I would advise pre-posting MPI_Irecv's if you know that messages will be coming.

Barriers should only be used if you need to synchronize between groups of 
processes.  Keep in mind that MPI is a lossless/reliable message delivery 
mechanism, so there's no need for you to call additional error-correcting API 
functions to ensure that you get all the messages that were sent.

Additionally, Iprobe is typically only useful if you don't know the size / 
sender of the message.  Calling Iprobe almost guarantees that MPI will need to 
make an internal copy of the message.  So avoid probe / iprobe if you can.



On Jun 28, 2010, at 5:16 PM, amjad ali wrote:

> I guess that if the receiver want to ensure that the sender should send data 
> only when the receiver will be able/free to receive data,
> 
> then use MPI-Barriers.
> 
>  
> 
> On Mon, Jun 28, 2010 at 12:53 PM, David Zhang <solarbik...@gmail.com> wrote:
> Use MPI_Iprobe.  It's a nonblocking probe that allow you to see if a message 
> is coming.  So in your receiver code you just periodically probe to see if 
> messages are coming your way.
> 
> 
> On Sun, Jun 27, 2010 at 9:11 PM, Jack Bryan <dtustud...@hotmail.com> wrote:
> thanks
> 
> I know that. 
> 
> MPI_irecv() ;
> 
> do other works;
> 
> MPI_wait();
> 
> But, my message receiver is much slower than sender. 
> 
> when the sender is doing its local  works, the sender has sent out their 
> messages. but at this time, 
> the sender is very busy doing its local work and cannot post MPI_irecv to get 
> the messages from 
> senders. 
> 
> Any help is appreciated. 
> 
> jack
> 
> 
> 
> From: jiangzuo...@gmail.com
> Date: Mon, 28 Jun 2010 11:55:32 +0800
> 
> To: us...@open-mpi.org
> Subject: Re: [OMPI users] openMPI asychronous communication
> 
> OK, then i think you also know using MPI_Wait to wait the asynchronous 
> requests to complete. if sender works faster then receiver(or reverse), then 
> the MPI_Wait will do wait, not just deallocted. you should keep the buffer 
> content before MPI_Wait.
> 
>                                                      Changsheng Jiang
> 
> 
> On Mon, Jun 28, 2010 at 11:41, Jack Bryan <dtustud...@hotmail.com> wrote:
> thanks
> 
> I know this. 
> 
> but, what if sender can send a lot of messages to receivers faster than what 
> receiver can receive ? 
> 
> it means that sender works faster than receiver. 
> 
> Any help is appreciated. 
> 
> jack 
> 
> 
> From: jiangzuo...@gmail.com
> Date: Mon, 28 Jun 2010 11:31:16 +0800
> To: us...@open-mpi.org
> Subject: Re: [OMPI users] openMPI asychronous communication
> 
> 
> MPI_Isend - Starts a standard-mode, nonblocking send.
> 
> BTW, are there any asynchronous collective operations?
> 
>                                                      Changsheng Jiang
> 
> 
> On Mon, Jun 28, 2010 at 11:22, Jack Bryan <dtustud...@hotmail.com> wrote:
> Dear All:
> 
> How to do asychronous communication among nodes by openMPI or boot.MPI  in 
> cluster ?
> 
> I need to set up a kind of asychronous communication protocol such that 
> message senders and receivers can communicate asychronously without losing any
> messages between them. 
> 
> I do not want to use blocking MPI routines because the processors can do other
> operations when they wait for new messages coming.
> 
> I donot find this kind of MPI routines that support this 
> asychronous communication. 
> 
> Any help is appreciated. 
> 
> thanks
> 
> Jack 
> 
> June 27  2010
> 
> The New Busy think 9 to 5 is a cute idea. Combine multiple calendars with 
> Hotmail. Get busy.
> 
> _______________________________________________
> users mailing list
> us...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/users
> 
> 
> The New Busy think 9 to 5 is a cute idea. Combine multiple calendars with 
> Hotmail. Get busy.
> 
> _______________________________________________
> users mailing list
> us...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/users
> 
> 
> The New Busy think 9 to 5 is a cute idea. Combine multiple calendars with 
> Hotmail. Get busy.
> 
> _______________________________________________
> users mailing list
> us...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/users
> 
> 
> 
> -- 
> David Zhang
> University of California, San Diego
> 
> _______________________________________________
> users mailing list
> us...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/users
> 
> _______________________________________________
> users mailing list
> us...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/users


-- 
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/


Reply via email to