Coda File System

Re: using coda for large servers

From: Ivan Popov <pin_at_math.chalmers.se>
Date: Tue, 19 Aug 2003 21:45:22 +0200 (MET DST)
Hello Ronald,

> I am investigating whether or not coda is suitable for my application.
> I have two questions about the workings of coda. I would like to
> read-write mirror a rather large filesystem (about 800 gb) over a WAN to
> 4 distinct nodes.  I have sufficient RAM to be able to create a RVM
> metadata partition which is still 4% of the total filessytem size.

There are two issues about it.

 - you would have to run about 30 processes (possibly on the same
host) serving parts of that data, as for the moment we cannot have more
that 1G RVM per server process

 - Coda servers are supposed to have good connectivity to each other,
while the clients do well with slow/flaky likes. So it is not a good idea
to replicate overseas - but no problem to run clients from very far away

> 2)  Also how do the filesystem semantics work?  if someone attempts to
> open a file at location A, and someone else tries to open the same file
> at location B, how does the filesystem respond in this situation?

open() for reading: sees the contents set by last happened close() after
write()

open() for writing at different sites: creates a conflict and make the
file mostly inaccessible until manual resolution of which update is the
"right" one.

Hope it helps,
--
Ivan
Received on 2003-08-20 04:00:59