github.com/kaydxh/golang@v0.0.131/pkg/gocv/cgo/third_path/graphics-magick/share/doc/GraphicsMagick/www/benchmark.html (about) 1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" 2 "http://www.w3.org/TR/html4/loose.dtd"> 3 <html> 4 <head> 5 <meta HTTP-EQUIV="CONTENT-TYPE" CONTENT="text/html; charset=us-ascii"> 6 <title>GraphicsMagick GM Utility</title> 7 <style type=text/css> 8 <!-- 9 @page { size: 8.5in 11in } 10 TD P { color: #000000; font-family: "Verdana", "Arial", "Helvetica", sans-serif; font-size: 12pt } 11 P { color: #000000; font-family: "Verdana", "Arial", "Helvetica", sans-serif; font-size: 12pt } 12 A:link { color: #00B04F } 13 A:visited { color: #007B37 } 14 --> 15 </style> 16 17 </head> 18 <body LANG="en-US" TEXT="#000000" LINK="#00B04F" VLINK="#007B37" BGCOLOR="#ffffff"> 19 <a name="top"></a> 20 <table border=0 cellpadding=10 cellspacing=0 style="margin-top:-17px" width="100%"><tr><td> 21 <br> <br> 22 <table BORDER=0 WIDTH="100%"> 23 <tr> 24 <td ALIGN=LEFT bgcolor="#FFFFFF"><img 25 SRC="images/right_triangle.png" ALT=">" BORDER=0 26 height=14 width=15><b><font face="Helvetica, Arial"><font 27 color="#00B04F"><font size="+1"> 28 <a NAME="benchmark"></a>gm benchmark 29 </font></font></font></b></td></tr></table> 30 <table width="94%" border="0" cellspacing="0" cellpadding="8"> 31 <tr><td width="3%"><br></td><td> 32 <table BORDER=0 WIDTH="100%"> 33 <tr> 34 <td ALIGN=LEFT bgcolor="#FFFFFF"><img 35 SRC="images/right_triangle.png" ALT=">" BORDER=0 36 height=14 width=15><b><font face="Helvetica, Arial"><font 37 color="#00B04F"><font size="+1"> 38 <a NAME="bench-top"></a>NAME 39 </font></font></font></b></td></tr></table> 40 <table width="94%" border="0" cellspacing="0" cellpadding="8"> 41 <tr><td width="3%"><br></td><td> 42 benchmark - benchmark the execution of a gm command 43 </td></tr></table> 44 <table BORDER=0 WIDTH="100%"> 45 <tr> 46 <td ALIGN=LEFT bgcolor="#FFFFFF"><img 47 SRC="images/right_triangle.png" ALT=">" BORDER=0 48 height=14 width=15><b><font face="Helvetica, Arial"><font 49 color="#00B04F"><font size="+1"> 50 <a NAME="bench-contents"></a>Contents 51 </font></font></font></b></td></tr></table> 52 <table width="94%" border="0" cellspacing="0" cellpadding="8"> 53 <tr><td width="3%"><br></td><td> 54 <dl> 55 <dt> 56 <a href="#bench-syno">Synopsis</a> 57 </dt> 58 <dt> 59 <a href="#bench-desc">Description</a> 60 </dt> 61 <dt> 62 <a href="#bench-exam">Examples</a> 63 </dt> 64 <dt> 65 <a href="#bench-opti">Options</a> 66 </dt> 67 </dl> 68 </td></tr></table> 69 <table BORDER=0 WIDTH="100%"> 70 <tr> 71 <td ALIGN=LEFT bgcolor="#FFFFFF"><img 72 SRC="images/right_triangle.png" ALT=">" BORDER=0 73 height=14 width=15><b><font face="Helvetica, Arial"><font 74 color="#00B04F"><font size="+1"> 75 <a NAME="bench-syno"></a>Synopsis 76 </font></font></font></b></td></tr></table> 77 <table width="94%" border="0" cellspacing="0" cellpadding="8"> 78 <tr><td width="3%"><br></td><td> 79 <p> 80 <strong>gm benchmark</strong> <strong>[</strong> <em>options ...