Fast Breeding (guru management) - Page 51

fivemack · 2016-05-21, 15:30

OK, I have updated the list. I see that yoyo@home has finished at least one of the C188 GNFS targets (C188_132_125); I'm still a couple of weeks away from finishing my ECM task, so my GPU is occupied.

C286_139_122 is ready to be queued.

http://www.rechenkraft.net/yoyo/down...m/xy/wu_status is a bit confusing because it's inconsistent with the XYYXF tab at http://www.rechenkraft.net/yoyo/y_status_ecm.php which says C188_140_113 has had only 32095/42000 curves: does 'done' mean different things in the two files?

yoyo · 2016-05-21, 18:56

Quote:

Originally Posted by fivemack

http://www.rechenkraft.net/yoyo/down...m/xy/wu_status is a bit confusing because it's inconsistent with the XYYXF tab at http://www.rechenkraft.net/yoyo/y_status_ecm.php which says C188_140_113 has had only 32095/42000 curves: does 'done' mean different things in the two files?

Yes it's meaning is slightly different.
The first DONE means that those amount of curves are now converted into BOINC workunits.
The second url show the status of all these Boinc workunits, which in your case above means that 32096 curves or 42000 were already done. Now 32935 are done.

yoyo

XYYXF · 2016-05-21, 19:48

Quote:

Originally Posted by fivemack

C286_139_122 is ready to be queued.

It is already sieving (16e).

wblipp · 2016-05-21, 21:39

Here's a batch that has completed at yoyo and is ready NFS

241^125-1 (Quartic - perhaps a big 14e or a 15e)
5659^67-1
6007^67-1
7282882732751770301917311394675342168569930282640579^5-1 (quartic)
37141302977105581064921322418772202338727337015016057^5-1 (quartic)
230605827695659261374183298936900535402657822464812511^5-1 (quartic)
1398036042749871729843146160774339503205345274956506531^5-1 (quartic)

Additional numbers are on their final yoyo pass and should be ready in a week or two.

pinhodecarlos · 2016-05-23, 19:11

Just been contacted by Silverman and he is saying all of the composites at http://www.chiark.greenend.org.uk/uc...mack/homcun.pl have
hit the "sweet spot" for lasieved: they are SNFS in the 235-250 digit range.
All have had at least t50. jyb has added more curves and Silverman has just finished a pass of 3000 curves with B1 = 200 x 10^6.

Lionel, can you take a look on those composites please.

jyb · 2016-05-23, 23:26

Quote:

Originally Posted by pinhodecarlos

Just been contacted by Silverman and he is saying all of the composites at http://www.chiark.greenend.org.uk/uc...mack/homcun.pl have
hit the "sweet spot" for lasieved: they are SNFS in the 235-250 digit range.
All have had at least t50. jyb has added more curves and Silverman has just finished a pass of 3000 curves with B1 = 200 x 10^6.

Lionel, can you take a look on those composites please.

Ah, this is interesting. In the past I had calculated that these numbers had all received the equivalent of a t50, plus approximately 5000 curves at B1 = 110e6. 3000 additional curves at B1 = 200e6 would be the equivalent of a little more than 5500 curves at 110e6, bringing the total up over 10K curves. And that's the threshold that Lionel has been asking for these numbers.

So yes, it does appear that they're all ready for NFS.

And just another data point: I have run another 4000 curves at B1 = 110e6 on the following six of these numbers:

Code:

7-5,289
7-4,289
9+5,256
9+7,256
8-3,271
8-7,271

so they may qualify as the best candidates with which to start.

Lionel, would you like me to generate the .poly files for them?

debrouxl · 2016-05-24, 06:25

Jon: yup, it would be great if you could generate polys for the 4 last of these 6 numbers. I used phi 0.1.4 to generate polynomials for the two first ones and queued them, out of order, for diversity. I'll get back to numbers from posts #533 and #534, before XYYXF numbers from posts #513/#528 (14e) and #531, #532 (15e).

