mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Aliquot Sequences (https://www.mersenneforum.org/forumdisplay.php?f=90)
-   -   Team sieve #2: c136 from 4788:2381 (https://www.mersenneforum.org/showthread.php?t=11690)

schickel 2009-04-05 06:50

Team sieve #2: c136 from 4788:2381
 
This is the thread for the team factorization of the c136 from 4788, line 2381.

[b][i]Post-processing will be handled by mdettweiler[/i][/b]

[b][i]Post-processing scheduled for 4/26.....[/i][/b]

[i][b]Sieve this one using lasieve4I13e.[/b][/i]

[b]Poly[/b]:[code]n: 4917400029751399370968199585848845965968509843013458776241467726948301116079944821945421874145871641363912960613047045854615665784286241
c5: 69660
c4: 47597534097
c3: -91091868260909608
c2: -9944838367785250261704
c1: 5950705355435020029486636408
c0: 292726224984467082644800227083392
Y1: 1225481998370981
Y0: -147825692404965220984995903
skew: 499783.93
rlim: 12000000
alim: 12000000
lpbr: 28
lpba: 28
mfbr: 56
mfba: 56
rlambda: 2.6
alambda: 2.6[/code]

Save this in a ".poly" file, "4788-2381.poly" for example, then execute the siever with this command line:[code]gnfs-lasieve4I13e -a 4788-2381.poly -f <start-of-range> -c <length-of-range> -o <output-file>[/code]

[i]Check msg #5 for upload instructions.[/i]

Estimated relations needed: [b]20 million[/b] (Very rough guess....)

[b]Please check downthread for reservations not transferred up here.[/b]
[code][b]
Range Who[/b]
* 1.1M- 1.7M jrk (done, 985675 rels)
* 1.7M- 1.8M Andi47 (done, 209451 rels)
* 1.8M- 2.2M jrk (done, 824107 rels)
* 2.2M- 2.6M jrk (done, 859794 rels)
* 2.6M- 3.0M jrk (done, 875473 rels)
* 3.0M- 3.4M jrk (done, 904464 rels)
* 3.4M- 3.8M jrk (done, 899233 rels)
* 3.8M- 4.2M jrk (done, 918679 rels)
* 4.2M- 4.6M jrk (done, 912624 rels)
* 4.6M- 5.0M jrk (done, 923318 rels)
* 5.0M- 5.4M Andi47 (done, 938271 rels)
* 5.4M- 6.0M jrk (done, 1391384 rels)
* 6.0M- 6.6M jrk (done, 1388768 rels)
* 6.6M- 7.2M jrk (done, 1373265 rels)
* 7.2M- 7.8M jrk (done, 1387303 rels)
* 7.8M- 8.8M henryzz (done, 2354238 rels)
* 8.8M- 9.4M jrk (done, 1370031 rels)
* 9.4M- 9.5M Andi47 (done, 230350 rels)
*10.1M-10.7M jrk (done, 1386273 rels)
*10.7M-11.7M jrk (done, 2259687 rels)
*11.7M-12.4M jrk (done, 1581432 rels)
*12.4M-12.8M jrk (done, 897146 rels)

Total relations received: 26277524 (131.4% of 20M rels)

* line sieve 1-2500 [-4000000,4000000] mdettweiler (done, 36282 rels) [/code]

fivemack 2009-04-14 16:49

In this particular case, all that's important is that you sieve the same range; you're getting roughly the same set of candidates out and just rejecting different proportions of them.

I get:

28-bit, 13: total yield: 1848, q=15001001 (0.04003 sec/rel)
28-bit, 14: total yield: 3966, q=15001001 (0.04397 sec/rel)
27-bit, 13: total yield: 922, q=15001001 (0.07872 sec/rel)
27-bit, 14: total yield: 1965, q=15001001 (0.08762 sec/rel)

and with small prime bounds at 12 million (note I'm still searching Q around 15M)

28-bit, 13: total yield: 1706, q=15001001 (0.03774 sec/rel)
28-bit, 14: total yield: 3646, q=15001001 (0.04481 sec/rel)
27-bit, 13: total yield: 856, q=15001001 (0.07520 sec/rel)
27-bit, 14: total yield: 1824, q=15001001 (0.08388 sec/rel)

and just try one with small prime bounds at 9 million since 12 was better than 15

28-bit, 13: total yield: 1505, q=15001001 (0.03880 sec/rel)

So: siever 13e seems to be the right one, small prime bound 12M looks good, and since ~2 relations per Q appears to be a sweet spot, I'd use 28-bit large primes. The job will take under a million CPU-seconds, say one quad-core for four days.

mdettweiler 2009-04-14 17:47

[quote=10metreh;169229]Now where should we start sieving? 5M?[/quote]
5M sounds approximately good (though what do I know about this stuff? :wink:). Also, who's doing the postprocessing? I'd be glad to do it like last time, though if anyone else is interested that's fine with me too.

