Message boards :
Number crunching :
Opinions wanted
Message board moderation
Author | Message |
---|---|
Mike Sherman* Send message Joined: 27 Jan 10 Posts: 6 Credit: 510,927 RAC: 0 |
SuperMicro X8DAL-i dual processor to be used strictly for crunching. Better suggestions? Thanks, Mike |
hiamps Send message Joined: 23 May 99 Posts: 4292 Credit: 72,971,319 RAC: 0 |
A board with as many PCIe slots with big video cards and a big Power Supply. GPU crunching is faster than CPU. If I had room for just one more video card.....LOL Official Abuser of Boinc Buttons... And no good credit hound! |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
A board with as many PCIe slots with big video cards and a big Power Supply. GPU crunching is faster than CPU. If I had room for just one more video card.....LOL So long as it's not VHAR or VLAR work :P SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14653 Credit: 200,643,578 RAC: 874 |
A board with as many PCIe slots with big video cards and a big Power Supply. GPU crunching is faster than CPU. If I had room for just one more video card.....LOL VHAR runs absolutely brilliantly on GPU, especially with the v2.3 runtime DLLs. VLAR, on the other hand - there I agree with you. Avoid at all costs. |
hiamps Send message Joined: 23 May 99 Posts: 4292 Credit: 72,971,319 RAC: 0 |
A board with as many PCIe slots with big video cards and a big Power Supply. GPU crunching is faster than CPU. If I had room for just one more video card.....LOL That is what I thought, and Jason is working on the VLar situation or at least was... Official Abuser of Boinc Buttons... And no good credit hound! |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
A board with as many PCIe slots with big video cards and a big Power Supply. GPU crunching is faster than CPU. If I had room for just one more video card.....LOL Ah, I thought VHARs were not so good as well. Since they included the "Force HVAR to CPU also" option in Reschedule v1.9. I use that option on the one machine I have at work with a CUDA card. Just let it D/L GPU work and dump VHARs & VLARs to the CPU. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
hiamps Send message Joined: 23 May 99 Posts: 4292 Credit: 72,971,319 RAC: 0 |
A board with as many PCIe slots with big video cards and a big Power Supply. GPU crunching is faster than CPU. If I had room for just one more video card.....LOL I do the same but only dump the VLars to CPU and that can still be too many. Official Abuser of Boinc Buttons... And no good credit hound! |
jason_gee Send message Joined: 24 Nov 06 Posts: 7489 Credit: 91,093,184 RAC: 0 |
A board with as many PCIe slots with big video cards and a big Power Supply. GPU crunching is faster than CPU. If I had room for just one more video card.....LOL VHAR will depend on GPU vs CPU vs overhead performance as to where it's best to send. VLAR, yes I have been working on the situation in ultra slow motion. Apart from personal commitments delaying things, some redesign is required (pointed to by some recent experiments) and that'll need to factor in soon to be released hardware (lot's of reading and 'guessing'). "Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions. |
Fred J. Verster Send message Joined: 21 Apr 04 Posts: 3252 Credit: 31,903,643 RAC: 0 |
VHAR runs absolutely brilliantly on GPU, especially with the v2.3 runtime DLLs. VLAR, on the other hand - there I agree with you. Avoid at all costs. That is what I thought, and Jason is working on the VLar situation or at least was... VHAR will depend on GPU vs CPU vs overhead performance as to where it's best to send. VLAR, yes I have been working on the situation in ultra slow motion. Apart from personal commitments delaying things, some redesign is required (pointed to by some recent experiments) and that'll need to factor in soon to be released hardware (lot's of reading and 'guessing'). Posting about (CUDA)-Error's A DUAL-CPU xATX Board*, with 3 or 4 PCI-E x16/16/16/8, slots or a PCI-E X-Slot. And if available and not too expensivea Back-Plane and 2 or 3 PSU's. *2x XEON; 2x i7(?) or 2x Q(X)9650/9770/9775 or newer, 6 core CPU. |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.