SAR (was Re: disks stats through /proc)

Stephen C. Tweedie (
Fri, 4 Dec 1998 15:07:07 GMT


On Thu, 3 Dec 1998 13:39:42 +0000 (GMT), Tigran Aivazian
<> said:

> IMHO, this should be done on a fine-grained (partition level) rather than
> coarse (drive) level. There is enough information in
> ll_rw_blk.c/add_request() to do it now but, of course, kernel_stat
> structure will have to be seriously modified.

> There is something frustrating about the quality and speed of Linux
> development. I.e. the quality is too high and the speed is too high,
> in other words, I can implement this disk stat feature, but I bet
> someone else has already done it and is just about to release his
> patch to Linus soon...

Yep, done that. :)

I have patches for 2.0.34/35 and 2.1.125 already working, outputing
both per-partition and per-spindle access stats via /proc/partitions.
A "sard" front-end gives human-readable output. (It is modeled on the
output of the SVR4 "sar -d" output.)

Currently you get average %utilisation; average request queue length;
number of K transfered plus number of distinct IOs for reads, writes
and combined; and average request service time for each disk and each

Look for

for the patches and sard source code. The plan is to integrate this
into 2.3 once we have got a more sensible kdev_t in the kernel.


To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Please read the FAQ at