Andi47 2009-04-14 18:03

[QUOTE=10metreh;169229]Now where should we start sieving? 5M?[/QUOTE]

5M sounds good for the beginning - we can still go down to a lower starting point later if we need to.

[B]Reserving 5M to 5.4M[/B]

Edit: I'm getting well above 2 relations per Q at Q=5M (for 1700 sieved Q's I get 2.65 rels/Q), so I guess we can start well below 5M.

schickel 2009-04-14 21:49

Upload instructions for relations
 
Here are directions for uploading:[quote=mdettweiler]In other news: I am pleased to announce that we now have a working FTP server set up for relations uploads! :grin: Instructions as follows will show how to use it with the built-in "ftp" command line program that comes with all Windows and Linux systems; feel free to adapt them as necessary for use with other FTP programs.

---------------------------------------------------
-Open a command prompt/terminal window and type the following command:
[FONT=Courier New]ftp nplb-gb1.no-ip.org[/FONT]

-ftp will ask for a username. Type "aliquot" (capitals as shown, no quotes) and press Enter.

-ftp will ask for a password. Type "aqupload-4788" (capitals as shown, no quotes) and press Enter.

-ftp will now output the following:
[FONT="Courier New"]230 Login successful.
ftp> [/FONT]

-Type "cd c136-relations" and press Enter.

-Type "binary" and press Enter.

-Now type the following command:
[FONT="Courier New"]put relations.gz[/FONT]
...replacing "relations.gz" with the name (and, if necessary, pathname) of the compressed relations file you're uploading. Tip: navigate to the directory your relations file(s) are in *before* starting the FTP command, so you don't have to type long and clumsy file paths in your "put" commands for each file you upload.

-Wait until the file's finished uploading. Depending on the size of the file, it may take a while; please note that ftp doesn't show any progress or status info until the upload has completed, so it can appear to be "frozen" if you haven't used ftp before.

-Verify that your file's made it to the server by typing the command "ls" and seeing if your file shows up in the directory listing that results. If you see, it, you're all set!

-Type "quit" and press Enter. Close the command prompt/terminal window.
---------------------------------------------------

Some of you may have your own personal favorite GUI-based FTP program that you'd rather use than the command-line "ftp" application that I've used in my above instructions. In that case, simply tell your program to connect to server nplb-gb1.no-ip.org (port 21, the default) and log in as user [I]aliquot[/I] with password [I]aqupload-4788[/I], and upload your relations files to the c132-relations directory.

Max :smile:[/quote]

schickel 2009-04-14 21:54

[QUOTE=fivemack;169219]So: siever 13e seems to be the right one, small prime bound 12M looks good, and since ~2 relations per Q appears to be a sweet spot, I'd use 28-bit large primes. The job will take under a million CPU-seconds, say one quad-core for four days.[/QUOTE]Thanks for the assist, Tom. Next time I'll have an extra PC that I throw at the pre-testing.....

jrk 2009-04-15 05:14

[quote] -Type "cd c132-relations" and press Enter. [/quote]What directory should we be using here? Make one named c136-relations instead?

But anyway, I can't "ls"
[code]$ ftp nplb-gb1.no-ip.org
Connected to nplb-gb1.no-ip.org (76.92.239.120).
220 Welcome to the Aliquot Sequences FTP server!
Name (nplb-gb1.no-ip.org:jayson): aliquot
331 Please specify the password.
Password:
230 Login successful.
Remote system type is UNIX.
Using binary mode to transfer files.
ftp> ls
227 Entering Passive Mode (76,92,239,120,32,71).
ftp: connect: Connection timed out
ftp>
[/code]

Also tried using active mode with the appropriate firewall change and it doesn't work either. Is something wrong on the other end?

mdettweiler 2009-04-15 06:16

[quote=jrk;169307]What directory should we be using here? Make one named c136-relations instead?

But anyway, I can't "ls"
[code]$ ftp nplb-gb1.no-ip.org
Connected to nplb-gb1.no-ip.org (76.92.239.120).
220 Welcome to the Aliquot Sequences FTP server!
Name (nplb-gb1.no-ip.org:jayson): aliquot
331 Please specify the password.
Password:
230 Login successful.
Remote system type is UNIX.
Using binary mode to transfer files.
ftp> ls
227 Entering Passive Mode (76,92,239,120,32,71).
ftp: connect: Connection timed out
ftp>
[/code]

Also tried using active mode with the appropriate firewall change and it doesn't work either. Is something wrong on the other end?[/quote]
Hmm...that's strange. I don't know why it's doing that time-out thing. Let me try it myself, hold on...

Okay, I've just tried it and it worked fine (including both ls and cd commands). Maybe it was a temporary downtime on one end or the other--give it another try as soon as you get the chance.

