Shared diskspace between nodes

[Greg Lindahl]

On Mon, Jan 07, 2002 at 02:36:42PM -0500, Jon E. Mitchiner wrote:

> Right now, we are saving data on various nodes, and moving it around when
> space gets tight on a machine.  This is getting time consuming as some of us
> have to look on different nodes to find out where your data is currently
> residing.  I am considering saving all directory names in a database and
> then making a GUI interface via the web so its easy to find the location of
> data directories, rather than looking for it (especially if someone moved my
> directory to another machine without letting me know).

Other people have suggested PVFS, which is a bit less reliable than
what you have now. Another route to go is a simple perl script,
with an interface like:

mkdatadir foo 100M

This creates a symlink "foo" to diskspace somewhere on the cluster
that has 100 megabytes free. The script starts by checking the local
node, and if there isn't enough room there, it can randomly look
around the cluster for a node that's up, and create a symlink to the
appropriate NFS directory (assuming you have N^2 automounts.)

That's basically automating what you do today, and it's as reliable as
what you do today. BTW, if you move someone's files, leave behind a
symlink.

greg