[rrd-developers] Re: Improving RRD tool scalability

[Stanislav Sinyagin]

Shalom Sasha!

--- Sasha Mikheev <sasha at avalon-net.co.il> wrote:
> Our
> modification reduced number of seeks by several times but it did not
> overcome
> fundamental problem.

Did you hear about the small program called "fastfs"?
It switches some hidden attributes in Solaris filesystem, 
so that when fastfs is ON, the file closing operation becomes asynchronous. 
The current process goes on, and the physical file closing is 
performed some time later. 

This is "unofficial" hack to UFS, but: you can find fastfs on Solaris 
installation disk! Solaris installer uses it, although they say that 
it's unsafe and not supported.

I suppose this could significantly increase the performance in 
your case, though with less safety. I know that it dramatically 
increases mailserver performance in case of large number of mailboxes.

> Also I am very surprised that SNMP collection/CPU usage did not become a
> bottleneck
> before the disk. According to RTG article this was supposed to be a major
> problem.
> On the other hand our collector has about the same performance as RTG even
> though it is written in perl.

I'd be interested to take a look into your collector. Though, the one 
I use in RRFW is fast enough too. 

Cheers, 
Stanislav

--
Unsubscribe mailto:rrd-developers-request at list.ee.ethz.ch?subject=unsubscribe
Help        mailto:rrd-developers-request at list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/rrd-developers
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi