October 2009 - Mpich2-maint - mailman3.cels.anl.gov

10 annuaires pour votre prospection commercial pour mpich2-maint de mcs
by ymmoo.org 15 Oct '09

15 Oct '09

charset="utf-8" Content-Transfer-Encoding: 8bit Bonjour, Pour developper votre activite, vous avez besoin de trouver de nouveaux clients. L�annuaire-perpetuel vous permet de constituer votre base de donnees prospectives facilement, rapidement et a moindre cout. L�annuaire perpetuel est un logiciel qui capture dans l�annuaire universel* francais ou dans 6 autres annuaires les coordonnees de toutes les entreprises de France, des associations, des professions liberales, des artisans... C�est facile, il suffit de choisir l�annuaire, la categorie et le logiciel va creer votre fichier de prospect. C�est complet, puisque vous disposez de toutes les informations commerciales utiles : nom, adresse, telephone, fax, email... C�est efficace, car les coordonnees de vos clients sont mises a jour a chaque action commerciale. C�est bon marche, car vous n�avez pas de mise a jour ou d�abonnement a payer tous les ans ou tous les semestres. Retrouvez le descriptif complet de notre logiciel sur www.ymmoo.org Christian Clerck Ps : pour ne plus recevoir d�information de notre part, cliquez sur www.ymmoo.org/scripts/d.php?client=mpich2-maint*mcs.anl.gov&s=a

1 0

Re: [mpich2-maint] #897: smpd cannot be installed on windows with a user specified port
by Seifer Lin 15 Oct '09

15 Oct '09

Hi: Even with the bug-fixed, I notice that our problem still exists. Because for a Windows Service (smpd.exe -install as a service), the service name and display name MUST be unique. >From MSDN, http://msdn.microsoft.com/en-us/library/ms682450(VS.85).aspx, the API CreateService(..) SC_HANDLE WINAPI CreateService( __in SC_HANDLE *hSCManager*, __in LPCTSTR *lpServiceName*, __in_opt LPCTSTR *lpDisplayName*, __in DWORD *dwDesiredAccess*, __in DWORD *dwServiceType*, __in DWORD *dwStartType*, __in DWORD *dwErrorControl*, __in_opt LPCTSTR *lpBinaryPathName*, __in_opt LPCTSTR *lpLoadOrderGroup*, __out_opt LPDWORD *lpdwTagId*, __in_opt LPCTSTR *lpDependencies*, __in_opt LPCTSTR *lpServiceStartName*, __in_opt LPCTSTR *lpPassword* ); I believe that every version of MPICH2 - windows edition has the same service name which makes that only one smpd can be installed as the service under Windows. I think the best way is 1. We wait for a bug-fixed, released version of MPICH2 (maybe 1.2.1 ?). 2. We build this new version from source code by ourselves. (to change the service name and display name). By the way, we may have to know how to modify the version of MPICH2. (I mean that the mpiexec.exe and smpd.exe must be the same version, otherwise the "version mismatch" error message will be shown.) best regards, Seifer 2009/10/15 mpich2 <mpich2-maint(a)mcs.anl.gov> > #897: smpd cannot be installed on windows with a user specified port > > --------------------+------------------------------------------------------- > Reporter: jayesh | Owner: jayesh > Type: bug | Status: new > Priority: major | Milestone: > Component: mpich2 | Keywords: > > --------------------+------------------------------------------------------- > > Comment(by jayesh): > > Hi, > This bug should be fixed in the truck now - r5462. This fix will be > included in the next patch release for 1.2 . > Do you build MPICH2 from the source (Do you want me to provide you a > private build of smpd.exe from trunk - Since you distribute your code to > other customers it is recommended that you base your code on a released > version of MPICH2. However you can test your setup with the patch/trunk)? > > Regards, > Jayesh > > -- > Ticket URL: <http://trac.mcs.anl.gov/projects/mpich2/ticket/897#comment:1> > mpich2 <http://www.mcs.anl.gov/research/projects/mpich2> > How to edit this ticket: > http://wiki.mcs.anl.gov/mpich2/index.php/Creating_New_Bug_Tickets >

1 0

ROMIO
by Henry Freund 13 Oct '09

13 Oct '09

Hello World, Is there any documentation for "romio". The issue is that I have a large amount of data on each processor that I would like to output to a single text file. Is it possible to have each processor write to a single file. The order of the writing is unimportant. I have been told that there are calls to a romio library that can do that. Yes? Thanks, Henry Freund