Regarding the directory for uploads: oops! :rolleyes: I've deleted the c132-relations directory and made a new one, c136-relations; use that for uploads.

Max :smile:

jrk 2009-04-15 06:34

Strange. I still can't "ls". Just as a test I FTP'd to another server and it still works there. hmm. I'll investigate it more tomorrow.

jrk 2009-04-15 06:41

4.6M to 5.0M done. 923318 relations. (but not uploaded yet...)

[B]Reserving 4.2M to 4.6M[/B].

mdettweiler 2009-04-15 14:19

[quote=jrk;169321]Strange. I still can't "ls". Just as a test I FTP'd to another server and it still works there. hmm. I'll investigate it more tomorrow.[/quote]
Hmm...that's weird. I take it you're using a standard Linux machine with standard command-line ftp (based on your console snippet)?

jrk 2009-04-15 15:56

Yes. I'm using Fedora 10.

jrk 2009-04-15 19:51

4.2M to 4.6M done. 912624 relations.

[B]Reserving 3.8M to 4.2M[/B].

mdettweiler 2009-04-15 22:20

[quote=jrk;169365]Yes. I'm using Fedora 10.[/quote]
Very odd indeed. Have you tried using an alternate FTP client (such as a graphical one like gFTP)?

Note: if all else fails, you can send them to me via [url]http://www.rapidshare.com[/url] like we did when the FTP server went down during the last team sieve.

jrk 2009-04-16 03:33

I found some space on my server to upload the completed ranges to. You can get the files here:

[URL="http://jaysonking.com/aq/4788-2381-c136-4.2M_to_4.6M.out.bz2"]4.2M_to_4.6M[/URL] (44.8 MB)

[URL="http://jaysonking.com/aq/4788-2381-c136-4.6M_to_5.0M.out.bz2"]4.6M_to_5.0M[/URL] (45.4 MB)

[quote="mdettweiler"]Very odd indeed. Have you tried using an alternate FTP client (such as a graphical one like gFTP)?[/quote]Yes I tried kftpgrabber also but it had the same result with your host. :sad:

mdettweiler 2009-04-16 03:54

[quote=jrk;169433]I found some space on my server to upload the completed ranges to. You can get the files here:

[URL="http://jaysonking.com/aq/4788-2381-c136-4.2M_to_4.6M.out.bz2"]4.2M_to_4.6M[/URL] (44.8 MB)

[URL="http://jaysonking.com/aq/4788-2381-c136-4.6M_to_5.0M.out.bz2"]4.6M_to_5.0M[/URL] (45.4 MB)

Yes I tried kftpgrabber also but it had the same result with your host. :sad:[/quote]
Okay, thanks. BTW, guys--any other volunteers for the postprocessing? If nobody else posts regarding this within the next 2 hours, I'll assume that I'm doing it myself and go ahead and grab the files.

Joshua2 2009-04-16 03:59

I think it should fit in ur memory, else I could do it. I think my cpu will probably be tied up in my beal conjecture honors project search still.

mdettweiler 2009-04-16 04:01

[quote=Joshua2;169437]I think it should fit in ur memory, else I could do it. I think my cpu will probably be tied up in my beal conjecture honors project search still.[/quote]
Yes, this should fit plenty well within 2 GB of memory. If memory serves, the previous c132 used approximately 500 MB, so this one shouldn't be much more than that.

jrk 2009-04-16 04:52

3.8M to 4.2M done. 918679 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-3.8M_to_4.2M.out.bz2"]3.8M_to_4.2M[/URL] (45.0 MB)

[B]Reserving 3.4M to 3.8M[/B].

I'm still getting slightly better timings for the lower Q ranges.

jrk 2009-04-16 04:54

So that's 2,754,621 relations total yet.

mdettweiler 2009-04-16 05:59

Okay, it's been about two hours since I made my last call for postprocessing volunteers so I guess I'll be doing it myself. I'll go ahead and pull down jrk's relations shortly...

Edit: Frank, you can mark down the following in the reservations table:
*3.8M- 4.2M jrk (done, 918679 rels)
*4.2M- 4.6M jrk (done, 912624 rels)
*4.6M- 5.0M jrk (done, 923318 rels)

Total rels received: 2754621 (13.8% of 20M rels)

schickel 2009-04-16 08:51

[QUOTE=mdettweiler;169436]Okay, thanks. BTW, guys--any other volunteers for the postprocessing? If nobody else posts regarding this within the next 2 hours, I'll assume that I'm doing it myself and go ahead and grab the files.[/QUOTE]I was offline when you posted this, so I didn't see it. If your resources are booked, I can make room on my system for the post-processing.......just let me know.

mdettweiler 2009-04-16 14:54

[quote=schickel;169467]I was offline when you posted this, so I didn't see it. If your resources are booked, I can make room on my system for the post-processing.......just let me know.[/quote]
Don't worry, I've definitely got the available resources for the post-processing--I primarily wanted to make sure others got the chance to volunteer if they were interested. :smile:

jrk 2009-04-16 17:36

3.4M to 3.8M done. 899233 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-3.4M_to_3.8M.out.bz2"]3.4M_to_3.8M[/URL] (44.0 MB)

[B]Reserving 3.0M to 3.4M[/B].

