mersenneforum.org  

Go Back   mersenneforum.org > Factoring Projects > Cunningham Tables

Reply
 
Thread Tools
Old 2021-11-24, 03:24   #309
frmky
 
frmky's Avatar
 
Jul 2003
So Cal

32×13×19 Posts
Default

Quote:
Originally Posted by henryzz View Post
although it would need to be a high memory GPU.
It might squeeze onto a 48GB card, but more likely two 32GB+ cards would be needed.
frmky is online now   Reply With Quote
Old 2021-11-24, 05:01   #310
VBCurtis
 
VBCurtis's Avatar
 
"Curtis"
Feb 2005
Riverside, CA

117138 Posts
Default

I think your CPU will do the job in under a month. We should have factors just before the new year.
Looks like we reach our relations target Wednesday- I suggest Ed point his farm elsewhere tomorrow night!
I'll be shutting down a client or two Wednesday to run other tasks, so that when the CADO server is taken down my machine is still crunching on something.
VBCurtis is online now   Reply With Quote
Old 2021-11-24, 14:00   #311
EdH
 
EdH's Avatar
 
"Ed Hall"
Dec 2009
Adirondack Mtns

3·372 Posts
Default

I currently have a crontab set to shut down my clients starting an hour before Oliver's announced server shutdown. I won't be able to change that until at least late tonight (local). If Oliver still plans to keep his stated time, I'll run until then, but if he changes that, I can redirect late tonight (local).
EdH is offline   Reply With Quote
Old 2021-11-24, 14:11   #312
kruoli
 
kruoli's Avatar
 
"Oliver"
Sep 2017
Porta Westfalica, DE

13768 Posts
Default

Quote:
Originally Posted by kruoli View Post
  • in 36 h
This is unfortunately still running, it takes more time than expected. For some reason the last run was much quicker with near-identical load in the background, even much quicker than the runs before it. Interesting.

Quote:
Originally Posted by kruoli View Post
  • maybe once on Wednesday (not sure when)
I will skip that, especially if some drop out now. It simply takes too long without having much more information for us.

Quote:
Originally Posted by VBCurtis View Post
I think your CPU will do the job in under a month. We should have factors just before the new year.
Yes, I am not worried about the time it takes. After we are finished sieving, I need to move the machine inside the house and then will do the benchmarking (MPI or not), so I will give a final update when I have started LA with final settings, which should be this weekend.
Quote:
Originally Posted by VBCurtis View Post
Looks like we reach our relations target Wednesday- I suggest Ed point his farm elsewhere tomorrow night!
I'll be shutting down a client or two Wednesday to run other tasks, so that when the CADO server is taken down my machine is still crunching on something.
(also @EdH) If all clients are pulled early, I will stop the server early. Otherwise, it will run at least to the stated time. Then, we would be getting close to 1.5B relations. In that case, the relation target set in the parameter file will be reached.

Last fiddled with by kruoli on 2021-11-24 at 14:24 Reason: Spelling. Grammar.
kruoli is online now   Reply With Quote
Old 2021-11-25, 01:51   #313
EdH
 
EdH's Avatar
 
"Ed Hall"
Dec 2009
Adirondack Mtns

3×372 Posts
Default

I was away all day today, but I see all my clients are still. I must have set my crontab for Mittwok instead of Donnerstag. I will move them to something else and let you work the server with whomever is left. (I haven't checked every client, but the few I looked at were stopped.)

Thanks for heading this up and good luck getting all the LA details sorted out.
EdH is offline   Reply With Quote
Old 2021-11-25, 10:13   #314
kruoli
 
kruoli's Avatar
 
"Oliver"
Sep 2017
Porta Westfalica, DE

2·383 Posts
Default

Quote:
Originally Posted by kruoli View Post
  • in 36 h
Sorry for the late result, it ran for more than a day. TD 130 and 1.42B relations gave:
Code:
found 388715044 duplicates and 1029756638 unique relations
…
Linear algebra completed 254 of 61004058 dimensions (0.0%, ETA 12275h28m)
kruoli is online now   Reply With Quote
Old 2021-11-25, 18:01   #315
kruoli
 
kruoli's Avatar
 
"Oliver"
Sep 2017
Porta Westfalica, DE

2×383 Posts
Default

Sieving is closed! Thanks to everyone who took part in this!
kruoli is online now   Reply With Quote
Old 2021-11-25, 20:08   #316
charybdis
 
charybdis's Avatar
 
Apr 2020

5·109 Posts
Default

Quote:
Originally Posted by kruoli View Post
Sieving is closed! Thanks to everyone who took part in this!
Thank you for organising! Hope the linear algebra goes smoothly.
charybdis is offline   Reply With Quote
Old 2021-11-28, 17:09   #317
kruoli
 
kruoli's Avatar
 
"Oliver"
Sep 2017
Porta Westfalica, DE

2×383 Posts
Default

Testing has finished:
Code:
found 409429916 duplicates and 1068575440 unique relations

