped" data-tip-id="target">Target</span></span> <span><span class="tooltipped" data-tip-id="delta">Delta</span></span></section><section class="aspect" data-id="MPI_TIME"><span class="tooltipped" data-id="MPI_TIME_TITLE" data-tip-id="MPI_TIME_DESCR">MPI Time</span> <span class="aspect_value" data-id="MPI_TIME_PERC" data-tip-id="MPI_TIME_ISSUE"></span> <span class="aspect_target"><10%</span> <span class="aspect_bar"><span data-id="MPI_TIME_BAR"></span></span></section><section class="aspect" data-id="OMP_SERIAL_TIME"><span class="tooltipped" data-id="OMP_SERIAL_TIME_TITLE" data-tip-id="OMP_SERIAL_TIME_DESCR">Serial Time</span> <span class="aspect_value" data-id="OMP_SERIAL_TIME_PERC" data-tip-id="OMP_SERIAL_TIME_ISSUE"></span> <span class="aspect_target"><15%</span> <span class="aspect_bar"><span data-id="OMP_SERIAL_TIME_BAR"></span></span></section><section class="aspect" data-id="OMP_IMBALANCE_TIME"><span class="tooltipped" data-id="OMP_IMBALANCE_TIME_TITLE" data-tip-id="OMP_IMBALANCE_TIME_DESCR">OpenMP Imbalance</span> <span class="aspect_value" data-id="OMP_IMBALANCE_TIME_PERC" data-tip-id="OMP_IMBALANCE_TIME_ISSUE"></span> <span class="aspect_target"><10%</span> <span class="aspect_bar"><span data-id="OMP_IMBALANCE_TIME_BAR"></span></span></section><section class="aspect" data-id="CPU_USAGE"><span class="tooltipped" data-id="CPU_USAGE_TITLE" data-tip-id="CPU_USAGE_DESCR">CPU Utilization</span> <span class="aspect_value" data-id="CPU_USAGE_PERC" data-tip-id="CPU_USAGE_ISSUE"></span> <span class="aspect_target">>90%</span> <span class="aspect_bar"><span data-id="CPU_USAGE_BAR"></span></span></section><section class="aspect" data-id="PHYSICAL_CORE_USAGE"><span class="tooltipped" data-id="PHYSICAL_CORE_USAGE_TITLE" data-tip-id="PHYSICAL_CORE_USAGE_DESCR">Physical Core Utilization</span> <span class="aspect_value" data-id="PHYSICAL_CORE_USAGE_PERC" data-tip-id="PHYSICAL_CORE_USAGE_ISSUE"></span> <span class="aspect_target">>80%</span> <span class="aspect_bar"><span data-id="PHYSICAL_CORE_USAGE_BAR"></span></span></section><section class="aspect" data-id="MEMORY_BOUND"><span class="tooltipped" data-id="MEMORY_BOUND_TITLE" data-tip-id="MEMORY_BOUND_DESCR">Memory Stalls</span> <span class="aspect_value" data-id="MEMORY_BOUND_PERC" data-tip-id="MEMORY_BOUND_ISSUE"></span> <span class="aspect_target"><20%</span> <span class="aspect_bar"><span data-id="MEMORY_BOUND_BAR"></span></span></section><section class="aspect" data-id="BACK_END_BOUND"><span class="tooltipped" data-id="BACK_END_BOUND_TITLE" data-tip-id="BACK_END_BOUND_DESCR">Back-End Stalls</span> <span class="aspect_value" data-id="BACK_END_BOUND_PERC" data-tip-id="BACK_END_BOUND_ISSUE"></span> <span class="aspect_target"><20%</span> <span class="aspect_bar"><span data-id="BACK_END_BOUND_BAR"></span></span></section><section class="aspect" data-id="FPU_UTILIZATION"><span