[Wien] parallel calc. without NFS

[Jeff Spirko]

On Wed, Jul 09, 2003 at 07:13:43PM +1000, Steven Homolya wrote:
> Local jobspace is allocated dynamically by the PBS and each node has its
> own jobspace and path (accesible only to processes that are part of the
> job and are running on that node), so there is no good choice for SCRATCH.
> Besides, reading/writing large chunks of data are not a problem. It's
> "trickly" I/O that many of the scripts are doing (eg. sed, awk) that are
> causing the problems. Things like writing a large array one element at a
> time, or reading a file line by line.

You might try smoothing out some of that I/O.  Using GNU versions of
sed and awk may help.  Or, piping the input our output through cat
or dd (with some blocksize ~10k) may force that I/O to be buffered.

-- 
Jeff Spirko   spirko at lehigh.edu   spirko at yahoo.com   WD3V   |=>

The study of non-linear physics is like the study of non-elephant biology.

All theoretical chemistry is really physics;
and all theoretical chemists know it. -- Richard P. Feynman