1 The following is a demonstration of the diskhits command. 2 3 4 Here we run diskhits on a large file, /extra1/contents with is 46 Mb, and 5 currently hasn't been accessed (so isn't in any cache). 6 7 While diskhits is running, the file is grep'd in another window. This causes 8 the entire file to be read, 9 10 # ./diskhits /extra1/contents 11 Tracing... Hit Ctrl-C to end. 12 ^C 13 Location (KB), 14 15 value ------------- Distribution ------------- count 16 < 0 | 0 17 0 |@@ 47 18 2303 |@@ 41 19 4606 |@@ 41 20 6909 |@@ 42 21 9212 |@@ 41 22 11515 |@@ 41 23 13818 |@@ 42 24 16121 |@@ 43 25 18424 |@@ 42 26 20727 |@@ 41 27 23030 |@@ 41 28 25333 |@@ 41 29 27636 |@@ 41 30 29939 |@@ 42 31 32242 |@@ 44 32 34545 |@@ 41 33 36848 |@@ 41 34 39151 |@@ 41 35 41454 |@@ 41 36 43757 |@@ 40 37 >= 46060 | 0 38 39 Size (KB), 40 41 value ------------- Distribution ------------- count 42 4 | 0 43 8 | 6 44 16 | 10 45 32 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ 818 46 64 | 0 47 48 Total RW: 46064 KB 49 50 Ok, so the file was read evently with each access around 32 to 63 Kb in size, 51 and a total of 46 Mb read. This all makes sense, as it is reading the file 52 for the first time. 53 54 55 56 Now the same file is grep'd with diskhits running, this time we can see what 57 effect caching the file has made, 58 59 # ./diskhits /extra1/contents 60 Tracing... Hit Ctrl-C to end. 61 ^C 62 Location (KB), 63 64 value ------------- Distribution ------------- count 65 2303 | 0 66 4606 | 5 67 6909 |@ 67 68 9212 |@@@@ 170 69 11515 |@@@@@ 216 70 13818 |@@@@@ 224 71 16121 |@@@@@@ 287 72 18424 |@@@@@ 227 73 20727 |@@@ 144 74 23030 |@@ 75 75 25333 |@ 59 76 27636 |@ 42 77 29939 |@ 41 78 32242 |@ 44 79 34545 |@ 41 80 36848 |@ 41 81 39151 |@ 41 82 41454 |@ 41 83 43757 |@ 39 84 >= 46060 | 0 85 86 Size (KB), 87 88 value ------------- Distribution ------------- count 89 2 | 0 90 4 |@@@@@@@@@@@@@@@@@@@@@@@@@ 1137 91 8 |@@@@@ 211 92 16 |@@ 111 93 32 |@@@@@@@@ 345 94 64 | 0 95 96 Total RW: 29392 KB 97 98 The difference is dramatic. This time only 29 Mb is read, leaving around 99 17 Mb that was read from the cache. The way the file is read differs - 100 in the later half of the file it looks the same, but in the first half there 101 are many more events; oddly enough, this is because the early part of the 102 file is cached more, the extra events are likely to be much smaller in size - 103 as indicated in the difference in the size distribution. 104 105 It appears that everything less that 4606 Kb has remained in the cache, with 106 zero hits for that range. 107 108
Indexes created Wed Apr 22 00:22:44 UTC 2026