class="tooltipped" data-id="FPU_UTILIZATION_TITLE" data-tip-id="FPU_UTILIZATION_DESCR">FPU Utilization</span> <span class="aspect_value" data-id="FPU_UTILIZATION_PERC" data-tip-id="FPU_UTILIZATION_ISSUE"></span> <span class="aspect_target">>50%</span> <span class="aspect_bar"><span data-id="FPU_UTILIZATION_BAR"></span></span></section><section class="aspect" data-id="SIMD_PER_CYCLE"><span class="tooltipped" data-id="SIMD_PER_CYCLE_TITLE" data-tip-id="SIMD_PER_CYCLE_DESCR">SIMD Instr. per Cycle</span> <span class="aspect_value" data-id="SIMD_PER_CYCLE_VAL" data-tip-id="SIMD_PER_CYCLE_ISSUE"></span> <span class="aspect_target">>1</span> <span class="aspect_bar"><span data-id="SIMD_PER_CYCLE_BAR"></span></span></section><section class="aspect" data-id="IO"><span class="tooltipped" data-id="IO_TITLE" data-tip-id="IOWAIT_USEC_DESCR" class="tooltipped">I/O Bound</span> <span class="aspect_value" data-id="IOWAIT_USEC_PERC"></span> <span class="aspect_target"><10%</span> <span class="aspect_bar"><span data-id="IO_BAR"></span></span></section></section></article><article data-id="INFO"><div data-id="APPLICATION_NAME_STR">Application: <span data-id="APPLICATION_NAME"></span></div><div data-id="CREATION_DATETIME_STR">Report creation date: <span data-id="CREATION_DATETIME"></span></div><div data-id="RANK_STR">Rank: <span data-id="RANK"></span></div><div data-id="NUM_RANKS_STR">Number of ranks: <span data-id="NUM_RANKS"></span></div><div data-id="RANKS_PER_NODE_STR">Ranks per node: <span data-id="RANKS_PER_NODE"></span></div><div data-id="NUM_OMP_THREADS_STR">OpenMP threads<span data-id="POSTFIX"></span>: <span data-id="NUM_OMP_THREADS"></span></div><div data-id="HW_PLATFORM_STR">HW Platform: <span data-id="HW_PLATFORM"></span></div><div data-id="NUM_LOGIC_CORE_STR">Logical Core Count per node: <span data-id="NUM_LOGIC_CORE"></span></div><div data-id="COLLECTOR_TYPE_STR">Collector type: <span data-id="COLLECTOR_TYPE"></span></div></article><article data-id="WALLCLOCK"><span data-id="WALLCLOCK_VAL"></span> <span class="tooltipped" data-tip-id="WALLCLOCK_DESCR">Elapsed Time</span></article><article data-id="GFLOPS"><span data-id="GFLOPS_TOTAL_VAL"></span> <span data-tip-id="GFLOPS_DESCR" class="tooltipped">SP GFLOPS</span></article><article data-id="CPI_RATE"><span data-id="CPI_RATE_VAL" data-tip-id="CPI_RATE_ISSUE"></span> <span data-tip-id="CPI_RATE_DESCR" class="tooltipped">CPI</span><br><span data-id="CPI_INFO">(<span class="tooltipped" data-tip-id="CPI_RATE_MAX_NODE">MAX</span> <span data-id="CPI_RATE_MAX_VAL"></span>, <span class="tooltipped" data-tip-id="CPI_RATE_MIN_NODE">MIN</span> <span data-id="CPI_RATE_MIN_VAL"></span>)</span></article><div id="bricks"><article data-id="MPI_TIME"><h3 data-id="MPI_TIME_TITLE" data-tip-id="MPI_TIME_DESCR" class="tooltipped">MPI Time</h3><span