Total relations: 3,653,854

mdettweiler 2009-04-16 18:03

[quote=jrk;169520]3.4M to 3.8M done. 899233 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-3.4M_to_3.8M.out.bz2"]3.4M_to_3.8M[/URL] (44.0 MB)

[B]Reserving 3.0M to 3.4M[/B].

Total relations: 3,653,854[/quote]
Frank, you can mark the following down:

* 3.4M- 3.8M jrk (done, 899233 rels)
Total relations received: 3653854 (18.3% of 20M rels)

jrk 2009-04-17 03:36

3.0M to 3.4M done. 904464 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-3.0M_to_3.4M.out.bz2"]3.0M_to_3.4M[/URL] (44.1 MB)

[B]Reserving 2.6M to 3.0M[/B].

Total relations: 4,558,318

mdettweiler 2009-04-17 05:36

[quote=jrk;169570]3.0M to 3.4M done. 904464 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-3.0M_to_3.4M.out.bz2"]3.0M_to_3.4M[/URL] (44.1 MB)

[B]Reserving 2.6M to 3.0M[/B].

Total relations: 4,558,318[/quote]
* 3.0M- 3.4M jrk (done, 904464 rels)
Total relations uploaded: 4558318 (22.8% of 20M rels)

(P.S.: Frank, could you leave off the *'s from the reservations until I've reported having downloaded and processed them? Those denotations are often somewhat useful to me for determining what I've received and what I haven't. If you just mark the "done" on the line if it's been done but not confirmed received yet, and then mark the * when it's also confirmed received, that would be great. :smile:)

schickel 2009-04-17 06:39

[QUOTE=mdettweiler;169576]* 3.0M- 3.4M jrk (done, 904464 rels)
Total relations uploaded: 4558318 (22.8% of 20M rels)

(P.S.: Frank, could you leave off the *'s from the reservations until I've reported having downloaded and processed them? Those denotations are often somewhat useful to me for determining what I've received and what I haven't. If you just mark the "done" on the line if it's been done but not confirmed received yet, and then mark the * when it's also confirmed received, that would be great. :smile:)[/QUOTE]Done, didn't realize you were tracking which you've downloaded.....

jrk 2009-04-17 15:56

2.6M to 3.0M done. 875473 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-2.6M_to_3.0M.out.bz2"]2.6M_to_3.0M[/URL] (42.6 MB)

[B]Reserving 2.2M to 2.6M[/B].

Total relations: 5,433,791

mdettweiler 2009-04-17 16:47

* 2.6M- 3.0M jrk (done, 875473 rels)[FONT=verdana]
[/FONT]Total relations uploaded: 5433791 (27.2% of 20M rels)

Edit: I'll reserve b=1-2500 for line sieving. That should pretty well cover all of the range for which line sieving is at all optimal.

mdettweiler 2009-04-17 18:15

* line sieve 1-2500 [-4000000,4000000] mdettweiler (done, 36282 rels)
Total relations received: 5470073 (27.3% of 20M rels)

jrk 2009-04-18 00:25

2.2M to 2.6M done. 859794 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-2.2M_to_2.6M.out.bz2"]2.2M_to_2.6M[/URL] (41.7 MB)

[B]Reserving 1.8M to 2.2M[/B].

Total relations: 6,329,867

mdettweiler 2009-04-18 03:44

* 2.2M- 2.6M jrk (done, 859794 rels)
Total relations received: 6329867 (31.6% of 20M relations)

Frank, did you forget to mark 2.6M-3.0M as received by me? I see you've noted some changes that have happened since I posted about that, so I was wondering if maybe you missed it.

schickel 2009-04-18 04:08

[QUOTE=mdettweiler;169695]Frank, did you forget to mark 2.6M-3.0M as received by me? I see you've noted some changes that have happened since I posted about that, so I was wondering if maybe you missed it.[/QUOTE]Yes, I did miss it. Sorry.....:cry:

jrk 2009-04-18 06:45

Max, what are you going to use for the post-processing?

