Atop System and Process Monitor

Atop – A top System and Process Monitor

Just learned about this useful tool called atop.  Better than top and just as useful as vmstat; atop will track processes as they grow and shrink memory. Very handy tool to determine what is going on a system and where to begin when looking for places to tune. Found this interesting and highly recommended paper discussing the tool using a case study of a process that has gone memory bound on the atop website.

To install on Centos.  Choose one of the following commands depending on 32 or 64 bit Centos:

# rpm -Uvh http://download.fedora.redhat.com/pub/epel/5/i386/epel-release-5-4.noarch.rpm
or
# rpm -Uvh http://download.fedora.redhat.com/pub/epel/5/x86_64/epel-release-5-4.noarch.rpm
followed by:
# yum install atop

Note: Don’t forget to enable the repository if it is disabled.

Once installed, you can type the following:

# atop

which will yield the following output:

ATOP - mimir              2011/01/27  18:23:00          2858101 seconds elapsed
PRC | sys 123h31m | user  80h32m | #proc    336 | #zombie    0 | #exit      0 |
CPU | sys     13% | user     13% | irq       0% | idle    771% | wait	   3% |
cpu | sys      3% | user      2% | irq       0% | idle     95% | cpu000 w  0% |
cpu | sys      2% | user      2% | irq       0% | idle     95% | cpu001 w  0% |
cpu | sys      2% | user      1% | irq       0% | idle     95% | cpu005 w  2% |
cpu | sys      1% | user      1% | irq       0% | idle     97% | cpu006 w  0% |
cpu | sys      1% | user      1% | irq       0% | idle     98% | cpu003 w  0% |
cpu | sys      1% | user      2% | irq       0% | idle     97% | cpu002 w  0% |
cpu | sys      1% | user      1% | irq       0% | idle     98% | cpu007 w  0% |
cpu | sys      2% | user      2% | irq       0% | idle     96% | cpu004 w  0% |
CPL | avg1   0.00 | avg5    0.02 | avg15   0.04 | csw 179951e5 | intr 80499e5 |
MEM | tot   11.8G | free  405.0M | cache   7.0G | buff  208.5M | slab  432.8M |
SWP | tot   23.4G | free   23.1G |              | vmcom   8.6G | vmlim  29.3G |
PAG | scan 5625e5 | stall      0 |              | swin 1120890 | swout 1220e3 |
DSK |         sdb | busy      5% | read 25104e3 | write 8528e3 | avio    4 ms |
DSK |         sdc | busy      4% | read 26837e3 | write 8501e3 | avio    3 ms |
DSK |         sda | busy      4% | read 26190e3 | write 8600e3 | avio    3 ms |
DSK |         sdd | busy      4% | read 26680e3 | write 8400e3 | avio    3 ms |
NET | transport   | tcpi 58338e4 | tcpo 22045e4 | udpi  438593 | udpo  160644 |
NET | network     | ipi 584080e3 | ipo 220687e3 | ipfrw      0 | deliv 5840e5 |
NET | eth0     0% | pcki 59401e4 | pcko 22900e4 | si 1778 Kbps | so 2093 Kbps |
NET | lo     ---- | pcki   30472 | pcko   30472 | si    0 Kbps | so    0 Kbps |
NET | virbr0 ---- | pcki       0 | pcko    9199 | si    0 Kbps | so    0 Kbps |
               	*** system and process activity since boot ***
  PID  SYSCPU  USRCPU  VGROW  RGROW  RDDSK  WRDSK  ST EXC S  CPU CMD     1/34
 3397  83h44m  46h33m   2.9G   1.2G 147.8G 1054.6G  N-   - S  16% VirtualBox
 5803 736m59s  19h37m   2.6G 244.5M 212.3M   4.4G  N-   - S   4% chrome
 2232 771m40s 101m04s 209.5M 34536K 157.6M  3324K  N-   - S   2% Xorg
 2527 214m08s 201m18s 566.7M 41900K 195.5M    12K  N-   - S   1% pulseaudio
  495 279m49s   0.00s     0K     0K     0K     0K  N-   - S   1% md1_raid5
 2749  22m30s 123m30s 964.5M 43936K 24420K  1436K  N-   - S   0% chrome
 2514  35m03s 104m01s   1.5G 349.3M   2.3G  19.4G  N-   - S   0% chrome
 1462  28m16s  27m14s 242.7M  5024K 17788K  12.5G  N-   - S   0% rsyslogd
 2594   2m29s  41m53s   1.4G 200.5M 140.7M   256K  N-   - S   0% chrome
 3365  24m51s  19m27s 815.3M  3432K 407.0M 103.0M  N-   - S   0% VBoxSVC