data-id="MPI_TIME_VAL"></span><br><span data-id="MPI_TIME_PERC" data-tip-id="MPI_TIME_ISSUE"></span> of Elapsed Time<br><div class="indent"><h4 data-id="MPI_IMBALANCE_TITLE" data-tip-id="MPI_IMBALANCE_DESCR" class="tooltipped">MPI Imbalance</h4><span data-id="MPI_IMBALANCE_VAL"></span><br><span data-id="MPI_IMBALANCE_PERC" data-tip-id="MPI_IMBALANCE_ISSUE"></span> of Elapsed Time<br></div><table data-id="TOP_5"><tr><th>TOP 5 MPI Functions</th><th><span class="tooltipped" data-tip-id="MPI_PERC_DESCR">%</span></th></tr></table></article><article data-id="OMP_SERIAL_TIME"><h3 data-id="OMP_SERIAL_TIME_TITLE" data-tip-id="OMP_SERIAL_TIME_DESCR" class="tooltipped">Serial Time</h3><span data-id="OMP_SERIAL_TIME_VAL"></span><br><span data-id="OMP_SERIAL_TIME_PERC" data-tip-id="OMP_SERIAL_TIME_ISSUE"></span> of Elapsed Time<br></article><article data-id="OMP_IMBALANCE_TIME"><h3 data-id="OMP_IMBALANCE_TIME_TITLE" data-tip-id="OMP_IMBALANCE_TIME_DESCR" class="tooltipped">OpenMP Imbalance</h3><span data-id="OMP_IMBALANCE_TIME_VAL"></span><br><span data-id="OMP_IMBALANCE_TIME_PERC" data-tip-id="OMP_IMBALANCE_TIME_ISSUE"></span> of Elapsed Time<br></article><article data-id="CPU_USAGE"><h3 data-id="CPU_USAGE_TITLE" data-tip-id="CPU_USAGE_DESCR" class="tooltipped">CPU Utilization</h3><span data-id="CPU_USAGE_PERC" data-tip-id="CPU_USAGE_ISSUE"></span><div class="indent"><h4 data-id="AVERAGE_CPU_USE_DESCR" class="tooltipped">Average CPU Utilization</h4><span data-id="AVERAGE_CPU_USE_VAL"></span> Out of <span data-id="CPU_OUT_OF_VAL"></span> logical CPUs</div></article><article data-id="PHYSICAL_CORE_USAGE"><h3 data-id="PHYSICAL_CORE_USAGE_TITLE" data-tip-id="PHYSICAL_CORE_USAGE_DESCR" class="tooltipped">Physical Core Utilization</h3><span data-id="PHYSICAL_CORE_USAGE_PERC" data-tip-id="PHYSICAL_CORE_USAGE_ISSUE"></span><div class="indent"><h4 data-id="PH_CORE_USE_DESCR" class="tooltipped">Average Physical Core Utilization</h4><span data-id="PH_CORE_USE_VAL"></span> out of <span data-id="PH_CORE_OUT_OF_VAL"></span> physical cores</div></article><article data-id="MEMORY_BOUND"><h3 data-id="MEMORY_BOUND_TITLE" data-tip-id="MEMORY_BOUND_DESCR" class="tooltipped">Memory Stalls</h3><span data-id="MEMORY_BOUND_PERC" data-tip-id="MEMORY_BOUND_ISSUE"></span> of pipeline slots<div class="indent"><h4 data-id="CACHE_BOUND_TITLE" data-tip-id="CACHE_BOUND_DESCR" class="tooltipped">Cache Stalls</h4><span data-id="CACHE_BOUND_PERC" data-tip-id="CACHE_BOUND_ISSUE"></span> of cycles<h4 data-id="DRAM_BOUND_TITLE" data-tip-id="DRAM_BOUND_DESCR" class="tooltipped">DRAM Stalls</h4><span data-id="DRAM_BOUND_PERC" data-tip-id="DRAM_BOUND_ISSUE"></span> of cycles<h4 data-id="DRAM_BANDWIDTH_TITLE" data-tip-id="DRAM_BANDWIDTH_DESCR" class="tooltipped">Average DRAM Bandwidth</h4><span data-id="DRAM_BANDWIDTH_VAL" data-tip-id="DRAM_BANDWIDTH_VAL_DESCR" class="tooltipped">Not