Fast Breeding (guru management) - Page 108

RichD · 2017-10-03, 01:05

queued C192 from the OPN t600 file.

Code:

n: 414293302092344017554661357329277202485278098895744995614721477118078507284151687520161679071795586391997371495642459917192518894462751183367401397961778517836686348168898838352646064970952651
# 249541^41-1, difficulty: 226.68, skewness: 7.93, alpha: 0.00
# cost: 1.32785e+18, est. time: 632.31 GHz days (not accurate yet!)
skew: 7.935
c6: 1
c0: -249541
Y1: -1
Y0: 60255039831511724428874194299741430381
m: 60255039831511724428874194299741430381
type: snfs
rlim: 67000000
alim: 67000000
lpbr: 30
lpba: 30
mfbr: 60
mfba: 60
rlambda: 2.6
alambda: 2.6

Trial sieving 2K blocks.

Code:

RichD · 2017-10-03, 03:49

Correction to the above post.
Trial sieving 5K blocks.

swellman · 2017-10-03, 14:49

queued C222_141_62 is ready for SNFS as a 15e/31 job.

Code:

n: 972321105743081656979376997770755626021809788244648150121626742682893357818356078629936773839300186077348832447590209404560750435134166496197530649467144877594353187819630800027157225839553392840662199269876640383429889629
# 141^62+62^141, difficulty: 254.52, anorm: 2.22e+033, rnorm: 8.64e+055
# scaled difficulty: 258.28, suggest sieving rational side
# size = 8.099e-018, alpha = 0.000, combined = 4.746e-014, rroots = 1
type: snfs
size: 254
skew: 3.1711
c5: 62
c0: 19881
Y1: -61748917974902741368975281
Y0: 153803885110405674678434597293100547399764930461696
rlim: 134000000
alim: 134000000
lpbr: 31
lpba: 31
mfbr: 62
mfba: 62
rlambda: 2.7
alambda: 2.7

Test sieving on the -r side with Q in blocks of 2K.

Code:

Q=20M    rels=2116
Q=80M    rels=1958
Q=150M   rels=2078
Q=200M   rels=2119
Q=250M   rels=1904

Suggesting a sieving range for Q of 20M-260M with target # rels = 240M.

RichD · 2017-10-04, 15:30

QUEUED C161 from the OPN t550 file.

Code:

n: 18240051261562921058567070311861891715172678007177798917429354382357183300885632657444483979332189347244110527138705974473160994107001905334487636771142410048269
# 3613045528091^17-1 - GNFS-161
lss: 0
Y0: -13937843400604410542946244759611
Y1: 313390996726672401239
c0: -381675828186169124255846117946454368480
c1: 2371909260583519175557816411813204
c2: 39299843186916850157004264
c3: -58080752133083914273
c4: -328301327070
c5: 104040
skew: 11621349.00763
rlim: 67000000
alim: 67000000
lpbr: 30
lpba: 30
mfbr: 60
mfba: 60
rlambda: 2.6
alambda: 2.6
type: gnfs

Trial sieving 5K blocks.

Code:

  Q  Yield
 20M 10298
 50M 10442
 80M  8633

swellman · 2017-10-05, 10:46

Quote:

Originally Posted by fivemack

Thank you for all the options!

I have put this one in as a 14/33, I think 400M is rather low for 33LP; if you don't mind, could you try post-processing when it hits the Q=400M mark, and we can add more relations if that doesn't work.

Attempted to postprocess C202_137_75 @TD=70 with Q=400M and 542M+ raw relations. Didn't even make it through filtering, msieve demands more relations.

How about bumping Q up to 450M? done

VBCurtis · 2017-10-05, 17:10

Quote:

Originally Posted by swellman

Attempted to postprocess C202_137_75 @TD=70 with Q=400M and 542M+ raw relations. Didn't even make it through filtering, msieve demands more relations.

How about bumping Q up to 450M? done

I'm afraid that won't be enough; 400M was set as a checkpoint for us to estimate how much Q would be needed to reach 720-750M rels. I'm filtering the other 14e/33 job tonight/tomorrow, so I'll have data on how many rels will be needed for this. Q=450M won't be enough, as you're gonna need 700M rels at minimum; 750M is likely necessary to get a tractable matrix. My rule of thumb is that for same-difficulty jobs, increasing LP by 1 increases raw relations needed by 70%. 32 -> 33 is a region I haven't yet explored, but we would want 450M(?) 32LP relations for this job; this suggests we'll want 750-760M 33LP rels. We'll find out soon how accurate the 70% estimate is!

swellman · 2017-10-05, 17:12

FYI, C215_147_53 is ready for SNFS on 15e. Just pointing it out before it gets lost in the rear view.

