[YLUG] Server purchase

[Robert Hulme]

> It might also be worth looking at Xen, which can migrate Xen instances
> to backup nodes in the case of failure.  Having seen this demonstrated
> by the Xen author, I can attest that this is really impressive.
At my work place all our servers run on Xen - it is really neat, but
to have the migration feature you mention you need to run your disks
off a SAN which is I suspect more complicated and more expensive than
Richard wants.

> BTW, hardware RAID is often slower than software RAID, because with
> software RAID you can split the disks so that each one can saturate a
> single IDE/SATA/SCSI controller/bus.
You can have a PCI-E RAID card that has lots of SATA controllers on
it. PCI-E has enough bw to deal with lots of disks... the real
bottleneck I think (or at least in our experience [which admittedly is
with the < £1,000 cards!]) is with the RAID accelerator (XOR
processor?) on the card...

> With hardware RAID, you often
> have a bottleneck in the RAID controller.  I've been to a talk by one
> of the Linux SW RAID guys, showing that its performance can increase
> linearly with each new controller you add.  OTOH, software RAID may
> potentially be less reliable, and (depending on the hardware) may not
> be hot-swappable.
Yeah that's true. I /think/ SATA howswap is now in the libata driver,
but if so it's *very* recent.

> Also, for extremely fault tolerant filesystems, Sistina/RedHat's GFS
> (Global File System) clustered filesystem might also be something to
> investigate.  This gives you a filesystem distributed over all the
> nodes in the network, with no central server.  It can cope with nodes
> failing, and you can use RAID or normal disks on each node in the
> cluster.  It can be thought of as RAID over an entire cluster of
> networked systems, but it's more than that.
Yeah - there is also DRDB, NDB (http://www.it.uc3m.es/~ptb/nbd/), and
a few others.

-Rob
-- 
------------------------------------------------------
"98.5% of DNA is considered to be junk DNA with no known purpose.
Maybe it's XML tags." -- Anon

"Debugging is twice as hard as writing the code in the first place.
Therefore, if you write the code as cleverly as possible, you are, by
definition, not smart enough to debug it." - Kernighan

http://www.robhulme.com/
http://robhu.livejournal.com/