Available</span><span data-id="DRAM_BANDWIDTH_UNIT"></span><h4 data-id="MCDRAM_BANDWIDTH_TITLE" data-tip-id="MCDRAM_BANDWIDTH_DESCR" class="tooltipped">Average MCDRAM Bandwidth</h4><span data-id="MCDRAM_BANDWIDTH_VAL" data-tip-id="MCDRAM_BANDWIDTH_VAL_DESCR" class="tooltipped">Not Available</span><span data-id="MCDRAM_BANDWIDTH_UNIT"></span><h4 data-id="NUMA_USE_TITLE" data-tip-id="NUMA_USE_DESCR" class="tooltipped">NUMA</h4><span data-id="NUMA_USE_PERC" data-tip-id="NUMA_USE_ISSUE"></span> of remote accesses</div></article><article data-id="BACK_END_BOUND"><h3 data-id="BACK_END_BOUND_TITLE" data-tip-id="BACK_END_BOUND_DESCR" class="tooltipped">Back-End Stalls</h3><span data-id="BACK_END_BOUND_PERC" data-tip-id="BACK_END_BOUND_ISSUE"></span> of pipeline slots<div class="indent"><h4 data-tip-id="L2_HIT_BOUND_DESCR" class="tooltipped">L2 Hit Bound</h4><span data-id="L2_HIT_BOUND_PERC" data-tip-id="L2_HIT_BOUND_ISSUE"></span> of cycles<h4 data-tip-id="L2_MISS_BOUND_DESCR" class="tooltipped">L2 Miss Bound</h4><span data-id="L2_MISS_BOUND_PERC" data-tip-id="L2_MISS_BOUND_ISSUE"></span> of cycles<h4 data-id="DRAM_BANDWIDTH_TITLE" data-tip-id="DRAM_BANDWIDTH_DESCR" class="tooltipped">Average DRAM Bandwidth</h4><span data-id="DRAM_BANDWIDTH_VAL" data-tip-id="DRAM_BANDWIDTH_VAL_DESCR" class="tooltipped">Not Available</span><span data-id="DRAM_BANDWIDTH_UNIT"></span><h4 data-id="MCDRAM_BANDWIDTH_TITLE" data-tip-id="MCDRAM_BANDWIDTH_DESCR" class="tooltipped">Average MCDRAM Bandwidth</h4><span data-id="MCDRAM_BANDWIDTH_VAL" data-tip-id="MCDRAM_BANDWIDTH_VAL_DESCR" class="tooltipped">Not Available</span><span data-id="MCDRAM_BANDWIDTH_UNIT"></span></div></article><article data-id="FPU_UTILIZATION"><h3 data-id="FPU_UTILIZATION_TITLE" data-tip-id="FPU_UTILIZATION_DESCR" class="tooltipped">FPU Utilization</h3><span data-id="FPU_UTILIZATION_PERC" data-tip-id="FPU_UTILIZATION_ISSUE"></span><div class="indent"><h4 data-tip-id="FLOPS_PER_CYCLE_DESCR" class="tooltipped">SP FLOPs per Cycle</h4><span data-id="FLOPS_PER_CYCLE_VAL" data-tip-id="FLOPS_PER_CYCLE_ISSUE"></span> Out of <span data-id="FLOPS_OUT_OF_VAL"></span><h4 data-tip-id="VECTOR_CAPACITY_DESCR" class="tooltipped">Vector Capacity Usage</h4><span data-id="VECTOR_CAPACITY_PERC" data-tip-id="VECTOR_CAPACITY_ISSUE"></span><h4>FP Instruction Mix</h4>% of <span data-tip-id="FP_PACKED_DESCR" class="tooltipped">Packed FP Instr.</span>: <span data-id="FP_PACKED_PERC"></span><div class="indent">% of <span data-tip-id="FP_128_BIT_DESCR" class="tooltipped">128-bit</span>: <span data-id="FP_128_BIT_PERC" data-tip-id="FP_128_BIT_ISSUE"></span><br>% of <span data-tip-id="FP_256_BIT_DESCR" class="tooltipped">256-bit</span>: <span data-id="FP_256_BIT_PERC" data-tip-id="FP_256_BIT_ISSUE"></span><br><span data-id="FP_512_BIT">% of <span data-tip-id="FP_512_BIT_DESCR" class="tooltipped">512-bit</span>: <span data-id="FP_512_BIT_PERC"></span></span></div>% of <span data-tip-id="SCALAR_FP_DESCR" class="tooltipped">Scalar FP Instr.