mfaktc: a CUDA program for Mersenne prefactoring - Page 184

Roy_Sirl · 2013-01-06, 22:41

Excellent - a red letter day indeed! Many thanks to all those involved.

swl551 · 2013-01-06, 22:49

Fasther throughput on i5 (bottlnecked with 4 instances using 0.19)
CPU doesn't even throttle up (power draw 6.4 watts)
LOWER overall system power consumption by 100 watts!

OUTSTANDING RESULTS!

Oh and nvidia 310.90 just came out.

firejuggler · 2013-01-06, 22:57

Geforce 560, option by default, 200 Ghz-d by day

kladner · 2013-01-06, 23:01

Still running -st2 on both GPUs. GTX 570 at about 55%, 460 about 75%. Affinity not set. CPU at about 20-25%. The 570 finished. With the 460 still running it's pulling 10-12% CPU.

EDIT: Now running an assignment on the 570, and there is negligible CPU usage for that instance. Just -st2 is using more CPU.

TheJudger · 2013-01-06, 23:05

Quote:

Originally Posted by kladner

Still running -st2 on both GPUs. GTX 570 at about 55%, 460 about 75%. Affinity not set. CPU at about 20-25%. The 570 finished. With the 460 still running it's pulling 10-12% CPU.

-st/-st2 tests both, CPU-sieve and GPU-sieve kernels.

Oliver

Chuck · 2013-01-07, 00:02

Quote:

Originally Posted by swl551

Fasther throughput on i5 (bottlnecked with 4 instances using 0.19)
CPU doesn't even throttle up (power draw 6.4 watts)
LOWER overall system power consumption by 100 watts!

OUTSTANDING RESULTS!

Oh and nvidia 310.90 just came out.

I upgraded to 310.70 two weeks ago, and it gave me slightly poorer performance (with the non-GPU sieving mfaktc) than the earlier 306.97 version. I reverted back to the earlier version. I guess I could change drivers again with this new mfaktc and see what the results are.

Chuck

swl551 · 2013-01-07, 00:07

Quote:

Originally Posted by Chuck

I upgraded to 310.70 two weeks ago, and it gave me slightly poorer performance (with the non-GPU sieving mfaktc) than the earlier 306.97 version. I reverted back to the earlier version. I guess I could change drivers again with this new mfaktc and see what the results are.

Chuck

310.70 had a short life, suggesting it did have some problems.

kladner · 2013-01-07, 00:21

Quote:

Originally Posted by swl551

Oh and nvidia 310.90 just came out.

310.70 had a short life, suggesting it did have some problems.

That's very good to know. Thanks! I'm running 310.70, but will upgrade.

@ Oliver- Thanks for the explanation. I'm now running assignments on both cards:

Code:

64.8M 69-73 assignments -both cards
GTX 570 6.004s 415 GHz-D/D  GPU 95%  823 MHz  71 C
GTX 460 12.135s  205 GHz-D/D  GPU 98%  830 MHz  70 C

CPU is essentially idle.

The temperatures above are about 2-3 C higher than with 4 instances of mfaktc 0.19 on the 570, and 2 instances on the 460. Considering that the CPU is idle, they seem to be working harder. On the other hand, a cool CPU means that the case fans throttled back.

This looks like a big jump in throughput, and I haven't even gotten P95 running yet. I now have up to six cores to run P-1, or something.

swl551 · 2013-01-07, 00:59

0.19 running 4 instances on GTX-570 988mv@850mhz with i7-2600k @ 4.2ghz
yielded 118ghzDay per instance = 472ghzDay total

0.20 running 1 instance on GTX-570 988mv@850mhz with i7-2600k @ 4.2ghz
yielding 427ghzDay. down 45ghzDay

However on my i5 2500k 0.20 increased by 25ghzDay compared to 0.19. (cpu was bottlenecked)

kracker · 2013-01-07, 01:03

Quote:

Originally Posted by swl551

0.20 running 1 instance on GTX-570 988mv@850mhz with i7-2600k @ 4.2ghz
yielding 427ghzDay. down 45ghzDay

But now maybe you can run something on the cpu!

EDIT: Er, that is using the part of da cpu that 0.19 *used* to take.

kladner · 2013-01-07, 01:50

With 6x 0.19 on the 570, and 2x on the 460 I was getting a combined GHz-D/D of 530-535 (per mfaktc readings). With 2.0 (32 bit -slightly better than 64 bit) those mfaktc readings total about 625 GHz-D/D with 1 instance per GPU. I now have 5 workers running P-1 in P95. This last item did not seem make any difference in the mfaktc 2.0 performance.

