2018 14e post processing reservations and results - Page 3

fivemack · 2018-05-24, 14:30

Code:

Thu May 24 14:41:46 2018  p81 factor: 659111171162587511001945202207366819602644411964225065081001137913325383295838909
Thu May 24 14:41:46 2018  p148 factor: 1672581374250702824738947742798076127330974868661411511817845162122621949046858328275460764421317767711434935811885774187181548411457277922503957077

72.8 hours for 12.00M density-130 (140 didn't work) matrix on 7 threads E5-2650v2. Log (containing density-100 and density-120 filtering - I ran 120/130/140 in parallel and didn't keep all three logs) attached and at https://pastebin.com/MT99LRXT

I just took the first 400M relations rather than trying more complicated pre-filtering analyses.

fivemack · 2018-05-24, 14:33

Quote:

Originally Posted by swellman

Thank you! For the record, TD=90 built a matrix with an ETA of 1265 hrs.

OK, those runtimes are enormous, I have added a further 100MQ of sieving.

I hope there is some degree of 'running on this potato' in the runtimes ... 1300 hours would deal with a 50M matrix on my larger machines. If it's still >1000 hours for the job then I would happily take it over and run it on something chunkier.

swellman · 2018-05-24, 16:19

Quote:

Originally Posted by fivemack

OK, those runtimes are enormous, I have added a further 100MQ of sieving.

I hope there is some degree of 'running on this potato' in the runtimes ... 1300 hours would deal with a 50M matrix on my larger machines. If it's still >1000 hours for the job then I would happily take it over and run it on something chunkier.

32-bit jobs can sometimes take on a life of their own. My first two jobs on the current 14e queue both had initial ETAs over 1000 hours on my personal hardware. The first job is a bit undersieved, and the second is just plain hard (SNFS 270). I will re-download the C209_701xx141_5 job tonight when it has 480M+ relations and get an ETA, or several at ever increasing TD.

Jarod · 2018-05-25, 08:34

Reserving C198_756xx781_11 Thanks

swellman · 2018-05-25, 12:05

Quote:

Originally Posted by swellman

I will re-download the C209_701xx141_5 job tonight when it has 480M+ relations and get an ETA, or several at ever increasing TD.

Many thanks for generating more relations.

With ~484M relations, TD=130 shows an ETA of 937 hrs. Not sure this time can be improved significantly. Reading in more relations and/or higher TD would likely help a bit but when traded against the time to optimize conditions, all under the cloud of possible oversieving lead me to think going with what I got might be best. But I’m open to investigating all scenarios if it’s helpful to others.

swellman · 2018-05-25, 12:45

Quote:

Originally Posted by pinhodecarlos

Please hold C212_157xx799_13. Making a test.

Quote:

Test done, don't need to reserve this composite.

I double checked with Carlos - he confirms he is not working C212_157xx799_13. Can a gatekeeper edit the 14e queue to show this composite as unreserved?

debrouxl · 2018-05-25, 13:30

I have removed the reservation for C212_157xx799_13.

swellman · 2018-05-25, 14:06

Thank you.

richs · 2018-05-26, 15:14

Quote:

Originally Posted by richs

Reserving C191_295xx029_13

Code:

p80 factor: 13745796595569657487046976086814644298188282962356651514984076906025347008147003
p111 factor: 930761774694254454289678294527788495654974849663089855154505082398340025888249556096804633684441558318426871251

Approximately 29.6 hours on 2 threads Core i3-2310M with 4 GB memory for a 4.36M matrix at TD = 70 (TD = 130 didn't work).
Log attached and log at https://pastebin.com/1QsvP734

Factors reported to factor database.

Jarod · 2018-05-29, 04:42

Quote:

Originally Posted by Jarod

Reserving C198_756xx781_11 Thanks

I am going to have a guess and say this job took 4 1/2 hours. Starting the job with 8 cores and finished with 15 just to see if there was much time difference and there wasn't

Code:

p70 factor: 1989774706154817243345988310685771699308514675731245500652616051195507
p129 factor: 187207251745133606157391653086769259037917392116430983143576943217585815557363013773039151703724781592509357752386497852063522527

fivemack · 2018-05-29, 09:10

As a person making essential use of Pastebin, and because they have a cheap offer at the moment, I've bought a lifetime PRO account for sixteen pounds. Apologies if I can't figure out how to drive it and accidentally produce, for example, pastes that aren't world-readable.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
2018 15e post-processing reservations and results	fivemack	NFS@Home	221	2019-01-04 13:08
16e Post Processing Progress	pinhodecarlos	NFS@Home	8	2018-11-28 13:45
Crash doing large post-processing job	wombatman	Msieve	22	2013-12-04 01:37
Update on 7^254+1 post processing	dleclair	NFSNET Discussion	4	2005-04-05 09:51
Post processing for 2,757-	xilman	NFSNET Discussion	3	2003-11-06 14:23

2018-05-25, 08:34	#26
Jarod AKA Speedy51 Oct 2012 New Zealand 3×73 Posts	Reserving C198_756xx781_11 Thanks

2018-05-25, 13:30	#29
debrouxl Sep 2009 3D1₁₆ Posts	I have removed the reservation for C212_157xx799_13.

2018-05-25, 14:06	#30
swellman Jun 2012 5²×7×17 Posts	Thank you.

2018-05-29, 09:10	#33
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 1100011101111₂ Posts	As a person making essential use of Pastebin, and because they have a cheap offer at the moment, I've bought a lifetime PRO account for sixteen pounds. Apologies if I can't figure out how to drive it and accidentally produce, for example, pastes that aren't world-readable.