</span>: <span data-id="SCALAR_FP_PERC" data-tip-id="SCALAR_FP_ISSUE"></span><h4 data-id="FP_ARITH_MEM_RD" data-tip-id="FP_ARITH_MEM_RD_DESCR" class="tooltipped">FP Arith/Mem Rd Instr. Ratio</h4><span data-id="FP_ARITH_MEM_RD_VAL" data-tip-id="FP_ARITH_MEM_RD_ISSUE"></span><h4 data-id="FP_ARITH_MEM_WR" data-tip-id="FP_ARITH_MEM_WR_DESCR" class="tooltipped">FP Arith/Mem Wr Instr. Ratio</h4><span data-id="FP_ARITH_MEM_WR_VAL" data-tip-id="FP_ARITH_MEM_WR_ISSUE"></span></div></article><article data-id="SIMD_PER_CYCLE"><h3 data-id="SIMD_PER_CYCLE_TITLE" data-tip-id="SIMD_PER_CYCLE_DESCR" class="tooltipped">SIMD Instr. per Cycle</h3><span data-id="SIMD_PER_CYCLE_VAL" data-tip-id="SIMD_PER_CYCLE_ISSUE"></span><div class="indent"><h4>FP Instruction Mix</h4><div class="indent">% of <span data-tip-id="PACKED_SIMD_DESCR" class="tooltipped">Packed SIMD Instr.</span>: <span data-id="PACKED_SIMD_PERC"></span><br>% of <span data-tip-id="SCALAR_SIMD_DESCR" class="tooltipped">Scalar SIMD Instr.</span>: <span data-id="SCALAR_SIMD_PERC" data-tip-id="SCALAR_SIMD_ISSUE"></span></div></div></article><article data-id="IO"><h3 data-id="IO_TITLE" data-tip-id="IOWAIT_USEC_DESCR" class="tooltipped">I/O Bound</h3><div data-id="IO_TEXT_BLOCK">These metrics are not available for Pcontrol.</div><div data-id="IO_DATA_BLOCK"><span data-id="IOWAIT_USEC_PERC"></span><br>(AVG <span data-id="IOWAIT_USEC_AVG_VAL"></span>, <a class="tooltipped" data-tip-id="IOWAIT_USEC_MAX_RANK">PEAK</a> <span data-id="IOWAIT_USEC_MAX_VAL"></span>)<div class="indent"><h4 data-tip-id="READ_BYTES_DESCR" class="tooltipped">Read</h4>AVG <span data-id="READ_BYTES_AVG_VAL"></span>, <a class="tooltipped" data-tip-id="READ_BYTES_MAX_RANK">MAX</a> <span data-id="READ_BYTES_MAX_VAL"></span><h4 data-tip-id="WRITTEN_BYTES_DESCR" class="tooltipped">Write</h4>AVG <span data-id="WRITTEN_BYTES_AVG_VAL"></span>, <a class="tooltipped" data-tip-id="WRITTEN_BYTES_MAX_RANK">MAX</a> <span data-id="WRITTEN_BYTES_MAX_VAL"></span></div></div></article><article data-id="MEMORY"><h3 data-tip-id="MEM_DESCR" class="tooltipped">Memory Footprint</h3><div data-id="MEMORY_TEXT_BLOCK">These metrics are not available for Pcontrol.</div><div data-id="MEMORY_DATA_BLOCK"><div data-id="RSS_APP">Resident total: <span data-id="RSS_VAL"></span></div><table><tr><th>Resident</th><th>PEAK</th><th>AVG</th></tr><tr><td>Per node:</td><td><span data-id="PEAK_RSS_PER_NODE_VAL" class="tooltipped" data-tip-id="PEAK_RSS_NODE"></span></td><td><span data-id="MEAN_RSS_PER_NODE_VAL"></span></td></tr><tr><td>Per rank:</td><td class="cell"><span data-id="PEAK_RSS_VAL" class="tooltipped" data-tip-id="PEAK_RSS_RANK"></span></td><td