[torqueusers] Job: Consultant needed

[notinh notien]

Hi, all.  Our company would like to bring in a consultant to help us and I 
thought that it is best to look for on here since many people on this list 
are doing similar things to us.  I am sorry for possible misuse of this 
mailing list but we have failed to bring in one after meeting with 3 
consultants because they lack the specific skills.

We are looking for a consultant with hand-on experiences on Linux cluster 
and MPI (Mpich 1.2.7p1), mpiexec, Torque, Maui, VASP (big plus), Linux 
Kernel, NFS.

Currently, we have some problems with our settings that some jobs got killed 
with SIGTERM and some jobs run for hours and no additional output got 
produced.

-------------------------------------------------------------------
    p4_error: latest msg from perror: Bad file descriptor
    p4_error: latest msg from perror: Bad file descriptor
    p4_error: latest msg from perror: Bad file descriptor
    p4_error: latest msg from perror: Bad file descriptor
forrtl: error (78): process killed (SIGTERM)
forrtl: error (78): process killed (SIGTERM)
mpiexec: Warning: tasks 0-5 died with signal 11 (Segmentation fault).

We would like this person to help us to resolve these issues.

We would also like to have the consultant double-check our customized 
vanilla Linux kernel (2.6.19.2) on CentOS 4.4 to make sure we configured 
everything correctly.

Besides that, we would also want to have some recommendations on how to 
expand our number of processors to around 300 in terms of architecture, 
networking equipments, cooling, power, and general infrastructure.  
Currently we have  110 proccessors in 39 (2) and 8(4) nodes.

Please contact me if you are in the Bay Area, CA because our company is in 
Redwood City, CA.

Thanks,
NN.

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE! 
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/