mersenneforum.org Shall we adopt another sequence?
 Register FAQ Search Today's Posts Mark Forums Read

 2011-10-27, 06:23 #1 schickel     "Frank <^>" Dec 2004 CDP Janesville 2×1,061 Posts Shall we adopt another sequence? I've been considering adding another sequence to the community pool, and in light of the recent activity on 4788, maybe now is the time to do it. 283752 is the shortest of the sequences currently greater than 7000 lines. It is also in a potentially fortuitous position. It is driven by 2^3 * 3, but the current line has the form 2^3 * 3^2 * c156. Accordingly, there is a 50% chance that on the next line, the driver can change. unconnected reports that he has completed 1.5*t55 worth of ECM with no success so far. Unfortunately, I am currently working on a GNFS job for a c154 of my own, so I'm going to be tied up for ~10 days or so....what do you all think?
 2011-10-29, 10:20 #2 henryzz Just call me Henry     "David" Sep 2007 Cambridge (GMT/BST) 2×2,861 Posts It might be nice to extend this sequence as a team to see if it can lose the driver with the current 2^3*3^2. Running that size sequence with a driver seems like a bit too much effort though.
 2011-10-29, 12:39 #3 LaurV Romulan Interpreter     Jun 2011 Thailand 2×11×397 Posts Any poly available for C156? (crunching for one right now, but the process is slow, one core only).
 2011-11-04, 18:26 #4 LaurV Romulan Interpreter     Jun 2011 Thailand 2×11×397 Posts Got a poly after about a week. Is it (relatively) good? How good? How long the sieving will take on two cores of a core2 3Gigs? (about) Can that be taken as a "one man job?" Maybe I will give it a try.... Code: n: 150334450606011724019777200211010468220565590046299234402254345532711750018652367487259651931850319063498312781804011647293058067263942651704486104870980321 skew: 14230655.24 c0: -520134197785689262516595081751266984241 c1: 809982347485060642501973343231813 c2: 321021768654971361269241683 c3: -4637554577292436357 c4: -1395390947346 c5: 17040 Y0: -1545689251703330366629739901868 Y1: 956597419325010283 rlim: 30000000 alim: 30000000 lpbr: 29 lpba: 29 mfbr: 58 mfba: 58 rlambda: 2.500000 alambda: 2.500000
2011-11-04, 19:21   #5
jrk

May 2008

3×5×73 Posts

Quote:
 Originally Posted by LaurV Got a poly after about a week. Is it (relatively) good? How good? How long the sieving will take on two cores of a core2 3Gigs? (about) Can that be taken as a "one man job?" Maybe I will give it a try....
Yes that is a good poly. My expected murphy score for a number of this size is about 2.71e-12 and your poly is 2.81e-12, so that is good. You should use ggnfs siever 14e and sieve on the -a side starting from specialq 15M until you have gathered 56M raw relations. Sieving will take about 3 to 4 weeks on your two cores, and the matrix about 2 days using msieve.

Last fiddled with by jrk on 2011-11-04 at 19:22

2011-11-04, 20:11   #6
Andi47

Oct 2004
Austria

2×17×73 Posts

Quote:
 Originally Posted by jrk Yes that is a good poly. My expected murphy score for a number of this size is about 2.71e-12 and your poly is 2.81e-12, so that is good. You should use ggnfs siever 14e and sieve on the -a side starting from specialq 15M until you have gathered 56M raw relations. Sieving will take about 3 to 4 weeks on your two cores, and the matrix about 2 days using msieve.
I am not sure whether LaurV can finish this one as an one-man job.

One of my biggest "one man" factorizations so far was a c153 cofactor of 2,1766M, and I used a similar computer for it (core 2, 2 gigs RAM).

After finishing sieving I attempted to do the matrix, and what happened then, can be seen in that thread beginning from post #157: Matrix generation failed with a "failed to reallocate 1.63 GB" error. On a 32 bit windows system it was not possible to do this linalg job, not even when I moved to a P4 with 3 Gigs RAM (also with 32-bit windows), because 32-bit windows is not able to handle the address space (reallocating huge chunks of data) which is needed for this matrix. When I switched to linux, it happily worked out.

So if LaurV's computer is running on a 64 bit system and with a 64-bit binary of msieve, 156 digit GNFS should be possible (but close to the edge of available RAM) (will take approx. 2 GB RAM).
For 32-bit systems it could end up in a balancing act between extreme oversiving for squeezing the matrix in smaller amount of RAM an failing when it's way too oversieved and runs out of cliques in the filtering step.

Last fiddled with by Andi47 on 2011-11-04 at 20:34 Reason: typo, clarification, ...

2011-11-04, 20:20   #7
schickel

"Frank <^>"
Dec 2004
CDP Janesville

2×1,061 Posts

Quote:
 Originally Posted by LaurV Got a poly after about a week. Is it (relatively) good? How good? How long the sieving will take on two cores of a core2 3Gigs? (about) Can that be taken as a "one man job?" Maybe I will give it a try.... [ snip poly ]
Thanks for the poly search LaurV; I was going to ask jrk to do a search when I got done with my current factorization.
Quote:
 Originally Posted by jrk Yes that is a good poly. My expected murphy score for a number of this size is about 2.71e-12 and your poly is 2.81e-12, so that is good. You should use ggnfs siever 14e and sieve on the -a side starting from specialq 15M until you have gathered 56M raw relations. Sieving will take about 3 to 4 weeks on your two cores, and the matrix about 2 days using msieve.
Thanks for the quality check, jrk.