</em> <strong>]</strong> <em>command</em> 81 </td></tr></table> 82 <table BORDER=0 WIDTH="100%"> 83 <tr> 84 <td ALIGN=LEFT bgcolor="#FFFFFF"><img 85 SRC="images/right_triangle.png" ALT=">" BORDER=0 86 height=14 width=15><b><font face="Helvetica, Arial"><font 87 color="#00B04F"><font size="+1"> 88 <a NAME="bench-desc"></a>Description 89 </font></font></font></b></td></tr></table> 90 <table width="94%" border="0" cellspacing="0" cellpadding="8"> 91 <tr><td width="3%"><br></td><td> 92 <p> 93 <strong>benchmark</strong> executes an arbitrary <strong>gm</strong> utility command 94 (e.g. <strong>convert</strong>) for one or more loops, and/or a specified 95 execution time, and reports many execution metrics. For builds using 96 OpenMP, a mode is provided to execute the benchmark with an increasing 97 number of threads and provide a report of speedup and multi-thread 98 execution efficiency. If <strong>benchmark</strong> is used to execute a 99 command without any additional benchmark options, then the command is 100 run once. 101 </td></tr></table> 102 <table BORDER=0 WIDTH="100%"> 103 <tr> 104 <td ALIGN=LEFT bgcolor="#FFFFFF"><img 105 SRC="images/right_triangle.png" ALT=">" BORDER=0 106 height=14 width=15><b><font face="Helvetica, Arial"><font 107 color="#00B04F"><font size="+1"> 108 <a NAME="bench-exam"></a>Examples 109 </font></font></font></b></td></tr></table> 110 <table width="94%" border="0" cellspacing="0" cellpadding="8"> 111 <tr><td width="3%"><br></td><td> 112 <p>To obtain benchmark information for a single execution of a 113 command: 114 <pre> 115 % gm benchmark convert input.ppm -gaussian 0x1 output.ppm 116 Results: 32 threads 1 iter 6.20s user 0.41s total 2.439 iter/s 0.161 iter/cpu 117 </pre> 118 <p>To obtain benchmark information from 100 iterations of the 119 command: 120 <pre> 121 % gm benchmark -iterations 100 convert input.ppm -gaussian 0x1 output.ppm 122 Results: 32 threads 100 iter 625.40s user 31.74s total 3.151 iter/s 0.160 iter/cpu 123 </pre> 124 <p>To obtain benchmark information by iterating the command until a 125 specified amount of time (in seconds) has been consumed: 126 <pre> 127 % gm benchmark -duration 30 convert input.ppm -gaussian 0x1 output.ppm 128 Results: 32 threads 91 iter 587.33s user 30.30s total 3.003 iter/s 0.155 iter/cpu 129 </pre> 130 <p>To obtain a full performance report with an increasing number of 131 threads (1-32 threads, stepping the number of threads by four each 132 time): 133 <pre> 134 % gm benchmark -duration 3 -stepthreads 4 convert input.ppm -gaussian 0x2 output.ppm 135 Results: 1 threads 1 iter 8.84s user 8.84s total 0.113 iter/s 0.113 iter/cpu 1.00 speedup 1.000 karp-flatt 136 Results: 4 threads 2 iter 18.37s user 4.89s total 0.409 iter/s 0.109 iter/cpu 3.62 speedup 0.035 karp-flatt 137 Results: 8 threads 3 iter 29.81s user 4.09s total 0.733 iter/s 0.101 iter/cpu 6.48 speedup 0.033 karp-flatt 138 Results: 12 threads 3 iter 30.81s user 3.14s total 0.955 iter/s 0.097 iter/cpu 8.45 speedup 0.038 karp-flatt 139 Results: 16 threads 3 iter 35.02s user 3.01s total 0.997 iter/s 0.086 iter/cpu 8.81 speedup 0.054 karp-flatt 140 Results: 20 threads 4 iter 52.92s user 3.53s total 1.133 iter/s 0.076 iter/cpu 10.02 speedup 0.052 karp-flatt 141 Results: 24 threads 4 iter 60.66s user 3.39s total 1.180 iter/s 0.066 iter/cpu 10.43 speedup 0.057 karp-flatt 142 Results: 28 threads 4 iter 73.10s user 3.35s total 1.194 iter/s 0.055 iter/cpu 10.56 speedup 0.061 karp-flatt 143 Results: 32 threads 4 iter 82.10s user 3.09s total 1.294 iter/s 0.049 iter/cpu 11.44 speedup 0.058 karp-flatt 144 </pre> 145 <p>Here is the interpretation of the output: 146 <ul> 147 <li><strong>threads</strong> - number of threads used. 148 <li><strong>iter</strong> - number of command iterations executed. 