network block device as replacement for NFS

[Chris Black]

We are trying to find the best way to share a small directory tree of 
files over the network from our head node to the compute nodes. There should 
be under a meg of data in this tree and most of it will be scripts and 
makefiles and such. We are considering using the linux kernel's network 
block device support instead of nfs. Is this a good idea? Would nfs be fine?
What would the benefits of either approach be? Would NFS v3 help at all?
I have never used NBD (network block device) stuff before and am not 
that familiar with how it works, but I get the impression that the nodes would 
just issue block read commands to the head node which would return the data.
I don't know if/why/how this would be better than NFS in terms of load on 
the server and network. I do know there is quite a bit of overhead with NFS 
though.

Please advise, thanks.

Chris Black
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
Url : http://www.scyld.com/pipermail/beowulf/attachments/20010430/897f2a15/attachment.bin