mersenneforum.org Msieve GPU LA
 Register FAQ Search Today's Posts Mark Forums Read

2021-09-24, 06:17   #56
frmky

Jul 2003
So Cal

22×547 Posts

Quote:
 Originally Posted by charybdis @frmky, for future reference, when I tested this I found that rational side sieving with *algebraic* 3LP was fastest. This shouldn't be too much of a surprise: the rational norms are larger, but not so much larger that 6 large primes across the two sides should split 4/2 rather than 3/3 (don't forget the special-q is a "free" large prime).
I'll try that, thanks!

2021-09-24, 06:21   #57
frmky

Jul 2003
So Cal

22·547 Posts

Quote:
 Originally Posted by frmky filtering yielded Code: matrix is 102063424 x 102063602 (51045.3 MB) with weight 14484270868 (141.91/col) Normally I'd try to bring this down, but testing on a quad V100 system with NVLink gives Code: linear algebra completed 2200905 of 102060161 dimensions (2.2%, ETA 129h 5m)
And it's done. LA on the 102M matrix with restarts took 5 days 14 hours.

2021-09-24, 12:36   #58
charybdis

Apr 2020

49310 Posts

Quote:
 Originally Posted by frmky I'll try that, thanks!
Also 250M is very low for alim/rlim at this size; some quick testing suggests the optimum is likely between 500M and 1000M. Is this done to keep memory use low? How many 16f contributors don't have the 1.5GB per thread needed to use lim=500M?

2021-09-24, 13:40   #59
pinhodecarlos

"Carlos Pinho"
Oct 2011
Milton Keynes, UK

497110 Posts

Quote:
 Originally Posted by charybdis Also 250M is very low for alim/rlim at this size; some quick testing suggests the optimum is likely between 500M and 1000M. Is this done to keep memory use low? How many 16f contributors don't have the 1.5GB per thread needed to use lim=500M?
95%.

2021-09-24, 15:13   #60
frmky

Jul 2003
So Cal

22·547 Posts

Quote:
 Originally Posted by charybdis Also 250M is very low for alim/rlim at this size; some quick testing suggests the optimum is likely between 500M and 1000M. Is this done to keep memory use low? How many 16f contributors don't have the 1.5GB per thread needed to use lim=500M?
A large fraction encounter issues when exceeding 1GB/thread, so I stay a little below that.

 2021-09-24, 15:50 #61 charybdis     Apr 2020 17·29 Posts If lims have to stay at 250M, it would probably be possible to stretch the upper limit of doable jobs a bit by using 3LP on both sides to catch some of the relations that are lost due to the low lims. This makes sec/rel ~30% worse but increases yield by ~50%, while also increasing the number of relations needed by some unknown amount (almost certainly below 50%) and making LA that bit harder as a result. But as long as you can cope with lpb 34/34 and 3LP on only one side, there shouldn't be any need for this.

 Similar Threads Thread Thread Starter Forum Replies Last Post frmky Msieve 3 2016-11-06 11:45 burrobert Msieve 9 2012-10-26 22:46 em99010pepe Msieve 23 2009-09-27 16:13 masser Sierpinski/Riesel Base 5 83 2007-11-17 19:39

All times are UTC. The time now is 04:34.

Mon Oct 18 04:34:03 UTC 2021 up 86 days, 23:03, 0 users, load averages: 1.25, 1.01, 1.10