Linux Process Monitoring – Getting Disk I/O Statistics for a Single Process

disk-usagelinuxmonitoringprocess

I need to monitor the I/O statistics of a process that writes to disk. The purpose is to avoid write rates too high for long periods.

I know there's iostat tool to accomplish this task on a system-wide perspective.

Is there something similar to monitor single process disk usage?

Best Answer

What you want is iotop. Most distributions have a package for it, usually called (logically enough) iotop.

One very cool command (at least, on a system that isn't very busy) is iotop -bo. This will show I/O as it occurs. It also has options to only monitor specific processes or processes owned by specified users.

Related Solutions

Linux – Detailed Per-Process Profiling

Meanwhile I wrote my own program - audria - capable of monitoring the resource usage of one or more processes, including current/average CPU usage, virtual memory usage, IO load and other information.

Run ‘top’ Command Once and Exit – How to Guide

In Linux, you can try this:

top -bn1 > output.txt

From man top:

-b : Batch-mode operation
            Starts top in 'Batch' mode, which could be useful for sending
            output from top to other programs or  to  a  file.   In  this
            mode, top will not accept input and runs until the iterations
            limit you've set with the '-n' command-line option  or  until
            killed.
....
-n : Number-of-iterations limit as:  -n number
            Specifies  the  maximum  number of iterations, or frames, top
            should produce before ending.

With OS X, try:

top -l 1

From top OSX manpage:

 -l <samples>
              Use logging mode and display <samples> samples, even if 
              standard output is a terminal. 0 is treated  as  infinity.   
              Rather than redisplaying, output is periodically printed in 
              raw form. Note that the first sample displayed will have an 
              invalid %CPU displayed for each process,  as it is calculated 
              using the delta between samples.

Best Answer

Related Solutions

Linux – Detailed Per-Process Profiling

Run ‘top’ Command Once and Exit – How to Guide

Related Question