Linear algebra reservations, progress and results - Page 157

pinhodecarlos · 2016-12-25, 22:36

Quote:

Originally Posted by VBCurtis

Thanks Rich! This worked perfectly. Not sure why the built-in-to-GUI archive manager failed; perhaps file size too big. Filtering now.

Is LA running Curtis?

VBCurtis · 2016-12-26, 06:26

Yes, it's complete. Took 22 hours for 6.8M matrix at TD=130. Unfortunately, I killed X by installing a second GPU, and haven't gotten X back up yet, so the factors are stuck on the machine temporarily. I should have it fixed Monday or Tuesday. I ran msieve in the console.

I ran about 1.3t50 of ECM, and I'm sad to say it had a p52 factor.

pinhodecarlos · 2016-12-26, 08:48

That's a pity Curtis.

In the meantime C208_127_84 LA is underway, just hope the factors come in before I leave to Cambdrige, on Wednesday morning, otherwise I'll post the factors this Friday afternoon.

Dubslow · 2016-12-27, 00:58

Quote:

Originally Posted by VBCurtis

Yes, it's complete. Took 22 hours for 6.8M matrix at TD=130. Unfortunately, I killed X by installing a second GPU, and haven't gotten X back up yet, so the factors are stuck on the machine temporarily. I should have it fixed Monday or Tuesday. I ran msieve in the console.

I ran about 1.3t50 of ECM, and I'm sad to say it had a p52 factor.

Can't get to a console and tail msieve.log?

Do you mean this p52 is for the one you did linalg for..?

VBCurtis · 2016-12-27, 07:11

Yes, the p52 is from the LA for 13*2^784-1. Yes, I can tail, or any other console command, but I've no way to get the log (or the screen output) to another machine until I get X back up. I don't have it set up to accept ssh connections, and would rather spend time fixing X than figuring out how to transfer a file from a console shell.

RichD · 2016-12-27, 20:41

162 hours to solve a 15.66M matrix, -t 4, TD=120. (TD=124 failed)

Code:

p107 factor: 66306695557313140049874237812099824092143104068796597405489471225486552252501661749058962921303529911190459
p141 factor: 300338650937306889791783385695010556994174491037184281645334453473675578174559072167694740059558885183445705871042812524682267057654085081617

pinhodecarlos · 2016-12-28, 07:31

Code:

Wed Dec 28 07:26:08 2016  p57 factor: 152399634435317152624511279321449057813195634735732566019
Wed Dec 28 07:26:08 2016  p152 factor: 38708956448686594595959450337996817935935767438934801816439101708794973824970666169191293576533368651161201966628877043795085850377793755421196444816533

http://pastebin.com/NnuwJKaM

VBCurtis · 2016-12-29, 02:11

Quote:

Originally Posted by VBCurtis

Yes, it's complete. Took 22 hours for 6.8M matrix at TD=130. Unfortunately, I killed X by installing a second GPU, and haven't gotten X back up yet, so the factors are stuck on the machine temporarily. I should have it fixed Monday or Tuesday. I ran msieve in the console.

I ran about 1.3t50 of ECM, and I'm sad to say it had a p52 factor.

My records indicate 1.6t50 rather than 1.3, which makes the missed p52 more disappointing...

Code:

prp52 factor: 3145095025048136688399558951107897016606343957919549
prp173 factor: 36214140129622755877783809369020064295424202666405688407097733248217469988148883021561162411968716554600845342451810800220434283522628163961120199187500619266122659758926319

Log attached.

YuL · 2016-12-29, 19:01

post-processing of C186_145_72, can I have it please?

jyb · 2016-12-30, 07:57

Quote:

Originally Posted by YuL

post-processing of C186_145_72, can I have it please?

Yes. Note that this number is a little light on sieving. Let us know if you have trouble building a matrix, or if it seems like it'll take a ridiculously long time to do the LA, and we can schedule some more sieving.

YuL · 2016-12-30, 13:52

Quote:

Originally Posted by jyb

Yes. Note that this number is a little light on sieving. Let us know if you have trouble building a matrix, or if it seems like it'll take a ridiculously long time to do the LA, and we can schedule some more sieving.

Thanks. It goes like this:
- using 428,484,349 relations filtering fails to produce a matrix at td=128
- using 429,259,244 relations filtering produces a matrix at td=124 but, using 8 threads
on Intel Xeon E5-2670 v2, I get

Code:

Fri Dec 30 10:34:41 2016  matrix is 20341268 x 20341445 (9856.9 MB) with weight 2823469472 (138.80/col)
....
Fri Dec 30 10:49:26 2016  linear algebra at 0.0%, ETA 384h22m

A 20.3e6 matrix is something I can deal with in about 9 days using my dual
Xeon E5-2620 but the machine is not available right now as it is processing
another number.
So I accept your offer, more sieving would be fine, however I don't know how much
maybe we could target 460M relations?

2016-12-29, 19:01	#1725
YuL Feb 2012 Paris, France 241₈ Posts	145^72+72^145 cofactor post-processing of C186_145_72, can I have it please?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
restarting nfs linear algebra	cubaq	YAFU	2	2017-04-02 11:35
Linear algebra at 600%	CRGreathouse	Msieve	8	2009-08-05 07:25
Linear algebra crashes	10metreh	Msieve	3	2009-02-02 08:34
Linear algebra proof	Damian	Math	8	2007-02-12 22:25
Linear algebra in MPQS	R1zZ1	Factoring	2	2007-02-02 06:45

2016-12-26, 06:26	#1718
VBCurtis "Curtis" Feb 2005 Riverside, CA 4874₁₀ Posts	Yes, it's complete. Took 22 hours for 6.8M matrix at TD=130. Unfortunately, I killed X by installing a second GPU, and haven't gotten X back up yet, so the factors are stuck on the machine temporarily. I should have it fixed Monday or Tuesday. I ran msieve in the console. I ran about 1.3t50 of ECM, and I'm sad to say it had a p52 factor.

2016-12-26, 08:48	#1719
pinhodecarlos "Carlos Pinho" Oct 2011 Milton Keynes, UK 3·17·97 Posts	That's a pity Curtis. In the meantime C208_127_84 LA is underway, just hope the factors come in before I leave to Cambdrige, on Wednesday morning, otherwise I'll post the factors this Friday afternoon.

2016-12-27, 07:11	#1721
VBCurtis "Curtis" Feb 2005 Riverside, CA 2·2,437 Posts	Yes, the p52 is from the LA for 13*2^784-1. Yes, I can tail, or any other console command, but I've no way to get the log (or the screen output) to another machine until I get X back up. I don't have it set up to accept ssh connections, and would rather spend time fixing X than figuring out how to transfer a file from a console shell.