Syncup call: review contribution proposals and technical issues

[Frost, Gary]

Ryan

If you believe that this experience with Aparapi can help us define direction/goals/milestones for Sumatra then I think that that input will be very valuable.

My guess is we can learn from users/implementers of many of the existing OpenCL/CUDA bindings.  

Gary

-----Original Message-----
From: sumatra-dev-bounces at openjdk.java.net [mailto:sumatra-dev-bounces at openjdk.java.net] On Behalf Of LaMothe, Ryan R
Sent: Tuesday, November 13, 2012 2:45 PM
To: Harle, Christophe; sumatra-dev at openjdk.java.net
Subject: Re: Syncup call: review contribution proposals and technical issues

We've been using and developing with Aparapi extensively over the past year on a number of high performance computing projects. Is this thread open to a list of our initial impressions with both Aparapi and Java which I think we should discuss tomorrow?

Sorry for the late reply, it has been a hectic couple of weeks.

__________________________________________________

Ryan LaMothe



On 11/8/12 6:18 AM, "Harle, Christophe" <christophe.harle at amd.com> wrote:

>All:
>
>We propose setting up an initial Sumatra project sync-up call next week 
>to review near term contribution plans and discuss outstanding 
>technical issues .
>If you are interested in attending this call, reply to this thread 
>indicating which of the following proposed dates/time you prefer: Nov 
>13 8-10am PST or Nov 14 8-10am PST; we will choose the date that gather 
>the largest number of votes,.
>
>In the spirit of seeding the discussion for this initial call, the AMD 
>team got together and would like to propose the following initial 
>milestone for Sumatra.
>
>This proposed milestone involves moving enough of Aparapi's 
>implementation down into the Project Lambda enabled JVM to allow Lambda 
>enabled Aparapi demos/workloads to execute directly (no Aparapi.dll).
>Whilst this effort would constrain demos/examples to Aparapi's existing 
>programming model constraints (parallel primitive arrays only) it would 
>act as a 'tracer bullet' through the code base, allowing us to 
>familiarize ourselves with the various major JVM components that we 
>think will to impacted via a Sumatra implementation.
>
>Maybe others might like to use this discussion thread to offer other 
>milestone proposals which we could also discuss in the upcoming call.
>
>Regards,
>
>Christophe