class="cell"><span data-id="MEAN_RSS_VAL"></span></td></tr></table><div data-id="MEMORY_APP">Virtual total: <span data-id="MEM_VAL"></span></div><table><tr><th>Virtual</th><th>PEAK</th><th>AVG</th></tr><tr><td>Per node:</td><td><span data-id="PEAK_MEM_PER_NODE_VAL" class="tooltipped" data-tip-id="PEAK_MEM_NODE"></span></td><td><span data-id="MEAN_MEM_PER_NODE_VAL"></span></td></tr><tr><td>Per rank:</td><td class="cell"><span data-id="PEAK_MEM_VAL" class="tooltipped" data-tip-id="PEAK_MEM_RANK"></span></td><td class="cell"><span data-id="MEAN_MEM_VAL"></span></td></tr></table></div></article></div></div></div><div class="tip"><div data-id="current_run">Metric <b>value</b> collected during the application profiling run.</div><div data-id="target">Metric threshold used to indicate possible performance issues. Threshold values are fixed and may not accurately reflect the nature of your application.</div><div data-id="delta">Visual representation of the current run value compared to the target threshold. The Delta is set to zero if the current run value is within the target threshold.</div><div data-id="WALLCLOCK_DESCR"></div><div data-id="GFLOPS_DESCR"></div><div data-id="CPI_RATE_MIN_NODE"></div><div data-id="CPI_RATE_MAX_NODE"></div><div data-id="CPI_RATE_DESCR"></div><div data-id="CPI_RATE_ISSUE"></div><div data-id="MEM_DESCR"></div><div data-id="RSS_DESCR"></div><div data-id="MEMORY_BOUND_MIN_NODE"></div><div data-id="MEMORY_BOUND_MAX_NODE"></div><div data-id="READ_BYTES_MAX_RANK"></div><div data-id="WRITTEN_BYTES_MAX_RANK"></div><div data-id="IOWAIT_USEC_MAX_RANK"></div><div data-id="IOWAIT_USEC_DESCR"></div><div data-id="PEAK_MEM_RANK"></div><div data-id="PEAK_MEM_NODE"></div><div data-id="PEAK_RSS_RANK"></div><div data-id="PEAK_RSS_NODE"></div><div data-id="READ_BYTES_DESCR"></div><div data-id="WRITTEN_BYTES_DESCR"></div><div data-id="MPI_TIME_DESCR"></div><div data-id="MPI_TIME_ISSUE"></div><div data-id="MPI_IMBALANCE_DESCR"></div><div data-id="MPI_IMBALANCE_ISSUE"></div><div data-id="OMP_IMBALANCE_TIME_ISSUE"></div><div data-id="OMP_IMBALANCE_TIME_DESCR"></div><div data-id="OMP_SERIAL_TIME_ISSUE"></div><div data-id="OMP_SERIAL_TIME_DESCR"></div><div data-id="MEMORY_BOUND_DESCR"></div><div data-id="MEMORY_BOUND_ISSUE"></div><div data-id="FPU_UTILIZATION_DESCR"></div><div data-id="FPU_UTILIZATION_ISSUE"></div><div data-id="FLOPS_PER_CYCLE_DESCR"></div><div data-id="FLOPS_PER_CYCLE_ISSUE"></div><div data-id="VECTOR_CAPACITY_DESCR"></div><div data-id="VECTOR_CAPACITY_ISSUE"></div><div data-id="FP_PACKED_DESCR"></div><div data-id="FP_128_BIT_DESCR"></div><div data-id="FP_128_BIT_ISSUE"></div><div data-id="FP_256_BIT_DESCR"></div><div data-id="FP_256_BIT_ISSUE"></div><div data-id="FP_512_BIT_DESCR"></div><div data-id="SCALAR_FP_DESCR"></div><div data-id="SCALAR_FP_ISSUE"></div><div data-id="FP_ARITH_MEM_RD_DESCR"></div><div