149 <li><strong>user</strong> - total user time consumed. 150 <li><strong>total</strong> - total elapsed time consumed. 151 <li><strong>iter/s</strong> - number of command iterations per second. 152 <li><strong>iter/cpu</strong> - amount of CPU time consumed per iteration. 153 <li><strong>speedup</strong> - speedup compared with one thread. 154 <li><strong>karp-flatt</strong> - Karp-Flatt measure of speedup efficiency. 155 </ul> 156 <p><em>Please note that the reported "speedup" is based on the 157 execution time of just one thread. A preliminary warm-up pass is used 158 before timing the first loop in order to ensure that the CPU is 159 brought out of power-saving modes and that system caches are warmed 160 up. Most modern CPUs provide a "turbo" mode where the CPU clock speed 161 is increased (e.g. by a factor of two) when only one or two cores are 162 active. If the CPU grows excessively hot (due to insufficient 163 cooling), then it may dial back its clock rates as a form of thermal 164 management. These factors result in an under-reporting of speedup 165 compared to if "turbo" mode was disabled and the CPU does not need to 166 worry about thermal management. The <strong>powertop</strong> utility available 167 under Linux and Solaris provides a way to observe CPU core clock rates 168 while a benchmark is running.</em> 169 </td></tr></table> 170 <p> 171 <i><a href="#top">Back to Contents</a></i> 172 </p> 173 <table BORDER=0 WIDTH="100%"> 174 <tr> 175 <td ALIGN=LEFT bgcolor="#FFFFFF"><img 176 SRC="images/right_triangle.png" ALT=">" BORDER=0 177 height=14 width=15><b><font face="Helvetica, Arial"><font 178 color="#00B04F"><font size="+1"> 179 <a NAME="bench-opti"></a>Options 180 </font></font></font></b></td></tr></table> 181 <table width="94%" border="0" cellspacing="0" cellpadding="8"> 182 <tr><td width="3%"><br></td><td> 183 <p> Options are processed from left to right and must appear before 184 any argument. 185 <table BORDER=0 WIDTH="94%"> 186 <tr> 187 <td width="3%"><br></td> 188 <td ALIGN=LEFT BGCOLOR="#FFFFFF"> 189 <img SRC="images/right_triangle_option.png" 190 ALT=">" BORDER=0 height=14 191 width=15><b><font face="Helvetica, Arial" 192 ><font color="#00B04F"><font size="+1"> 193 -duration <i>duration</i> 194 </font></font></font></b></td></tr></table> 195 <table width="90%" border="0" cellspacing="0" cellpadding="8"> <tr><td width="6%"><br></td><td>duration to run benchmark</td></tr></table> 196 <table width="90%" border="0" cellspacing="0" cellpadding="8"> <tr><td width="6%"><br></td><td><font size="-1">Specify the number of seconds to run the benchmark. The command is 197 executed repeatedly until the specified amount of time has 198 elapsed.