How to find per-process I/O statistics on Linux

http://www.xaprb.com/blog/2009/08/23/how-to-find-per-process-io-statistics-on-linux/

by jawngee • 17 years ago

49 17 17 years ago

Anyone has an idea if there is any way to distinguish random from sequential IO?

In my experience sequential IO is never the problem. Instead, it seems to me that random seeks are really the only performance problem nowadays. In the most extreme case throughput is only ~100Bits/s instead of ~100MB/s. Unfortunately random seeks are hidden behind abstraction layers and thus quite invisible to programmers (until the system freezes). Maybe we just need to wait for SSDs to become cheap.

voberoi• 17 years ago

Take a look at blktrace on Linux. It logs all block I/O layer activity from when an I/O request is sent to the disk to when it's serviced. It also outputs which sectors on the disk were written to service the request. People have written a few tools to read these logs and show meaningful numbers/pretty graphs -- btt is the official one, and here's another one that I like: http://oss.oracle.com/~mason/seekwatcher/.

Among other things, you can distinguish random from sequential I/O using blktrace and its accompanying set of tools by looking at the distance between subsequent I/O requests sent to the disk.

gnaritas2• 17 years ago

SSD's are cheap.

jacquesm2• 17 years ago

Would you care to quantify that ?

blasdel3• 17 years ago

...and cheap SSDs are pathologically bad at random writes -- at least an order of magnitude worse than an mediocre hard drive on average, with regular latency spikes of several seconds!

nailer3• 17 years ago

Shorter method: run 'iotop'.

This tool is packaged in every current Linux OS.

If you want to change IO priority, run ionice.

Dobbs• 17 years ago

I call your bluff. 'iotop' may be included in many linux distro's but don't say every. My Arch box doesn't have it, not by default anyway.

edit: What I'm trying to say, is that your statement is overly broad. Not all linux distro's have 'iotop' installed. It is not a guarantee. Particularly on specialized distros.

jacquesm2• 17 years ago

not by default for most distros means an

apt-get install iotop

yum install iotop

away. That took about 6 seconds on the two machines I tried it on. So even if it is not installed by default it hardly is a hindrance that it isn't.

warp• 17 years ago

He did say "packaged", which is rather different from "installed".

nostrademons• 17 years ago

The article acknowledges iotop and explains that this method is for older kernels that don't support whatever syscalls iotop uses.

nailer3• 17 years ago

Yes, but the headline is misleading - this is a terrible way to 'find per-process I/O statistics on Linux'.

'Linux with older kernels' would be more appropriate.

How to find per-process I/O statistics on Linux

Related Stories