data-id="FP_ARITH_MEM_RD_ISSUE"></div><div data-id="FP_ARITH_MEM_WR_DESCR"></div><div data-id="FP_ARITH_MEM_WR_ISSUE"></div><div data-id="CACHE_BOUND_DESCR"></div><div data-id="CACHE_BOUND_ISSUE"></div><div data-id="DRAM_BOUND_DESCR"></div><div data-id="DRAM_BOUND_ISSUE"></div><div data-id="NUMA_USE_DESCR"></div><div data-id="NUMA_USE_ISSUE"></div><div data-id="CPU_USAGE_DESCR"></div><div data-id="CPU_USAGE_ISSUE"></div><div data-id="PHYSICAL_CORE_USAGE_DESCR"></div><div data-id="PHYSICAL_CORE_USAGE_ISSUE"></div><div data-id="AVERAGE_CPU_USE_DESCR"></div><div data-id="BACK_END_BOUND_DESCR"></div><div data-id="BACK_END_BOUND_ISSUE"></div><div data-id="L2_HIT_BOUND_DESCR"></div><div data-id="L2_MISS_BOUND_DESCR"></div><div data-id="L2_HIT_BOUND_ISSUE"></div><div data-id="L2_MISS_BOUND_ISSUE"></div><div data-id="SIMD_PER_CYCLE_DESCR"></div><div data-id="SIMD_PER_CYCLE_ISSUE"></div><div data-id="PACKED_SIMD_DESCR"></div><div data-id="SCALAR_SIMD_DESCR"></div><div data-id="SCALAR_SIMD_ISSUE"></div><div data-id="FLOPS_PER_CYCLE_ISSUE"></div><div data-id="DRAM_BANDWIDTH_DESCR">Average amount of data transferred through DRAM memory controller per second.</div><div data-id="MCDRAM_BANDWIDTH_DESCR">Average amount of data transferred through MCDRAM memory controller per second.</div><div data-id="DRAM_BANDWIDTH_VAL_DESCR">Data for this metric is not collected since it requires system-wide performance monitoring. Make sure the sampling driver is properly installed on your system: <a href="https://software.intel.com/en-us/vtune-amplifier-help-sep-driver">https://software.intel.com/en-us/vtune-amplifier-help-sep-driver</a>. Otherwise, enable a driverless Perf-based sampling collection by setting the /proc/sys/kernel/perf_even_paranoid value to 0 or less.</div><div data-id="MCDRAM_BANDWIDTH_VAL_DESCR">Data for this metric is not collected since it requires system-wide performance monitoring. Make sure the sampling driver is properly installed on your system: <a href="https://software.intel.com/en-us/vtune-amplifier-help-sep-driver">https://software.intel.com/en-us/vtune-amplifier-help-sep-driver</a>. Otherwise, enable a driverless Perf-based sampling collection by setting the /proc/sys/kernel/perf_even_paranoid value to 0 or less.</div><div data-id="MCDRAM_BANDWIDTH_ISSUE"></div><div data-id="DRAM_BANDWIDTH_ISSUE"></div><div data-id="MPI_PERC_DESCR">Percentage from Elapsed Time</div></div><div class="not-supported-browser">Intel<sup>®</sup> MPI Performance Snapshot report cannot be opened with the current browser. Use any of these supported browsers:<ul><li>Mozilla Firefox* version 17 or later</li><li>Safari* version 8 or later</li><li>Microsoft Internet Explorer* version 11 or later</li><li>Microsoft Edge* version 12 or later</li><li>Google Chrome* version 40 or later</li></ul></div></body></html>