10metreh 2009-04-18 06:47

[quote=jrk;169718]Max, what are you going to use for the post-processing?[/quote]

I think he'll use msieve again.

mdettweiler 2009-04-18 10:52

[quote=jrk;169718]Max, what are you going to use for the post-processing?[/quote]
[quote=10metreh;169719]I think he'll use msieve again.[/quote]
Yes, I'll be using msieve v1.41 (unless v1.42 comes out between now and then).

jrk 2009-04-18 17:00

1.8M to 2.2M done. 824107 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-1.8M_to_2.2M.out.bz2"]1.8M_to_2.2M[/URL] (32.9 MB)

[B]Reserving 5.4M to 6.0M[/B].

[QUOTE=mdettweiler;169732]Yes, I'll be using msieve v1.41 (unless v1.42 comes out between now and then).[/QUOTE]
Ok. I started trying out the experimental siever, and it appears it doesn't output the smallest factors. Serge tells me that the change won't affect msieve.

mdettweiler 2009-04-18 20:25

[quote=jrk;169767]1.8M to 2.2M done. 824107 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-1.8M_to_2.2M.out.bz2"]1.8M_to_2.2M[/URL] (32.9 MB)

[B]Reserving 5.4M to 6.0M[/B].


Ok. I started trying out the experimental siever, and it appears it doesn't output the smallest factors. Serge tells me that the change won't affect msieve.[/quote]
Yes, I've been following that thread, too--sounds good. :smile:

* 1.8M- 2.2M jrk (done, 824107 rels)
Total rels received: 7153974 (35.7% of 20M rels)

jasonp 2009-04-18 22:30

[QUOTE=mdettweiler;169732]Yes, I'll be using msieve v1.41 (unless v1.42 comes out between now and then).[/QUOTE]

You flatter me; it will not :)

Note that many of the line sieving relations will be duplicated by the lattice siever.

mdettweiler 2009-04-19 00:44

[quote=jasonp;169804]Note that many of the line sieving relations will be duplicated by the lattice siever.[/quote]
Hmm...I see. So, does that essentially negate any efficiency savings otherwise gained by doing that little bit of line sieving at low b-levels?

jrk 2009-04-19 01:02

5.4M to 6.0M done. 1391384 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-5.4M_to_6.0M.out.bz2"]5.4M_to_6.0M[/URL] (55.7 MB)

[B]Reserving 6.0M to 6.6M[/B].

mdettweiler 2009-04-19 03:02

* 5.4M- 6.0M jrk (done, 1391384 rels)
Total relations received: 8545358 (42.7% of 20M rels)

jrk 2009-04-19 08:52

6.0M to 6.6M done. 1388768 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-6.0M_to_6.6M.out.bz2"]6.0M_to_6.6M[/URL] (55.7 MB)

[B]Reserving 6.6M to 7.2M[/B].

jasonp 2009-04-19 13:28

[QUOTE=mdettweiler;169816]Hmm...I see. So, does that essentially negate any efficiency savings otherwise gained by doing that little bit of line sieving at low b-levels?[/QUOTE]
If previous experience is any guide, about half the relations you found by line sieving are also going to be found by lattice sieving. There are differences in the sievers that mean they won't find exactly the same relations, but also the line sieve will find relations whose large primes are above the limit of the special-q's that you test.

37k relations is not very many in the grand scheme of things; the line siever did find them much more quickly than the lattice siever would have needed. For polynomials with large amounts of skew, you really need to tell the line siever to use much larger lines (maybe with width 10x larger than the default). If you want, open the msieve.fb file and multiply the number next to SLINE by 10, then re-sieve a few lines and note the difference. It will take close to 10x longer though...

jrk 2009-04-19 16:51

6.6M to 7.2M done. 1373265 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-6.6M_to_7.2M.out.bz2"]6.6M_to_7.2M[/URL] (55.2 MB)

[B]Reserving 7.2M to 7.8M[/B].

Andi47 2009-04-19 17:18

50M to 54M done, 938271 relations.

[B]Reserving 1.7 to 1.8M[/B]

10metreh 2009-04-19 17:52

This doesn't seem to be that much of a "team" sieve. (I'm concentrating on my own sequence.)

mdettweiler 2009-04-19 18:14

[quote=Andi47;169892]50M to 54M done, 938271 relations.

[B]Reserving 1.7 to 1.8M[/B][/quote]
Andi, I don't see your relations on the FTP server. Did you forget to upload them? (Or is the upload still in progress?)

