***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***


Hi,

I'm running a molrep job on a protein with approx ~4500 atoms and ~550 residues using the adv (slow) search algorithm and the log file shows that 31h 44m 43s remains.

1) Is that an accurate estimate?
2) Using the same algorithm on the same machine but using ccp4-4.2.2, this search took ~3.25 hours. What is the difference in the algorithm or default setting that causes such a difference in total processing time? Is it the # of cycles of rigid body refinements before and after translational search? I think the default in 5.99.5 is 10 after search, what was it in 4.2.2? 3) Is it possible to parallelize (e.g. using data decomposition ) the molrep algorithm or at least make an option for running N number of threads for people with SMP systems or SSI (single system image) clusters?

VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
209m  50m 2532 R 99.9  1.4   1296:24 molrep
..and still going..

Thanks,
Sabuj Pattanayek



Reply via email to