EDIT: [OT]PrimeNet still keeps sticking in 1 DC assignment (Worker #3) to 4 P-1s in P95. All the settings I can find, both online and in P95, are set to P-1. I have worked around this by moving all the assignments from the other 4 workers to Worker #3. This stops #3 from getting more assignments for now, and the other 4 then fill in with P-1s.[/OT]

EDIT2: Sieve is running 82,485 on both GPUs, with SievePrimesAdjust=1. Otherwise the settings are default except for CheckpointDelay=300, Stages=0, and StopAfterFactor=2.

2013-01-06, 22:49	#2015
swl551 Aug 2012 New Hampshire 2³·101 Posts	Outstanding Fasther throughput on i5 (bottlnecked with 4 instances using 0.19) CPU doesn't even throttle up (power draw 6.4 watts) LOWER overall system power consumption by 100 watts! OUTSTANDING RESULTS! Oh and nvidia 310.90 just came out. Attached Thumbnails

2013-01-06, 22:57	#2016
firejuggler Apr 2010 Over the rainbow 2×1,303 Posts	Geforce 560, option by default, 200 Ghz-d by day Attached Thumbnails

2013-01-06, 23:01	#2017
kladner "Kieren" Jul 2011 In My Own Galaxy! 27AE₁₆ Posts	Still running -st2 on both GPUs. GTX 570 at about 55%, 460 about 75%. Affinity not set. CPU at about 20-25%. The 570 finished. With the 460 still running it's pulling 10-12% CPU. EDIT: Now running an assignment on the 570, and there is negligible CPU usage for that instance. Just -st2 is using more CPU. Last fiddled with by kladner on 2013-01-06 at 23:06

2013-01-07, 00:59	#2022
swl551 Aug 2012 New Hampshire 2³·101 Posts	comparison 0.19 running 4 instances on GTX-570 988mv@850mhz with i7-2600k @ 4.2ghz yielded 118ghzDay per instance = 472ghzDay total 0.20 running 1 instance on GTX-570 988mv@850mhz with i7-2600k @ 4.2ghz yielding 427ghzDay. down 45ghzDay However on my i5 2500k 0.20 increased by 25ghzDay compared to 0.19. (cpu was bottlenecked) Last fiddled with by swl551 on 2013-01-07 at 01:00

2013-01-07, 01:50	#2024
kladner "Kieren" Jul 2011 In My Own Galaxy! 2×3×1,693 Posts	With 6x 0.19 on the 570, and 2x on the 460 I was getting a combined GHz-D/D of 530-535 (per mfaktc readings). With 2.0 (32 bit -slightly better than 64 bit) those mfaktc readings total about 625 GHz-D/D with 1 instance per GPU. I now have 5 workers running P-1 in P95. This last item did not seem make any difference in the mfaktc 2.0 performance. EDIT: [OT]PrimeNet still keeps sticking in 1 DC assignment (Worker #3) to 4 P-1s in P95. All the settings I can find, both online and in P95, are set to P-1. I have worked around this by moving all the assignments from the other 4 workers to Worker #3. This stops #3 from getting more assignments for now, and the other 4 then fill in with P-1s.[/OT] EDIT2: Sieve is running 82,485 on both GPUs, with SievePrimesAdjust=1. Otherwise the settings are default except for CheckpointDelay=300, Stages=0, and StopAfterFactor=2. Last fiddled with by kladner on 2013-01-07 at 02:10

2013-01-06, 22:41	#2014
Roy_Sirl Sep 2002 Cornwall, UK 2·5 Posts	Excellent - a red letter day indeed! Many thanks to all those involved.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
mfakto: an OpenCL program for Mersenne prefactoring	Bdot	GPU Computing	1676	2021-06-30 21:23
The P-1 factoring CUDA program	firejuggler	GPU Computing	753	2020-12-12 18:07
gr-mfaktc: a CUDA program for generalized repunits prefactoring	MrRepunit	GPU Computing	32	2020-11-11 19:56
mfaktc 0.21 - CUDA runtime wrong	keisentraut	Software	2	2020-08-18 07:03
World's second-dumbest CUDA program	fivemack	Programming	112	2015-02-12 22:51