Meanwhile:
* 6.0M- 6.6M jrk (done, 1388768 rels)
* 6.6M- 7.2M jrk (done, 1373265 rels)
Total relations received: 11307391 (56.5% of 20M rels)

henryzz 2009-04-19 18:22

How much longer is this sieve going to take?
I will reserve a 1M-2M range if it will last another week.

schickel 2009-04-19 18:23

[QUOTE=10metreh;169899]This doesn't seem to be that much of a "team" sieve. (I'm concentrating on my own sequence.)[/QUOTE]True, true, but next time I can provide some horsepower. I've got just one more NFS job for myself, so I'll have some spare cycles.

mdettweiler 2009-04-19 18:24

[quote=henryzz;169906]How much longer is this sieve going to take?
I will reserve a 1M-2M range if it will last another week.[/quote]
I'm not sure how much longer it will take; probably another 3 or 4 days at the rate we're going.

jrk, I presume you're using just one machine (possibly with multiple cores?) to run your ranges? If so, then I imagine a 1M range wouldn't take too long for henryzz to crunch on a similar amount of firepower.

henryzz 2009-04-19 18:34

[quote=mdettweiler;169908]I'm not sure how much longer it will take; probably another 3 or 4 days at the rate we're going.

jrk, I presume you're using just one machine (possibly with multiple cores?) to run your ranges? If so, then I imagine a 1M range wouldn't take too long for henryzz to crunch on a similar amount of firepower.[/quote]
I can definitely do a 1M range in that time on 2 cores.
Where is the thread for the c132? I could judge my timing from that.
reserving 7.8M-8.8M

Andi47 2009-04-19 19:02

[QUOTE=Andi47;169892]50M to 54M done, 938271 relations.
[/QUOTE]

[QUOTE=mdettweiler;169903]Andi, I don't see your relations on the FTP server. Did you forget to upload them? (Or is the upload still in progress?)
[/QUOTE]

Upload was still in progress, I uploaded them to Rapidshare:

Uploaded in [URL="http://rapidshare.com/files/223288648/alq4788.2381_50.52.out.bz2"]two[/URL] [URL="http://rapidshare.com/files/223291426/alq4788.2381_52.54.out.bz2"]parts[/URL].

jrk 2009-04-19 19:12

I'm using one core. So henryzz should be able to do his range in a few hours on two cores.

mdettweiler 2009-04-19 19:46

[quote=jrk;169914]I'm using one core. So henryzz should be able to do his range in a few hours on two cores.[/quote]
* 5.0M- 5.4M Andi47 (done, 938271 rels)
Total relations received: 12245662 (61.2% of 20M rels)

jrk 2009-04-20 00:47

7.2M to 7.8M done. 1387303 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-7.2M_to_7.8M.out.bz2"]7.2M_to_7.8M[/URL] (55.8 MB)

[B]Reserving 8.8M to 9.4M[/B].

mdettweiler 2009-04-20 02:37

* 7.2M-7.8M jrk (done, 1387303 rels)
Total relations received: 13632965 (68.2% of 20M rels)

henryzz 2009-04-20 08:37

There is no command line in the first post of this thread. Could it be added please.

schickel 2009-04-20 09:07

[QUOTE=henryzz;170076]There is no command line in the first post of this thread. Could it be added please.[/QUOTE]Done....sorry, I tend to run all my NFS jobs with a script.

henryzz 2009-04-20 09:47

[quote=schickel;170077]Done....sorry, I tend to run all my NFS jobs with a script.[/quote]
If the siever was kind enough to output it's command line options it wouldn't have been necessary. I eventually used the same method the fact*.pl scripts do by creating a job file rather than a poly file.

jrk 2009-04-20 12:15

8.8M to 9.4M done. 1370031 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-8.8M_to_9.4M.out.bz2"]8.8M_to_9.4M[/URL] (55.4 MB)

[B]Reserving 1.1M to 1.7M[/B].

Andi47 2009-04-20 17:25

[B]reserving 9.4 to 9.5M[/B]

mdettweiler 2009-04-20 17:44

* 8.8M-9.4M jrk (done, 1370031 rels)
Total rels received: 15002996 (75.0% of 20M relations)

jrk 2009-04-20 19:51

1.1M to 1.7M done. 985675 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-1.1M_to_1.7M.out.bz2"]1.1M_to_1.7M[/URL] (38.0 MB)

Terrible yield on that one (1.64 rel/Q). Spot checking can be misleading.

[B]Reserving 9.5M to 10.1M[/B].

These reservations should put us very close.

mdettweiler 2009-04-20 20:00

[quote=jrk;170173]These reservations should put us very close.[/quote]
Indeed. As soon as I receive the relations from your 9.5M-10.1M range I'll try an initial filtering run to give us a better idea of where we are in the grand scheme of things.

* 1.1M-1.7M jrk (done, 985675 rels)
Total relations received: 15988671 (79.9% of 20M rels)

jrk 2009-04-21 05:03

9.5M to 10.1M done. 1370276 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-9.5M_to_10.1M.out.bz2"]9.5M_to_10.1M[/URL] (55.5 MB)

Andi47 2009-04-21 05:18

1.7 to 1.8M is done and ftp'd. (209451 relations)

Edit: 9.4 to 9.5M is done too and ftp'd. (230350 relations)

jrk 2009-04-21 05:29

Assuming we want to be 5-10% oversieved, I'll do another batch.

