Dear Lam users,
We are trying to run clustalmpi on biocluster grid engine.
System configuration:
-------------------------------
1. Specification:
Operating System:
Red Hat Enterprise Linux 3.0 for AMD 64 Architecture
Sun Fire v20z machine
Bio Clusture Grid Engine
2. Lam version available : lam 7.1.1 ( installed properly).
3. clustalmpi: clustalw-mpi-0.13
4. Through mpirun we are able to run the job and we got the desired out put files. But we want to run same application using qsub(to submit it to grid)
5. We have come across that it requires some sge-lam script to integrate lam with sge.
But We need your assitance where to find the sge-lam script .
And whatever the changes we have to make in that script to integrate with GRID.
I have downloaded one sge-lam script from gridengine.sunsource.net and I tried with loose integration using rsh , but it says some requirement of /machines file.
Previously we have one hostfile in all the nodes which we use for lamboot -d hostfile booting to all the nodes.Is it necessary to again create a /machines file and what changes we need to make in startlam.sh and stoplam.sh ? I have created a PE as lam_loose_rsh in SGE and I have given Start_arg_proc path to startlam.sh and Stop_arg_proc path to stoplam.sh.
But the job is still pending if i run like qsub -pe lam_loose_rsh 4 script
I need your kind assitance.
With Thanks and Regards
Debasis
---------------------------------
Discover Yahoo!
Use Yahoo! to plan a weekend, have fun online & more. Check it out!
|
