Recommendations for Bugzilla hardware

[Eddie Xie]

On 2005 August 01 (Mon) 12:59:13pm PDT, David Miller <justdave at bugzilla.org> wrote:
> Eddie Xie wrote:
> >I maintain a customized version of Bugzilla for my organization.  We
> >are upgrading some of our equipment, and one of the things being
> >considered for upgrade is our Bugzilla server.  What sort of specs
> >would be good for our new server?  Bugzilla is used a lot in our
> >organization.  We have around 100 users and over 40,000 bugs in our
> >database.  More bugs get added all the time.  We also seem to like
> >generating graphs and reports.
> >
> >We don't want to invest too much in a Bugzilla server if we could put
> >that money and hardware elsewhere.  On the other hand, we don't want
> >our users complaining that Bugzilla is slow, because it is used all
> >the time.  I welcome any recommendations for this.
> 
> All depends on what hardware you have it running on now. :)

Currently, our system has a PIII 1Ghz CPU, 512MB RAM, 3x36Gig SCSI
drives in RAID5 and a 40GB local hard drive.  Everything runs on this
one system.

> Our Bugzilla is on a 2 x 3.4 GHz HT Xeon box, with hardware SCSI RAID, 
> and we use a replicated database for queries, the primary and replicated 
> MySQL servers are on two separate boxes which are both 2 x 2.8 GHz with 
> software RAID.  We've got about 120,000 total users (about 10,000 
> active) and over 300,000 bugs.  We survived for a long time with much 
> less-capable hardware (these machines are fairly new).
> 
> The database servers are both getting moved to new hardware soon (which 
> will be equivalent hardware with the web server).
> 
> Our current performance bottlenecks are software-related though (mostly 
> sending mail).  Changes to Bugzilla itself will be our next avenue for 
> performance improvements.

I see that you have your Bugzilla using dual processor systems.  How
much of a benefit is it to have two processors?  Do you have to do
anything special to get this benefit (e.g., use a specially compiled
version of MySQL, or use a specialized version of Linux or FreeBSD).

What is the benefit of using a replicated database for queries?  Is it
only when multiple queries are going on at the same time?

Thanks for your reply!

-- 
Eddie Xie