> }volumes for fault tolerance as well as load balancing (we currently
> }run 3 database servers, which is reasonable for our size. Andrew,
> }which has nearly an order of magnitude more users than we do, runs
> }with 5. You probably should be doing the same).
>
> We just moved them onto 3 machines doing nothing else -- this
> was a big reliability plus for us -- esp. since the file servers
> take so long to come up (fsck/salvage/volattach).
Our database servers are now (for the last 10 months or so) no longer
acting as fileservers, but they do handle a few other tasks. For example,
one acts as the central repository for AFS server software and does
periodic tasks like nightly replication of RO volumes; a second is a
nameserver for CMU.EDU, and a third is an "alpha-test" fileserver - it
has a single vice partition which I use for testing.
> }With reasonably fast servers, I'd reccommend not putting more than,
> }say, 50-100GB per server, and then only if your network can keep up
> }(it sounds like you don't have that problem).
>
> What I'd like to do is move to a smaller number of FileServers,
> but I'm unsure how much data you can put on an AFS server.
> What I have in mind are AXP1000s with a pair of the 3 channel
> RAID controllers which would give 144GB of space.
I'd love to hear of your experience with this kind of configuration.
-- Jeffrey T. Hutzelman (N3NHS) <[EMAIL PROTECTED]>
Systems Programmer, CMU SCS Research Facility
Please send requests and problem reports to [EMAIL PROTECTED]