[SIESTA-L] Strange mpi problem
Hello, Dear all, These days, I was trying to run a calculation with the parallel version of siesta on a PC cluster. It was compiled by mpich2 and pgi compiler. What surprises me is that, sometimes, it runs normally, and sometimes, the task enters a sleeping status right after I submit the job by mpiexec -n 8 siesta input output. In the output of the command top, I can see S as the status on the line of my job. At this time ,it never goes on. This happens very frequently, even for the same input file. I do not know why. Could you tell me how to avoid entering a sleeping state right after the submission of the job, please? Thank you very much!! Sincerely, Lakee
Re: [SIESTA-L] Strange mpi problem
I've got the same problem when I was trying to compute optical properties even in the serial version ... simply dont know what happens, but the calculation just sleep ... NH On Wed, Aug 20, 2008 at 4:06 PM, Lakee Johnson [EMAIL PROTECTED]wrote: Hello, Dear all, These days, I was trying to run a calculation with the parallel version of siesta on a PC cluster. It was compiled by mpich2 and pgi compiler. What surprises me is that, sometimes, it runs normally, and sometimes, the task enters a sleeping status right after I submit the job by mpiexec -n 8 siesta input output. In the output of the command top, I can see S as the status on the line of my job. At this time ,it never goes on. This happens very frequently, even for the same input file. I do not know why. Could you tell me how to avoid entering a sleeping state right after the submission of the job, please? Thank you very much!! Sincerely, Lakee
[SIESTA-L] Strange mpi problem
Dear Lakee, Try with one of the two following commands !! mpiexec -np 8 siesta input output or mpirun -np 8 siesta input output Hello, Dear all, These days, I was trying to run a calculation with the parallel version of siesta on a PC cluster. It was compiled by mpich2 and pgi compiler. What surprises me is that, sometimes, it runs normally, and sometimes, the task enters a sleeping status right after I submit the job by mpiexec -n 8 siesta input output. In the output of the command top, I can see S as the status on the line of my job. At this time ,it never goes on. This happens very frequently, even for the same input file. I do not know why. Could you tell me how to avoid entering a sleeping state right after the submission of the job, please? Thank you very much!! Sincerely, Lakee _ Envoyez avec Yahoo! Mail. Une boite mail plus intelligente http://mail.yahoo.fr
[SIESTA-L] Re : Strange mpi problem
Dear Lakee, Try with one of the two following commands !! mpiexec -np 8 siesta input output or mpirun -np 8 siesta input output Karim _ Envoyez avec Yahoo! Mail. Une boite mail plus intelligente http://mail.yahoo.fr
Re: [SIESTA-L] Strange mpi problem
Lakee, I cannt speak about the serial sleep, but I have encountered similar problems trying to run parallel SIESTA using MPICH2 as my MPI interface. I would recommend considering one of the other MPI programs and see if this doesn't solve the problem. Myself I use both OpenMPI and MVAPICH. Depending on how aggressively the libraries are compiled, the latter tends to run a little faster than the former, but it is not really that significant of a difference. I would be interested to hear what the other users have to say concerning this issue. Tom Sadowski University of Connecticut Date: Wed, 20 Aug 2008 22:06:20 +0800 From: [EMAIL PROTECTED] Subject: [SIESTA-L] Strange mpi problem To: SIESTA-L@listserv.uam.es Hello, Dear all, These days, I was trying to run a calculation with the parallel version of siesta on a PC cluster. It was compiled by mpich2 and pgi compiler. What surprises me is that, sometimes, it runs normally, and sometimes, the task enters a sleeping status right after I submit the job by mpiexec -n 8 siesta input output. In the output of the command top, I can see S as the status on the line of my job. At this time ,it never goes on. This happens very frequently, even for the same input file. I do not know why. Could you tell me how to avoid entering a sleeping state right after the submission of the job, please? Thank you very much!! Sincerely, Lakee _ See what people are saying about Windows Live. Check out featured posts. http://www.windowslive.com/connect?ocid=TXT_TAGLM_WL_connect2_082008
[SIESTA-L] Error in siesta compilation
Hi all, I am trying to compile siesta (actually the compilation goes fine) in an 8-processor intel machine (Intel(R) Xeon(R) CPU X5355 @ 2.66GHz). The compiler is ifort version 10.1.017 and libraries are mkl 10.0.3.020 plus the usual self-compiled blacs and scalapack. However, I get an error when I try to run siesta in parallel, no matter how many nodes (even at 1 node!): forrtl: severe (174): SIGSEGV, segmentation fault occurred Image PCRoutineLineSource siesta 010311C7 Unknown Unknown Unknown libpthread.so.02AD287BB8FB0 Unknown Unknown Unknown siesta 010311A3 Unknown Unknown Unknown libpthread.so.02AD287BB8FB0 Unknown Unknown Unknown Has anyone had this problem with the 10.0.3 version of the mkl? I ask that because I have parallel siesta running smoothly in dual-core laptops and desktops, and haven't had this problem. I see that with the newest version of the intel mkl, as usual the names of the libraries one has to add changed, and now I have to add -lmkl_core and -lpthread if I want the compilation to end with no errors. Has anyone had this problem, and knows how to solve it? Cheers, Marcos -- Dr. Marcos Verissimo Alves Post-Doctoral Fellow Unité de Physico-Chimie et de Physique des Matériaux (PCPM) Université Catholique de Louvain 1 Place Croix du Sud, B-1348 Louvain-la-Neuve Belgique -- Gort, Klaatu barada nikto. Klaatu barada nikto. Klaatu barada nikto. Free translation: Gort, Google is your friend. Google is your friend. Google is your friend.