Efika 5200B Project Review and Support, Application optimization in category Other proposed by Matthias Muench on 4th June 2006 (accepted on 7th October 2006) [View Full Project]

Blog Entry

	Stream Benchmarks posted by Matthias Muench on 10th April 2007 My next step was to do some benchmarks under Linux to check the RAM performance. I used the Stream test against Pegasos and Sam440ep board. EFIKA Linux 2.6.19 ------------------------------------------------------------- This system uses 8 bytes per DOUBLE PRECISION word. ------------------------------------------------------------- Array size = 2000000, Offset = 0 Total memory required = 45.8 MB. Each test is run 10 times, but only the *best* time for each is used. ------------------------------------------------------------- Printing one line per active thread.... ------------------------------------------------------------- Your clock granularity/precision appears to be 2 microseconds. Each test below will take on the order of 183141 microseconds. (= 91570 clock ticks) Increase the size of the arrays if this shows that you are not getting at least 20 clock ticks per test. ------------------------------------------------------------- WARNING -- The above is only a rough guideline. For best results, please be sure you know the precision of your system timer. ------------------------------------------------------------- Function----Rate (MB/s)---Avg time----Min time----Max time Copy:--------192.4569------0.1978------0.1663------0.3896 Scale:--------187.5988------0.1855------0.1706------0.2254 Add:----------195.1688------0.2569------0.2459------0.2920 Triad:---------188.3092------0.3100------0.2549------0.4959 ------------------------------------------------------------- Solution Validates ------------------------------------------------------------- Pegasos II G4/1GHz Linux 2.6.17 ------------------------------------------------------------- This system uses 8 bytes per DOUBLE PRECISION word. ------------------------------------------------------------- Array size = 2000000, Offset = 0 Total memory required = 45.8 MB. Each test is run 10 times, but only the *best* time for each is used. ------------------------------------------------------------- Printing one line per active thread.... ------------------------------------------------------------- Your clock granularity/precision appears to be 1 microseconds. Each test below will take on the order of 90864 microseconds. (= 90864 clock ticks) Increase the size of the arrays if this shows that you are not getting at least 20 clock ticks per test. ------------------------------------------------------------- WARNING -- The above is only a rough guideline. For best results, please be sure you know the precision of your system timer. ------------------------------------------------------------- Function----Rate (MB/s)---Avg time----Min time----Max time Copy:--------376.0548------0.0869------0.0851------0.0907 Scale:--------361.1246------0.0907------0.0886------0.0929 Add:----------359.3918------0.1359------0.1336------0.1376 Triad:---------361.1680------0.1357------0.1329------0.1394 ------------------------------------------------------------- Solution Validates ------------------------------------------------------------- Pegasos II G3/600MHz ------------------------------------------------------------- STREAM version $Revision: 5.6 $ ------------------------------------------------------------- This system uses 8 bytes per DOUBLE PRECISION word. ------------------------------------------------------------- Array size = 2000000, Offset = 0 Total memory required = 45.8 MB. Each test is run 10 times, but only the *best* time for each is used. ------------------------------------------------------------- Printing one line per active thread.... ------------------------------------------------------------- Your clock granularity/precision appears to be 2 microseconds. Each test below will take on the order of 251755 microseconds. (= 125877 clock ticks) Increase the size of the arrays if this shows that you are not getting at least 20 clock ticks per test. ------------------------------------------------------------- WARNING -- The above is only a rough guideline. For best results, please be sure you know the precision of your system timer. ------------------------------------------------------------- Function----Rate (MB/s)---Avg time---Min time---Max time Copy:--------132.6981------0.2416-----0.2411-----0.2421 Scale:-------112.7054------0.2849-----0.2839-----0.2859 Add:---------118.6606------0.4054-----0.4045-----0.4063 Triad:-------118.7202------0.4056-----0.4043-----0.4065 ------------------------------------------------------------- Solution Validates ------------------------------------------------------------- AMCC 440EP 533 Mhz Evaluation Board Linux 2.6.18 ------------------------------------------------------------- This system uses 8 bytes per DOUBLE PRECISION word. ------------------------------------------------------------- Array size = 2000000, Offset = 0 Total memory required = 45.8 MB. Each test is run 10 times, but only the *best* time for each is used. ------------------------------------------------------------- Printing one line per active thread.... ------------------------------------------------------------- Your clock granularity/precision appears to be 1 microseconds. Each test below will take on the order of 112887 microseconds. (= 112887 clock ticks) Increase the size of the arrays if this shows that you are not getting at least 20 clock ticks per test. ------------------------------------------------------------- WARNING -- The above is only a rough guideline. For best results, please be sure you know the precision of your system timer. ------------------------------------------------------------- Function----Rate (MB/s)---Avg time----Min time----Max time Copy:--------262.6503------0.1221------0.1218------0.1223 Scale:--------245.8928------0.1302------0.1301------0.1303 Add:----------253.5979------0.1893------0.1893------0.1893 Triad:---------251.2734------0.1911------0.1910------0.1912 ------------------------------------------------------------- Solution Validates ------------------------------------------------------------- For further tests, Application optimizations and guides I am waiting now for a MorphOS Release.