Perhaps if you, me, John C, mabshoff and the people he is working with all signed off on it.
I could also mention the "seed funding" EPSRC have given me through my grant, for hardware and my salary, specifically for developing "fast core arithmetic for parallel processors and platforms". We could actually make the application look quite impressive I think. Bill. 2008/11/23 Bill Hart <[EMAIL PROTECTED]>: > Sure I just emailed John C, being a professor and all, he could sign > off on it. We are also planning on buying some really fast hardware > between us early next year which could be made compatible. > > Note they really want to see that we are promoting/teaching CUDA and > have books on the subject. But I think that if we wrote up a pdf doc > outlining how CUDA would be promoted through the eMPIRe project and > Sage and hopefully work it's way upstream to Pari/FLINT/blah blah.... > > I think we should also ask for an unrestricted gift equal to 1/3 of a > year's salary to employ someone for build system/repo > management/website/promotion, etc. We would outline how part of their > job would be to promote CUDA/NVIDIA etc through the eMPIRe website, > and upstream partnerships. > > Note Mathematica already runs on these machines!! > > Bill. > > 2008/11/23 William Stein <[EMAIL PROTECTED]>: >> >> On Sun, Nov 23, 2008 at 12:14 PM, Bill Hart <[EMAIL PROTECTED]> wrote: >>> >>> Right, but call me odd or whatever, but I think this can be used for >>> exact integer arithmetic too. >>> >>> Thanks for the professor link. That looks very promising indeed! >>> >>> Bill. >> >> Let's put in an application!!! >> >>> >>> 2008/11/23 mabshoff <[EMAIL PROTECTED]>: >>>> >>>> On Nov 23, 12:04 pm, Bill Hart <[EMAIL PROTECTED]> wrote: >>>> >>>> Hi, >>>> >>>>> Here is a machine I would like to acquire and develop an eMPIRe port >>>>> for. >>>>> >>>>> http://www.nvidia.com/object/personal_supercomputing.html >>>>> >>>>> Anyone else interested in this sort of thing? >>>> >>>> Very much for fast exact linear algebra, i.e. think LinBox. Clement >>>> and I have worked on hooking CUDABLAS into LinBox and we got it >>>> working, but since the hardware we have access to was not IEEE conform >>>> and slower than the main CPU in that case we stepped back for a while. >>>> >>>> There is also http://www.nvidia.com/page/professor_partnership.html >>>> >>>>> Bill. >>>> >>>> Cheers, >>>> >>>> Michael >>>> > >>>> >>> >>> > >>> >> >> >> >> -- >> William Stein >> Associate Professor of Mathematics >> University of Washington >> http://wstein.org >> >> >> >> > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "mpir-devel" group. To post to this group, send email to mpir-devel@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/mpir-devel?hl=en -~----------~----~----~----~------~----~------~--~---