1 0

Re: [mpich2-maint] ROMIO
by Rajeev Thakur 13 Oct '09

13 Oct '09

Since you say "a single text file", I would like to point out that MPI-IO doesn't support formatted I/O like fprintf. Only binary I/O. You can write character arrays as characters. Rajeev > -----Original Message----- > From: mpich2-bugs-bounces(a)mcs.anl.gov > [mailto:[email protected]] On Behalf Of Rob Latham > Sent: Tuesday, October 13, 2009 3:57 PM > To: Henry Freund > Cc: mpich2-maint(a)mcs.anl.gov > Subject: Re: [mpich2-maint] ROMIO > > On Tue, Oct 13, 2009 at 04:14:54PM -0500, Henry Freund wrote: > > Is there any documentation for "romio". The issue is > that I have a > > large amount of data on each processor that I would like to output > > to a single text file. Is it possible to have each processor write > > to a single file. The order of the writing is unimportant. I have > > been told that there are calls to a romio library that can do that. > > Yes? > > Hi Henry Freund > > What you want to do is definitely possible. First, a tiny bit of > background. ROMIO is a library that implements a part of the MPI > standard. the MPI-2 standard contains some I/O routines, which ROMIO > implements. Nowadays, nearly every MPI library supports MPI-IO or > MPI-2 so you shouldn't have to worry about finding or installing > ROMIO > > The MPI standard provides the details of the routines you'd have to > make: > http://www.mpi-forum.org/docs/mpi21-report-bw/node257.htm#Node257 > > The "Using MPI-2" book is the best resource for someone looking to > learn more about MPI-IO features: > > http://mitpress.mit.edu/catalog/item/default.asp?ttype=2&tid=3490 > > It sounds like you want something like "MPI_FILE_WRITE_AT_ALL". > > ==rob > > -- > Rob Latham > Mathematics and Computer Science Division > Argonne National Lab, IL USA > _______________________________________________ > mpich2-maint mailing list > mpich2-maint(a)lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/mpich2-maint >

1 0

Re: [mpich2-maint] ROMIO
by Rob Latham 13 Oct '09

13 Oct '09

On Tue, Oct 13, 2009 at 04:14:54PM -0500, Henry Freund wrote: > Is there any documentation for "romio". The issue is that I have a > large amount of data on each processor that I would like to output > to a single text file. Is it possible to have each processor write > to a single file. The order of the writing is unimportant. I have > been told that there are calls to a romio library that can do that. > Yes? Hi Henry Freund What you want to do is definitely possible. First, a tiny bit of background. ROMIO is a library that implements a part of the MPI standard. the MPI-2 standard contains some I/O routines, which ROMIO implements. Nowadays, nearly every MPI library supports MPI-IO or MPI-2 so you shouldn't have to worry about finding or installing ROMIO The MPI standard provides the details of the routines you'd have to make: http://www.mpi-forum.org/docs/mpi21-report-bw/node257.htm#Node257 The "Using MPI-2" book is the best resource for someone looking to learn more about MPI-IO features: http://mitpress.mit.edu/catalog/item/default.asp?ttype=2&tid=3490 It sounds like you want something like "MPI_FILE_WRITE_AT_ALL". ==rob -- Rob Latham Mathematics and Computer Science Division Argonne National Lab, IL USA

1 0

Undelivered Mail Returned to Sender
by MAILER-DAEMON＠mcs.anl.gov 13 Oct '09

13 Oct '09

This is the Postfix program at host mailgw.mcs.anl.gov. I'm sorry to have to inform you that your message could not be be delivered to one or more recipients. It's attached below. For further assistance, please send mail to <postmaster> If you do so, please include this problem report. You can delete your own text from the attached returned message. The Postfix program <martin.audet(a)imi.nrc-cnrc.gc.ca>: connect to imi.nrc-cnrc.gc.ca[132.246.11.112]: Connection timed out

1 0

Undelivered Mail Returned to Sender
by MAILER-DAEMON＠mcs.anl.gov 13 Oct '09

13 Oct '09

This is the Postfix program at host mailgw.mcs.anl.gov. I'm sorry to have to inform you that your message could not be be delivered to one or more recipients. It's attached below. For further assistance, please send mail to <postmaster> If you do so, please include this problem report. You can delete your own text from the attached returned message. The Postfix program <martin.audet(a)imi.nrc-cnrc.gc.ca>: connect to imi.nrc-cnrc.gc.ca[132.246.11.112]: Connection timed out