I've just asked Greg for a copy of the current state of the 14e and 15e queue management code, which has evolved since I had him import it from RSALS. I'd like to work on it on my side, by e.g. removing redundant fields which can be deduced from the poly; I'll create a topic about infrastructure automation (posts #526 onwards) later.

pinhodecarlos · 2016-05-24, 14:18

Soon I'll request a call of arms to NFS@Home members to help us reach the 1000 factorisation on 14e side.

debrouxl · 2016-05-24, 14:20

Not until we have a more automated infrastructure, which will take a while.

pinhodecarlos · 2016-05-24, 14:27

Quote:

Originally Posted by debrouxl

Not until we have a more automated infrastructure, which will take a while.

How long, three weeks? I'll be on my annual leave on the first two weeks of June so I was thinking on doing that afterwards.

debrouxl · 2016-05-24, 15:07

Probably far more than this, let's be realistic...

2016-05-23, 19:11	#555
pinhodecarlos "Carlos Pinho" Oct 2011 Milton Keynes, UK 1353₁₆ Posts	Just been contacted by Silverman and he is saying all of the composites at http://www.chiark.greenend.org.uk/uc...mack/homcun.pl have hit the "sweet spot" for lasieved: they are SNFS in the 235-250 digit range. All have had at least t50. jyb has added more curves and Silverman has just finished a pass of 3000 curves with B1 = 200 x 10^6. Lionel, can you take a look on those composites please. Last fiddled with by pinhodecarlos on 2016-05-23 at 19:15

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
System management notes	kriesel	kriesel	7	2020-10-21 18:52
Improving the queue management.	debrouxl	NFS@Home	10	2018-05-06 21:05
Script-based Primenet assignment management	ewmayer	Software	3	2017-05-25 04:02
Do normal adults give themselves an allowance? (...to fast or not to fast - there is no question!)	jasong	jasong	35	2016-12-11 00:57
Power Management settings	PrimeCroat	Hardware	3	2004-02-17 19:11

2016-05-21, 15:30	#551
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 14430₈ Posts	OK, I have updated the list. I see that yoyo@home has finished at least one of the C188 GNFS targets (C188_132_125); I'm still a couple of weeks away from finishing my ECM task, so my GPU is occupied. C286_139_122 is ready to be queued. http://www.rechenkraft.net/yoyo/down...m/xy/wu_status is a bit confusing because it's inconsistent with the XYYXF tab at http://www.rechenkraft.net/yoyo/y_status_ecm.php which says C188_140_113 has had only 32095/42000 curves: does 'done' mean different things in the two files?

2016-05-21, 21:39	#554
wblipp "William" May 2003 New Haven 2×7×13² Posts	Here's a batch that has completed at yoyo and is ready NFS 241^125-1 (Quartic - perhaps a big 14e or a 15e) 5659^67-1 6007^67-1 7282882732751770301917311394675342168569930282640579^5-1 (quartic) 37141302977105581064921322418772202338727337015016057^5-1 (quartic) 230605827695659261374183298936900535402657822464812511^5-1 (quartic) 1398036042749871729843146160774339503205345274956506531^5-1 (quartic) Additional numbers are on their final yoyo pass and should be ready in a week or two.

2016-05-24, 06:25	#557
debrouxl Sep 2009 977 Posts	Jon: yup, it would be great if you could generate polys for the 4 last of these 6 numbers. I used phi 0.1.4 to generate polynomials for the two first ones and queued them, out of order, for diversity. I'll get back to numbers from posts #533 and #534, before XYYXF numbers from posts #513/#528 (14e) and #531, #532 (15e). I've just asked Greg for a copy of the current state of the 14e and 15e queue management code, which has evolved since I had him import it from RSALS. I'd like to work on it on my side, by e.g. removing redundant fields which can be deduced from the poly; I'll create a topic about infrastructure automation (posts #526 onwards) later.

2016-05-24, 14:18	#558
pinhodecarlos "Carlos Pinho" Oct 2011 Milton Keynes, UK 3·17·97 Posts	Soon I'll request a call of arms to NFS@Home members to help us reach the 1000 factorisation on 14e side.

2016-05-24, 14:20	#559
debrouxl Sep 2009 977₁₀ Posts	Not until we have a more automated infrastructure, which will take a while.

2016-05-24, 15:07	#561
debrouxl Sep 2009 977 Posts	Probably far more than this, let's be realistic...