 ~ MSIEVE compiled with AOCC, VBITS=128 ~
mpirun -np 2 --bind-to none ./msieve_mpi -nc2 1,2 -t 16 -v
~1200h (exact line was overriden by Ctrl-C by accident)

mpirun -np 2 --bind-to none ./msieve_mpi -nc2 "2,1 skip_matbuild=1" -t 16 -v
linear algebra completed 7124 of 58179007 dimensions (0.0%, ETA 1077h24m)

mpirun -np 4 --bind-to none ./msieve_mpi -nc2 "2,2 skip_matbuild=1" -t 8 -v
linear algebra completed 2291 of 58179007 dimensions (0.0%, ETA 1290h50m)

./msieve_nompi -nc2 skip_matbuild=1 -t 32 -v
linear algebra completed 1780 of 58179007 dimensions (0.0%, ETA 907h53m)

./msieve_nompi -nc2 skip_matbuild=1 -t 16 -v
linear algebra completed 2801 of 58179007 dimensions (0.0%, ETA 865h24m)

mpirun -np 2 --map-by l3cache ./msieve_mpi -nc2 "2,1 skip_matbuild=1" -t 16 -v
linear algebra completed 3804 of 58179007 dimensions (0.0%, ETA 913h20m)

mpirun -np 2 --map-by l3cache ./msieve_mpi -nc2 "2,1 skip_matbuild=1" -t 8 -v
linear algebra completed 1779 of 58179007 dimensions (0.0%, ETA 872h 3m)

 ~ MSIEVE compiled with GCC, VBITS=128 ~
./msieve_nompi -nc2 skip_matbuild=1 -t 16 -v
linear algebra completed 1272 of 58179007 dimensions (0.0%, ETA 863h55m)

 ~ MSIEVE compiled with AOCC and custom GMP build, VBITS=128 ~
./msieve_nompi -nc2 skip_matbuild=1 -t 16 -v
linear algebra completed 3056 of 58179007 dimensions (0.0%, ETA 861h56m)

(running again after running GMP's tuneup and recompiling)
./msieve_nompi -nc2 skip_matbuild=1 -t 16 -v
linear algebra completed 2292 of 58179007 dimensions (0.0%, ETA 874h17m)

 ~ MSIEVE compiled with AOCC, VBITS=256 ~
./msieve_nompi -nc2 skip_matbuild=1 -t 16 -v
linear algebra completed 2552 of 58179007 dimensions (0.0%, ETA 835h52m)
But there are also bad news... When I did the test filtering, my msieve invocation was always wrong. I had the TD parameter after -v and thus was ignored. I am in the process of rerunning this with the correct parameters.
kruoli is online now   Reply With Quote
Old 2021-11-29, 10:23   #318
kruoli
 
kruoli's Avatar
 
"Oliver"
Sep 2017
Porta Westfalica, DE

2·383 Posts
Default

LA has started!
The "correct" filtering run yielded a smaller matrix (51.45M dim) and an ETA of 748h. See you in about a month! This will become a tight squeeze to complete it this year…
kruoli is online now   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Using 16e on smaller numbers fivemack Factoring 3 2017-09-19 08:52
NFS on smaller numbers? skan YAFU 6 2013-02-26 13:57
Bernoulli(200) c204 akruppa Factoring 114 2012-08-20 14:01
checking smaller number fortega Data 2 2005-06-16 22:48
Factoring Smaller Numbers marc Factoring 6 2004-10-09 14:17

All times are UTC. The time now is 07:32.


Tue Nov 30 07:32:21 UTC 2021 up 130 days, 2:01, 0 users, load averages: 1.01, 0.96, 0.95

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

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.
A copy of the license is included in the FAQ.