Simple strides are a fairly specific use case that apply to dense matrices and 
regular problems.  I am not denying that those problems exist, but they are a 
subset.  As I understand it, regular applications are a shrinking fraction of 
the application domain.

Does any current hardware support strides natively?  If not, maybe we have a 
little time to work out that interface.

From: Jeff Hammond [mailto:[email protected]]
Sent: Thursday, October 29, 2015 3:09 PM
To: Underwood, Keith D <[email protected]>
Cc: Hefty, Sean <[email protected]>; Howard Pritchard <[email protected]>; 
[email protected]
Subject: Re: [ofiwg] strided IOVs

Both GA/ARMCI and SHMEM utilize regular aka simple strided operations in RMA.  
GA_Acc/ARMCI_AccS/MPI_Accumulate map to fi_atomic, that means those also need 
to support strided, assuming RMA does.

I think only simple strided is worth supporting, because it is O(1) metadata.  
General gather/scatter ala iovec is unlikely to optimizable relative to a 
higher-level software implementation, so I don't see a good reason to put it 
into OFI.

Jeff

On Thu, Oct 29, 2015 at 4:46 AM, Underwood, Keith D 
<[email protected]<mailto:[email protected]>> wrote:
Exactly.  However, we should be careful about what we are adding here.  A 
strided interface is a very specific special case that applies to regular 
problems.  There is a bready of MPI derived datatype capabilities that are not 
covered by a simple stride.  What are the objectives for covering datatypes?  
Is the IOVEC interface the right place to glue in the first handling of a 
special case of datatypes?

> -----Original Message-----
> From: 
> [email protected]<mailto:[email protected]>
>  [mailto:ofiwg-<mailto:ofiwg->
> [email protected]<mailto:[email protected]>] On 
> Behalf Of Hefty, Sean
> Sent: Wednesday, October 28, 2015 3:14 PM
> To: Hefty, Sean <[email protected]<mailto:[email protected]>>; Howard 
> Pritchard
> <[email protected]<mailto:[email protected]>>
> Cc: [email protected]<mailto:[email protected]>
> Subject: Re: [ofiwg] strided IOVs
>
> > I need more context to understand how writing X bytes every N bytes
> > into a one-time use receive buffer is useful.
>
> FWIW, someone sent me this link:
>
> http://people.csail.mit.edu/fred/ghost_cell.pdf
>
> as an example.
> _______________________________________________
> ofiwg mailing list
> [email protected]<mailto:[email protected]>
> http://lists.openfabrics.org/mailman/listinfo/ofiwg
_______________________________________________
ofiwg mailing list
[email protected]<mailto:[email protected]>
http://lists.openfabrics.org/mailman/listinfo/ofiwg



--
Jeff Hammond
[email protected]<mailto:[email protected]>
http://jeffhammond.github.io/
_______________________________________________
ofiwg mailing list
[email protected]
http://lists.openfabrics.org/mailman/listinfo/ofiwg

Reply via email to