[b]Reserving 10.1M to 10.7M[/b].

When that and henryzz's batch are done, it should be ready for postprocessing.

mdettweiler 2009-04-21 06:40

* 1.7M-1.8M Andi47 (done, 209451 rels)
* 9.4M-9.5M Andi47 (done, 230350 rels)
* 9.5M-10.1M jrk (done, 1370276 rels)
Total relations received: 17798748 (89.0% of 20M rels)

Regarding when to start postprocessing: okay, that sounds good. My computer is currently all booked up so I can't do the preliminary filtering run as planned just yet, but it should still be OK just to hold off until your and henryzz's results come in.

henryzz 2009-04-21 06:58

[quote=mdettweiler;170304]* 1.7M-1.8M Andi47 (done, 209451 rels)
* 9.4M-9.5M Andi47 (done, 230350 rels)
* 9.5M-10.1M jrk (done, 1370276 rels)
Total relations received: 17798748 (89.0% of 20M rels)

Regarding when to start postprocessing: okay, that sounds good. My computer is currently all booked up so I can't do the preliminary filtering run as planned just yet, but it should still be OK just to hold off until your and henryzz's results come in.[/quote]
7.8M- 8.8M is complete
I will start uploading in a minute. From experience it might take over a 1/2 hour to upload.

edit: upload started i forgot how long 7-zip takes to compress:sad:

jrk 2009-04-21 07:33

Expect my upload in about 6.5 hours.

henryzz 2009-04-21 09:20

[quote=henryzz;170306]7.8M- 8.8M is complete
I will start uploading in a minute. From experience it might take over a 1/2 hour to upload.

edit: upload started i forgot how long 7-zip takes to compress:sad:[/quote]
over a half hour :lol:
it took:
ftp: 119183652 bytes sent in 3442.58Seconds 34.62Kbytes/sec.

that's almost an hour:smile:

jrk 2009-04-21 13:48

10.1M to 10.7M done. 1386273 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-10.1M_to_10.7M.out.bz2"]10.1M_to_10.7M[/URL] (56.2 MB)

mdettweiler 2009-04-22 07:43

Thanks guys--unfortunately I wasn't able to start the postprocessing today (4/21) since I was really busy all day with helping to set up a new forum over at NPLB. However, now that I've got both of your relations I'll see about getting it started tomorrow morning--stay tuned! :smile:

* 7.8M-8.8M henryzz (done, 2354238 rels)
* 10.1M-10.7M jrk (done, 1386273 rels)
Total rels received: 21539259 (108% of 20M rels)

henryzz 2009-04-22 08:48

[quote=mdettweiler;170483]Thanks guys--unfortunately I wasn't able to start the postprocessing today (4/21) since I was really busy all day with helping to set up a new forum over at NPLB. However, now that I've got both of your relations I'll see about getting it started tomorrow morning--stay tuned! :smile:

* 7.8M-8.8M henryzz (done, 2354238 rels)
* 10.1M-10.7M jrk (done, 1386273 rels)
Total rels received: 21539259 (108% of 20M rels)[/quote]
according to my time zone you posted at 22 Apr 09 08:43 AM

10metreh 2009-04-22 11:46

[quote=henryzz;170492]according to my time zone you posted at 22 Apr 09 08:43 AM[/quote]

It was early morning his time, but he thought of it as the previous day.

mdettweiler 2009-04-22 15:51

[quote=10metreh;170511]It was early morning his time, but he thought of it as the previous day.[/quote]
Yes, I figured it would be somewhat on the borderline depending on timezones, so that's why I specified the date in my post specifically to be 4/21. :smile:

Meanwhile, I'll get the postprocessing started up shortly.

mdettweiler 2009-04-22 18:04

Well, would you know! Apparently msieve wasn't satisfied with the ~21M relations we've collected:

[i]filtering wants 1000000 more relations[/i]

Looks like we'll need another 1 or 2 million relations. Load 'em up! :smile:

henryzz 2009-04-22 18:44

its worth doing at least 2M relations then
my cores are busy at this moment unfortunately

jrk 2009-04-22 19:52

[b]Reserving 10.7M to 11.7M[/b].

Andi47 2009-04-23 05:48

[QUOTE=schickel;168099]
[code][b]
Range Who[/b]
1.1M- 1.7M jrk
1.7M- 1.8M Andi47
[/code][/QUOTE]

These are already done - see postings #66 and 70.

schickel 2009-04-23 05:57

[QUOTE=Andi47;170659]These are already done - see postings #66 and 70.[/QUOTE]You're right.....I didn't go back far enough when I changed it to "awaiting post-processing".....

fivemack 2009-04-23 07:44

[QUOTE=mdettweiler;169626]
Edit: I'll reserve b=1-2500 for line sieving. That should pretty well cover all of the range for which line sieving is at all optimal.[/QUOTE]

I should have mentioned this at the time, but if you have a skew of half a million, as you do in this case, the aspect ratio of the line-sieving rectangle ought to be 500000:1 ... so b=1..2500 would mean you'd want to target a=-625 million .. +625 million.