</font></td></tr></table> 199 <table BORDER=0 WIDTH="94%"> 200 <tr> 201 <td width="3%"><br></td> 202 <td ALIGN=LEFT BGCOLOR="#FFFFFF"> 203 <img SRC="images/right_triangle_option.png" 204 ALT=">" BORDER=0 height=14 205 width=15><b><font face="Helvetica, Arial" 206 ><font color="#00B04F"><font size="+1"> 207 -help 208 </font></font></font></b></td></tr></table> 209 <table width="90%" border="0" cellspacing="0" cellpadding="8"> <tr><td width="6%"><br></td><td><font size="-1"> 210 Prints benchmark command help.</font></td></tr></table> 211 <table BORDER=0 WIDTH="94%"> 212 <tr> 213 <td width="3%"><br></td> 214 <td ALIGN=LEFT BGCOLOR="#FFFFFF"> 215 <img SRC="images/right_triangle_option.png" 216 ALT=">" BORDER=0 height=14 217 width=15><b><font face="Helvetica, Arial" 218 ><font color="#00B04F"><font size="+1"> 219 -iterations <i>loops</i> 220 </font></font></font></b></td></tr></table> 221 <table width="90%" border="0" cellspacing="0" cellpadding="8"> <tr><td width="6%"><br></td><td>number of command iterations</td></tr></table> 222 <table width="90%" border="0" cellspacing="0" cellpadding="8"> <tr><td width="6%"><br></td><td><font size="-1">Specify the number of iterations to run the benchmark. The command 223 is executed repeatedly until the specified number of iterations has 224 been reached.</font></td></tr></table> 225 <table BORDER=0 WIDTH="94%"> 226 <tr> 227 <td width="3%"><br></td> 228 <td ALIGN=LEFT BGCOLOR="#FFFFFF"> 229 <img SRC="images/right_triangle_option.png" 230 ALT=">" BORDER=0 height=14 231 width=15><b><font face="Helvetica, Arial" 232 ><font color="#00B04F"><font size="+1"> 233 -rawcsv 234 </font></font></font></b></td></tr></table> 235 <table width="90%" border="0" cellspacing="0" cellpadding="8"> <tr><td width="6%"><br></td><td>Print results in CSV format</td></tr></table> 236 <table width="90%" border="0" cellspacing="0" cellpadding="8"> <tr><td width="6%"><br></td><td><font size="-1">Print results in a comma-separated value (CSV) format which is easy 237 to parse for plotting or importing into a spreadsheet or database. 238 The values reported are <strong>threads</strong>, <strong>iterations</strong>, 239 <strong>user_time</strong>, and <strong>elapsed_time</strong>.</font></td></tr></table> 240 <table BORDER=0 WIDTH="94%"> 241 <tr> 242 <td width="3%"><br></td> 243 <td ALIGN=LEFT BGCOLOR="#FFFFFF"> 244 <img SRC="images/right_triangle_option.png" 245 ALT=">" BORDER=0 height=14 246 width=15><b><font face="Helvetica, Arial" 247 ><font color="#00B04F"><font size="+1"> 248 -stepthreads <i>step</i> 249 </font></font></font></b></td></tr></table> 250 <table width="90%" border="0" cellspacing="0" cellpadding="8"> <tr><td width="6%"><br></td><td>execute a per-thread benchmark ramp</td></tr></table> 251 <table width="90%" border="0" cellspacing="0" cellpadding="8"> <tr><td width="6%"><br></td><td><font size="-1"> Execute a per-thread benchmark ramp, incrementing the number of 252 threads at each step by the specified value. The maximum number of 253 threads is taken from the standard <tt>OMP_NUM_THREADS</tt> 254 environment variable.</font></td></tr></table> 255 </td></tr></table> 256 <p> 257 <i><a href="#top">Back to Contents</a></i> 258 </p> 259 </td></tr></table>