1 0

Undelivered Mail Returned to Sender
by MAILER-DAEMON＠mcs.anl.gov 12 Oct '09

12 Oct '09

This is the Postfix program at host mailgw.mcs.anl.gov. I'm sorry to have to inform you that your message could not be be delivered to one or more recipients. It's attached below. For further assistance, please send mail to <postmaster> If you do so, please include this problem report. You can delete your own text from the attached returned message. The Postfix program <martin.audet(a)imi.nrc-cnrc.gc.ca>: connect to imi.nrc-cnrc.gc.ca[132.246.11.112]: Connection timed out

1 0

Undelivered Mail Returned to Sender
by MAILER-DAEMON＠mcs.anl.gov 12 Oct '09

12 Oct '09

This is the Postfix program at host mailgw.mcs.anl.gov. I'm sorry to have to inform you that your message could not be be delivered to one or more recipients. It's attached below. For further assistance, please send mail to <postmaster> If you do so, please include this problem report. You can delete your own text from the attached returned message. The Postfix program <martin.audet(a)imi.nrc-cnrc.gc.ca>: connect to imi.nrc-cnrc.gc.ca[132.246.11.112]: Connection timed out

1 0

Errors while install MPICH on a 64bit cluster
by Lei Yang 12 Oct '09

12 Oct '09

Hello, This is Lei Yang (at Guang Zhou, China) I am trying to install MPICH or MPICH2 on our cluster, but many errors came up, please help me!! attached c.txt, m.txt, mi.txt are for configuration, make and make install results. (i has been trying to install this for days, please do help!!~) I think the main issue is that it didn't pass configure test, and showed 'can't find -lfmpich' I use intel fortran compiler. my .bashrc file is like this export F90=/opt/intel/fc/10.0.023/bin/ifort export FC=F90 export CC=gcc export F77=F90 and my cluster information is: Processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 23 model name : Intel(R) Xeon(R) CPU E5430 @ 2.66GHz stepping : 10 cpu MHz : 2660.002 cache size : 6144 KB physical id : 0 siblings : 4 core id : 0 cpu cores : 4 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush d ts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm constant_tsc pni monitor ds_cpl vmx est tm2 cx16 x tpr lahf_lm bogomips : 5323.69 clflush size : 64 cache_alignment : 64 address sizes : 38 bits physical, 48 bits virtual power management: processor : 1 vendor_id : GenuineIntel cpu family : 6 model : 23 model name : Intel(R) Xeon(R) CPU E5430 @ 2.66GHz stepping : 10 cpu MHz : 2660.002 cache size : 6144 KB physical id : 0 siblings : 4 core id : 2 cpu cores : 4 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush d ts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm constant_tsc pni monitor ds_cpl vmx est tm2 cx16 x tpr lahf_lm bogomips : 5320.00 clflush size : 64 cache_alignment : 64 address sizes : 38 bits physical, 48 bits virtual power management: processor : 2 vendor_id : GenuineIntel cpu family : 6 model : 23 model name : Intel(R) Xeon(R) CPU E5430 @ 2.66GHz stepping : 10 cpu MHz : 2660.002 cache size : 6144 KB physical id : 0 siblings : 4 core id : 1 cpu cores : 4 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush d ts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm constant_tsc pni monitor ds_cpl vmx est tm2 cx16 x tpr lahf_lm bogomips : 5320.06 clflush size : 64 cache_alignment : 64 address sizes : 38 bits physical, 48 bits virtual power management: processor : 3 vendor_id : GenuineIntel cpu family : 6 model : 23 model name : Intel(R) Xeon(R) CPU E5430 @ 2.66GHz stepping : 10 cpu MHz : 2660.002 cache size : 6144 KB physical id : 0 siblings : 4 core id : 3 cpu cores : 4 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush d ts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm constant_tsc pni monitor ds_cpl vmx est tm2 cx16 x tpr lahf_lm bogomips : 5320.01 clflush size : 64 cache_alignment : 64 address sizes : 38 bits physical, 48 bits virtual power management: Thanks very much!!! ~ ------------------------------------------------------------ LED The South China Sea Institute of Oceanology, CAS Guangzhou, China, 510301 Tel:86-20-89023199

1 0