[mpich-discuss] Force MPICH2 to not use network adapter

Balaji, Pavan balaji at anl.gov
Fri Nov 7 17:14:06 CST 2014


Hi Igor,

Unfortunately, Windows is not supported anymore in MPICH.

http://wiki.mpich.org/mpich/index.php/Frequently_Asked_Questions#Q:_Why_can.27t_I_build_MPICH_on_Windows_anymore.3F

But 1.4.1p1 should be automatically detecting and using shared memory inside the node, and not using the network. Sorry, I know that doesn’t help you.  But this is an unsupported platform and version of MPICH.

Regards,

  — Pavan

> On Nov 7, 2014, at 10:07 AM, Igor Raskin <igor.raskin at weblakes.com> wrote:
> 
> Hi Pavan,
> 
> Thank you for your response.
> 
> I should have given more details about the platform and MPICH2 I am using.
> 
> The platform is Windows 7 Professional.
> 
> The version of MPICH2 is 1.4.1 (32-bit) which is latest version of MPICH2 that I can find.
> 
> To what version of MPICH2 should I upgrade?
> 
> Thank you and best regards,
> Igor
> 
> -----Original Message-----
> From: Balaji, Pavan [mailto:balaji at anl.gov] 
> Sent: Friday, November 07, 2014 12:50 AM
> To: discuss at mpich.org
> Subject: Re: [mpich-discuss] Force MPICH2 to not use network adapter
> 
> Igor,
> 
> Based on the fact that you are asking about smpd and -localonly options, I’m assuming you are using a super-ancient vesion of MPICH that dinosaurs used to compute with.  Perhaps you can upgrade to the latest version?  For the past 10 years or so, we have been automatically detecting shared memory and using it internally without going over the network.
> 
>  — Pavan
> 
>> On Nov 6, 2014, at 8:28 PM, Igor Raskin <igor.raskin at weblakes.com> wrote:
>> 
>> Hello,
>> 
>> I am running a multi-process calculations using MPICH2 from Argonne National Laboratory. The run is on a single machine, so -localonly option of mpiexec is used. Usually everything works.
>> If the network adapter is enabled when the run starts, and if I disable it during the run, the run fails with error stating:
>> 
>> op_read error on left context: Error = -1
>> 
>> op_read error on parent context: Error = -1
>> 
>> unable to read the cmd header on the left context, Error = -1 .
>> unable to read the cmd header on the parent context, Error = -1 .
>> Error posting readv, An existing connection was forcibly closed by the 
>> remote host.(10054) connection to my parent broken, aborting.
>> state machine failed.
>> However, if the network adapter is disabled when the run is started, I can enable/disable the adapter as many times as I want, and the run still proceeds to the end.
>> 
>> Is there a way to run mpiexec or modify smpd configuration such that MPICH2 is not using network adapter for inter-process communication for local runs even if the adapter is available when the run starts?
>> 
>> Thank you,
>> 
>> Igor
>> _______________________________________________
>> discuss mailing list     discuss at mpich.org
>> To manage subscription options or unsubscribe:
>> https://lists.mpich.org/mailman/listinfo/discuss
> 
> --
> Pavan Balaji  ✉️
> http://www.mcs.anl.gov/~balaji
> 
> _______________________________________________
> discuss mailing list     discuss at mpich.org
> To manage subscription options or unsubscribe:
> https://lists.mpich.org/mailman/listinfo/discuss
> _______________________________________________
> discuss mailing list     discuss at mpich.org
> To manage subscription options or unsubscribe:
> https://lists.mpich.org/mailman/listinfo/discuss

--
Pavan Balaji  ✉️
http://www.mcs.anl.gov/~balaji

_______________________________________________
discuss mailing list     discuss at mpich.org
To manage subscription options or unsubscribe:
https://lists.mpich.org/mailman/listinfo/discuss


More information about the discuss mailing list