Hi again
You're right. This could be quite useful especially for multi-processor
systems...
Also, as you say, there could be other levels of queuing (that people more
familiar with this topic than me could propose)...
Ciao
S?b
Selon Douglas Kojetin
Hi
There are other levels of queuing! In the devlopment branches of
relax there is a working implimentation ('the multi branch) that will do
most of relax's calculations in parallel either using threads or using
mpi or another transport. Now the good news is that i see almost ideal
theoretical increases in speed for the intrinisically parallelisable
calculations. However, even though it is useable it hans't been released
yet because I have been sidetracked onto another urgent projects for
about the last 6 months. However, I will soon be back on it (the next
month or two) and looking for volunteers to try it out ;-)
regards
gary
--
-------------------------------------------------------------------
Dr Gary Thompson
Astbury Centre for Structural Molecular Biology,
University of Leeds, Astbury Building,
Leeds, LS2 9JT, West-Yorkshire, UK Tel. +44-113-3433024
email: garyt@xxxxxxxxxxxxxxx Fax +44-113-2331407
-------------------------------------------------------------------