Posts by pharrg

21) Message boards : BOINC client : Designtate different projects for CPU and GPU? (Message 22463)
Posted 15 Jan 2009 by pharrg
Post:
Are there plans to add the ability to designtate that the CPU's run one project, and the GPU's run another? I would like to let my CPUs run NON-CUDA work while my nVidia GPUs run CUDA work from a different project. I'd prefer this over simply alternating between projects since this way each type of processor can simultaneously work on what it's best suited for non-stop.
22) Message boards : BOINC Manager : Extreme CUDA processing!!! (Message 22361)
Posted 9 Jan 2009 by pharrg
Post:
For the specs and more info, such as the power and heat you asked about, here's links to nVidia's pages

GTX 295 Video Cards
(note that many manufacturers, including EVGA, XFX, Gigabyte, and more are making them)
http://www.nvidia.com/object/product_geforce_gtx_295_us.html

Tesla (teraflop+ parallel multicore supercomputing)
http://www.nvidia.com/object/tesla_computing_solutions.html

CUDA web page
http://www.nvidia.com/object/cuda_what_is.html

Hope this helps!

23) Message boards : BOINC Manager : Extreme CUDA processing!!! (Message 22353)
Posted 9 Jan 2009 by pharrg
Post:
If someone is wanting the fastest processing of CUDA units, the new eVGA GTX 295 boards were released today. A dual GPU board with a whopping 480 cores processing in parallel, 2Gb of memory, and memory bandwidth of over 223Gb/sec! Heck, get two of them and hook them up in SLI.

If you want to get really extreme, install 4 of nVidia's Tesla C1060 boards into a single desktop pc and you now have a machine with 960 cores capable of an incredible 3.732 Teraflops (single precision). Yes, you'd essentially have a massively parallel supercomputer sitting on your desk.

Let's see who's really serious about cranking work units out!
24) Message boards : BOINC Manager : Will CUDA specific preferences be added? (Message 22345)
Posted 8 Jan 2009 by pharrg
Post:
Hi! I was wondering if there were plans to add new fields in the preferences on the web page and in the app menu to tweak the CUDA settings now that several projects support parallel processing on nVidia GPUs?

Here are some examples of what I'd like to see:

Ability to set BOINC to use nVidia GPU processors and CUDA only and not use main CPUs at all.

Ability to set usage settings and levels for nVidia GPUs seperate from CPUs.

Just like CPU settings, be able to set usage levels and times for GPU's as well (mainly to limit heat since some of the newer GPUs such as the GTX 200 series run hot anyway).

Display which units are working in CUDA vs regular CPUs.

Also, wondering if the graphics do, or will soon, be able to take advantage of directX or either nVidia or ATI acceleration. I still notice a major slowdown in unit processing speed when I leave graphics on even though I have high end video card. Wondering if it's even being used? Or is that specific to the project apps? Thanks!



Previous 20

Copyright © 2024 University of California.
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.