Lattice sieving remains noticeably more efficient than line sieving.

jrk 2009-04-23 09:52

10.7M to 11.7M done. 2259687 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-10.7M_to_11.7M.out.bz2"]10.7M_to_11.7M[/URL] (91.8 MB)

mdettweiler 2009-04-23 17:53

* 10.7M-11.7M jrk (done, 2259687 rels)
Total relations received: 23798946 (119% of 20M rels)

Postprocessing will recommence shortly. :smile:

mdettweiler 2009-04-23 18:46

Well, believe it or not, we still don't have enough relations:

[i]filtering wants 276621 more relations[/i]

Probably another million or so should do the trick. :smile:

Batalov 2009-04-23 19:01

What is important is how many unique relns you had at this point.
(Check the log. Then you may be able to make a better estimate.)
For a number of this size, you'd need at least 20-22M unique relns.

[I]filtering wants 276621 more relations[/I]
is not [I]always[/I] accurate.

jrk 2009-04-23 19:20

[b]Reserving 11.7M to 12.4M[/b].

mdettweiler 2009-04-23 20:00

[quote=Batalov;170730]What is important is how many unique relns you had at this point.
(Check the log. Then you may be able to make a better estimate.)
For a number of this size, you'd need at least 20-22M unique relns.

[I]filtering wants 276621 more relations[/I]
is not [I]always[/I] accurate.[/quote]
Okay, here you go:
[I]found 3904684 duplicates and 17754868 unique relations[/I]
In that case, you guys may as well load up 2 or 3 million more (total) relations. Looks like we'll need it. :smile:

Regarding "filtering wants more relations" not being accurate: yes, I've noticed that too. Seems that it's impossible to *really* know how many more relations are needed except by trial and error, and by rough estimates based on # of unique relations needed by similar jobs.

jasonp 2009-04-24 03:24

[QUOTE=mdettweiler;170737]
Regarding "filtering wants more relations" not being accurate: yes, I've noticed that too. Seems that it's impossible to *really* know how many more relations are needed except by trial and error, and by rough estimates based on # of unique relations needed by similar jobs.[/QUOTE]
Willemien Ekkelkamp is working on [url="http://www.springerlink.com/content/747k17x78rt63341/"]a method[/url] that can extrapolate the number of relations needed based on a little test sieving, and for line sieving her results are a very accurate approximation to the real world (within 1-2%). I think some folks at EPFL are trying to improve the technique to work with lattice sieving.

But for now we only have the lazy way, that relies on history.

jrk 2009-04-24 03:25

I'll have almost 1.6M relations done in about 90 minutes. Will this be enough for another filtering run or should I reserve another range?

mdettweiler 2009-04-24 04:38

[quote=jrk;170766]I'll have almost 1.6M relations done in about 90 minutes. Will this be enough for another filtering run or should I reserve another range?[/quote]
Probably won't hurt to do another range. I won't be able to start the postprocessing until tomorrow morning (4/24 EDT) anyway, so even if we would have enough otherwise, the extra relations would help produce a smaller matrix.

Batalov 2009-04-24 04:59

[quote=jasonp;170765]Willemien Ekkelkamp is working on [URL="http://www.springerlink.com/content/747k17x78rt63341/"]a method[/URL] that can extrapolate the number of relations needed based on a little test sieving, and for line sieving her results are a very accurate approximation to the real world (within 1-2%)...[/quote]
For a moment I felt that you were pulling our (collective) leg(s).

It sounded like Eddie Izzard's sketch about made-up names...
"-- What shall we call our son so he does not get the sh!t kicked out of him at school? We shall call him [I]Englebert Humperdinck[/I]! Yes, that'll work."

Yes, I know, that's a real name. Sorry!

jrk 2009-04-24 05:12

11.7M to 12.4M done. 1581432 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-11.7M_to_12.4M.out.bz2"]11.7M_to_12.4M[/URL] (64.3 MB)

[B]Reserving 12.4M to 12.8M[/B].

jrk 2009-04-24 10:45

12.4M to 12.8M done. 897146 relations. Uploaded: [URL="http://jaysonking.com/aq/4788-2381-c136-12.4M_to_12.8M.out.bz2"]12.4M_to_12.8M[/URL] (36.5 MB)

jasonp 2009-04-24 13:31

[QUOTE=Batalov;170774]For a moment I felt that you were pulling our (collective) leg(s).
[/QUOTE]
By publishing in Lecture Notes in Computer Science using an assumed name? That's a rather elaborate joke...


All times are UTC. The time now is 15:19.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.