@LaurV: Do you want to tackle the whole thing on your own? If yes, there's another community-sized project that I would like to look into. (660 is currently at i890 c181 = 2^3 * 3^2 * 5 * c178. No chance of escape on this line, but since it is the biggest sequence, it would definitely be a huge deal to see if the 3^2 holds for the next line.....)

If you tackle 283752, I'll email Paul Zimmerman to see if he's got anything going on with 660....

2011-11-04, 20:23   #8
schickel

"Frank <^>"
Dec 2004
CDP Janesville

2×1,061 Posts

Quote:
 Originally Posted by Andi47 I am not sure whether LaurV can finish this one as an one-man job. One of my biggest "one man" factorization so far was a c153 cofactor of 2,1766M, and I used a similar computer for it (core 2, 2 gigs RAM). After finishing sieving I attempted to do the matrix, and what happened then, can be seen in that thread beginning from post #157: Matrix generation failed with a "failed to reallocate 1.63 GB" error. On a 32 bit windows system it was not possible to do this linalg job, not even when I moved to a P4 with 3 Gigs RAM (also with 32-bit windows), because 32-bit windows is not able to handle the adress space which is needed for this matrix. When I switched to linux, it happily worked out. So if LaurV's computer is running on a 64 bit system and with a 64-bit binary of msieve, 156 digit GNFS should be possible (but close to the edge of available RAM). For 32-bit systems it could end up in a balancing act between extreme oversiving for squeezing the matrix in smaller amount of RAM an failing when it's way too oversieved and runs out of cliques in the filtering step.

Well, if it comes to that, there are several people that could do the post-processing on the job.

2011-11-04, 20:48   #9
Andi47

Oct 2004
Austria

46628 Posts

Quote:
 Originally Posted by schickel Ugh....hadn't even considered that. Well, if it comes to that, there are several people that could do the post-processing on the job.
One should consider that this requires that LaurV uploads huge amounts of data (I guess this would be 1-2 GB?). But this should be no problem if he has a flat rate with sufficiently high (or unlimited) data transfer per month.

2011-11-05, 03:24   #10
LaurV
Romulan Interpreter

Jun 2011
Thailand

221E16 Posts

Quote:
 Originally Posted by schickel @LaurV: Do you want to tackle the whole thing on your own?
I don't necessarily "want" to take the job by myself. I am kinda busy with my C185 from fib(1049) since about two months, and want to finish it only because of pride :D (the one for which I did not know what I am getting into, and fivemack helped me by computing the poly, still sieving on rational side, about 40M relations already). I am not really ready to take a new "long term" challenge, that is why I asked about duration. I could help, however, with small tasks, like for few days or so. Some of my cores are busy with P95, but some do other jobs and are free from time to time, that is why I stepped into aliquots, because the tasks do not last long, and you can see immediate results. I am kinda keeping my old 996666 (132 digits) and the newer 585000 (114 digits) for now on.

Finding the poly was not different, just an "over the weekend plus a day or two more", when the computer at job stays anyhow idle.

But taking another "one month job" by myself would be too much. As I said, I could help. But I won't promise anything.

Last fiddled with by LaurV on 2011-11-05 at 03:28

2011-11-05, 04:25   #11
schickel

"Frank <^>"
Dec 2004
CDP Janesville

2·1,061 Posts

Quote:
 Originally Posted by LaurV I don't necessarily "want" to take the job by myself. I am kinda busy with my C185 from fib(1049) since about two months, and want to finish it only because of pride :D (the one for which I did not know what I am getting into, and fivemack helped me by computing the poly, still sieving on rational side, about 40M relations already). I am not really ready to take a new "long term" challenge, that is why I asked about duration. I could help, however, with small tasks, like for few days or so. Some of my cores are busy with P95, but some do other jobs and are free from time to time, that is why I stepped into aliquots, because the tasks do not last long, and you can see immediate results. I am kinda keeping my old 996666 (132 digits) and the newer 585000 (114 digits) for now on. Finding the poly was not different, just an "over the weekend plus a day or two more", when the computer at job stays anyhow idle. But taking another "one month job" by myself would be too much. As I said, I could help. But I won't promise anything.
OK, I threw it open as a team job, and if no one else offers (henryzz?) I can do the post-processing on this. If it takes long enough, I can probably run a 6-core job on my new system.....it's only just now starting to stay cool enough upstairs during the day to contemplate running more than 3 cores.

 Similar Threads Thread Thread Starter Forum Replies Last Post sweety439 And now for something completely different 17 2017-06-13 03:49 devarajkandadai Math 1 2007-08-25 15:23 devarajkandadai Math 3 2007-03-20 19:43 Citrix Puzzles 5 2005-09-14 23:33 biwema Puzzles 13 2004-06-11 02:05

All times are UTC. The time now is 13:28.

Sat Sep 19 13:28:04 UTC 2020 up 9 days, 10:39, 1 user, load averages: 1.05, 1.22, 1.30