I will avoid proposing anymore 15e jobs for a while. We seem to have enough. Thanks.

VBCurtis · 2017-10-06, 14:41

Our first 14e/33 job was filtered last night. 729M raw relations produced 592M unique via remdups4. Filtering with target_density=140 produced a matrix with density 83, so the number is quite oversieved.

So, C202_137_75 might be good to go with just 700-720M raw relations. I'm re-running filtering today with 560M unique relations to see if I get a smaller matrix.

VBCurtis · 2017-10-07, 18:30

580M unique relations produced 23.5M matrix with density 95.
560M unique relations produced 22.6M matrix with density 111.
540M unique relations produced 21.9M matrix with density 136.
Each density is as reported during filtering phase; I began LA on the first matrix, and the log reports sparse part has density 72. So, I'm running a 23.5M matrix of density 72, with ETA around 600 hours (I use the machine for other tasks, so ETA isn't very accurate anyway).

swellman · 2017-10-08, 16:15

Quote:

Originally Posted by swellman

Attempted to postprocess C202_137_75 @TD=70 with Q=400M and 542M+ raw relations. Didn't even make it through filtering, msieve demands more relations.

How about bumping Q up to 450M? done

Well, adding 50M additional Q is projected to add ~55M relations for a total of 596M raw relations. And from the recent findings of VBCurtis that won’t be near enough for a 14e/33 job. Suggest Q be bumped up to 550M. From there we can bump it up in smaller increments until the matrix builds successfully at TD=70.

Or just bump Q up to 500M if there’s a risk of overshooting the minimum number of relations. I’ll keep at it until we build a good matrix. But 596M raw relations isn’t going to do it I think.

swellman · 2017-10-09, 00:52

QUEUED C228_137_69 is ready for SNFS as a 14e/32 job. Uses a tip suggested by VBCurtis. Thanks!

Code:

n: 288586883064120353283752244127022318083321958689563012149474062888848914428054547248965781086918169929312487672089154962646543872755426023525435015307429819367951025596622808788595557023408258148610859799175375821689424729570031
# 137^69+69^137, difficulty: 253.76, anorm: 2.66e+040, rnorm: 4.04e+047
# scaled difficulty: 254.96, suggest sieving rational side
# size = 3.595e-013, alpha = 0.000, combined = 5.871e-014, rroots = 0
type: snfs
size: 253
skew: 23.7047
c6: 1
c0: 177423357
Y1: -319099584516184696444313
Y0: 1965753632901132991452851230906979378964909
rlim: 268000000
alim: 450000000
lpbr: 32
lpba: 32
mfbr: 64
mfba: 64
rlambda: 2.8
alambda: 2.8

Test sieving on the -r side with Q in blocks of 2K

Code:

Q=20M   rels=3491
Q=80M   rels=2462
Q=150M  rels=2249
Q=250M  rels=2227
Q=350M  rels=1687
Q=450M  rels=1667

Suggesting a sieving range for Q of 20M-470M with a target # rels=480M.

2017-10-06, 14:41	#1185
VBCurtis "Curtis" Feb 2005 Riverside, CA 1001100001010₂ Posts	C226_143_57 update Our first 14e/33 job was filtered last night. 729M raw relations produced 592M unique via remdups4. Filtering with target_density=140 produced a matrix with density 83, so the number is quite oversieved. So, C202_137_75 might be good to go with just 700-720M raw relations. I'm re-running filtering today with 560M unique relations to see if I get a smaller matrix.

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

2017-10-03, 03:49	#1179
RichD Sep 2008 Kansas 3,391 Posts	Correction to the above post. Trial sieving 5K blocks.

2017-10-05, 17:12	#1184
swellman Jun 2012 11·281 Posts	FYI, C215_147_53 is ready for SNFS on 15e. Just pointing it out before it gets lost in the rear view. I will avoid proposing anymore 15e jobs for a while. We seem to have enough. Thanks.

2017-10-07, 18:30	#1186
VBCurtis "Curtis" Feb 2005 Riverside, CA 1001100001010₂ Posts	580M unique relations produced 23.5M matrix with density 95. 560M unique relations produced 22.6M matrix with density 111. 540M unique relations produced 21.9M matrix with density 136. Each density is as reported during filtering phase; I began LA on the first matrix, and the log reports sparse part has density 72. So, I'm running a 23.5M matrix of density 72, with ETA around 600 hours (I use the machine for other tasks, so ETA isn't very accurate anyway).