Documentation/admin-guide/iostats.rst

6 more extensive disk statistics have been introduced to help measure disk
7 activity. Tools such as ``sar`` and ``iostat`` typically interpret these and do
44 each snapshot of your disk statistics.
56 or long-lived system; so applications should be prepared to deal with
58 int (32 bit) or unsigned long (32-bit or 64-bit, depending on your
59 machine) as noted per-field below. Unless your observations are very
63 system-wide stats you'll have to find all the devices and sum them all up.
65 Field  1 -- # of reads completed (unsigned long)
68 Field  2 -- # of reads merged, field 6 -- # of writes merged (unsigned long)
71     ultimately handed to the disk, and so it will be counted (and queued)
74 Field  3 -- # of sectors read (unsigned long)
77 Field  4 -- # of milliseconds spent reading (unsigned int)
81 Field  5 -- # of writes completed (unsigned long)
84 Field  6 -- # of writes merged  (unsigned long)
87 Field  7 -- # of sectors written (unsigned long)
90 Field  8 -- # of milliseconds spent writing (unsigned int)
94 Field  9 -- # of I/Os currently in progress (unsigned int)
98 Field 10 -- # of milliseconds spent doing I/Os (unsigned int)
105 Field 11 -- weighted # of milliseconds spent doing I/Os (unsigned int)
112 Field 12 -- # of discards completed (unsigned long)
115 Field 13 -- # of discards merged (unsigned long)
118 Field 14 -- # of sectors discarded (unsigned long)
121 Field 15 -- # of milliseconds spent discarding (unsigned int)
125 Field 16 -- # of flush requests completed
129     This counts flush requests executed by disk. Not tracked for partitions.
131 Field 17 -- # of milliseconds spent flushing
141 almost a non-issue.  When the statistics are read, the per-CPU counters
144 user interface for accessing the per-CPU counters themselves.
150 -------------------
154 a disk address relative to a partition to the disk address relative to
155 the host disk happens much earlier.  All merges and timings now happen
156 at the disk level rather than at both the disk and partition level as
161 Field  1 -- # of reads issued
164 Field  2 -- # of sectors read
168 Field  3 -- # of writes issued
171 Field  4 -- # of sectors written
175 Note that since the address is translated to a disk-relative one, and no
176 record of the partition-relative address is kept, the subsequent success
189 disk and partition statistics are consistent again. Since we still don't
190 keep record of the partition-relative address, an operation is attributed to
196 ----------------
205 In 2.6+, all disk statistics were removed from ``/proc/stat``.  In 2.4, they
210 -- ricklind@us.ibm.com