bcc-biotop(8)

1biotop(8)                   System Manager's Manual                  biotop(8)
2
3
4

NAME

6       biotop - Block device (disk) I/O by process top.
7

SYNOPSIS

9       biotop [-h] [-C] [-r MAXROWS] [-p PID] [interval] [count]
10

DESCRIPTION

12       This is top for disks.
13
14       This  traces block device I/O (disk I/O), and prints a per-process sum‐
15       mary every interval (by default, 1 second). The summary  is  sorted  on
16       the top disk consumers by throughput (Kbytes). The PID and process name
17       shown are measured from when the I/O was first created,  which  usually
18       identifies the responsible process.
19
20       For  efficiency, this uses in-kernel eBPF maps to cache process details
21       (PID and comm) by I/O request, as well as a starting timestamp for cal‐
22       culating I/O latency, and the final summary.
23
24       This  works  by  tracing various kernel blk_*() functions using dynamic
25       tracing, and will need updating to match any  changes  to  these  func‐
26       tions.
27
28       Since this uses BPF, only the root user can use this tool.
29

REQUIREMENTS

31       CONFIG_BPF and bcc.
32

OPTIONS

34       -C     Don't clear the screen.
35
36       -r MAXROWS
37              Maximum number of rows to print. Default is 20.
38
39       -p PID Trace this PID only.
40
41       interval
42              Interval between updates, seconds.
43
44       count  Number of interval summaries.
45

EXAMPLES

47       Summarize block device I/O by process, 1 second screen refresh:
48              # biotop
49
50       Don't clear the screen:
51              # biotop -C
52
53       5 second summaries, 10 times only:
54              # biotop 5 10
55

FIELDS

57       loadavg:
58              The contents of /proc/loadavg
59
60       PID    Cached  process  ID, if present. This usually (but isn't guaran‐
61              teed) to identify the responsible process for the I/O.
62
63       COMM   Cached process name, if present. This usually (but isn't guaran‐
64              teed) to identify the responsible process for the I/O.
65
66       D      Direction: R == read, W == write. This is a simplification.
67
68       MAJ    Major device number.
69
70       MIN    Minor device number.
71
72       DISK   Disk device name.
73
74       I/O    Number of I/O during the interval.
75
76       Kbytes Total Kbytes for these I/O, during the interval.
77
78       AVGms  Average time for the I/O (latency) from the issue to the device,
79              to its completion, in milliseconds.
80

OVERHEAD

82       Since block device I/O  usually  has  a  relatively  low  frequency  (<
83       10,000/s), the overhead for this tool is expected to be low or negligi‐
84       ble. For high IOPS storage systems, test and quantify before use.
85

SOURCE

87       This is from bcc.
88
89              https://github.com/iovisor/bcc
90
91       Also look in the bcc distribution for a  companion  _examples.txt  file
92       containing example usage, output, and commentary for this tool.
93

OS

95       Linux
96

STABILITY

98       Unstable - in development.
99

AUTHOR

101       Brendan Gregg, Rocky Xing
102

INSPIRATION

104       top(1) by William LeFebvre
105