mersenneforum.org  

Go Back   mersenneforum.org > Prime Search Projects > No Prime Left Behind

Reply
 
Thread Tools
Old 2009-12-05, 01:27   #1
MyDogBuster
 
MyDogBuster's Avatar
 
May 2008
Wilmington, DE

1011001001002 Posts
Lightbulb Sieving drive for k=301-400 n=1M-2M

This is a sieving drive for k=301-400 and n=1M-2M. With testing complete to n=1M, this has become high priority so we would be grateful for any help that people can give.

Sr2sieve is what we will use for sieving. The easiest way to start the program is to specify the ranges and file name at the command prompt as in:

sr2sieve -p 4000e9 -P 4250e9 -i sieve300-400-1M-2M.txt

The above would be if you were sieving P=4000G-4250G.

When complete, you should have a factors.txt file. Please Email the file to us at:

gbarnes017 at gmail dot com

For new searchers, please post a reservation and we will Email you the latest sieve file. All factors have been removed up to P=108T. We will continue removing factors for contiguous P-ranges as we progress.

Reservations:
Code:
P-range         reserved by     status       est. completion date
    0G- 3000G   MyDogBuster     complete
 3000G- 5000G   Flatlander      complete
 5000G- 7000G   MyDogBuster     complete
 7000G- 8000G   Flatlander      complete
 8000G-10000G   Lennart         complete
10000G-14000G   MyDogBuster     complete
14000G-15000G   Flatlander      conplete
15000G-16000G   MyDogBuster     complete
16000G-17000G   Flatlander      complete
17000G-18000G   Lennart         complete
18000G-22000G   MyDogBuster     complete
22000G-23000G   Lennart         complete
23000G-38000G   MyDogBuster     complete
38000G-40000G   Flatlander      complete
40000G-55000G   MyDogBuster     complete
55000G-57000G   Flatlander      complete
57000G-93000G   MyDogBuster     complete
   93T-104T     gd_barnes       complete
  104T-113T     vaughan         complete
  113T-140T     gd_barnes       complete
Once this is done, we'll have a well-sieved file for the entire range of k=300-400 for n=1M-2M.

Ian

Last fiddled with by gd_barnes on 2010-03-21 at 01:14 Reason: status update
MyDogBuster is offline   Reply With Quote
Old 2009-12-15, 18:58   #2
mdettweiler
A Sunny Moo
 
mdettweiler's Avatar
 
Aug 2007
USA (GMT-5)

3·2,083 Posts
Default Sieve for k=301-400 n=1M-2M

Since the "new" forum at http://www.noprimeleftbehind.net/forum/ will soon be going the way of the dinosaur, I'm copying the main thread of general importance from there, the k=300-400 n=1M-2M sieve, over here. I'll set up the first post to have Ian's name on it shortly and load it up with the appropriate data.
mdettweiler is offline   Reply With Quote
Old 2009-12-15, 19:06   #3
mdettweiler
A Sunny Moo
 
mdettweiler's Avatar
 
Aug 2007
USA (GMT-5)

3×2,083 Posts
Default

Okay, that's all set now. The only thing left, of course, is to copy over all the other posts from the original thread, which would be way too much hassle to do one-by-one with the original poster's name on it as with Ian's initial post here. Instead, I've condensed the original thread into two files:

-A PDF file of the original thread in printable form:
http://www.noprimeleftbehind.net/dow...-400-sieve.pdf

-An archive containing all the factor files attached to posts in the original thread:
http://www.noprimeleftbehind.net/dow...actors.tar.bz2

All further discussion related to this sieve should take place in this thread, and not the original one.
mdettweiler is offline   Reply With Quote
Old 2009-12-15, 21:47   #4
gd_barnes
 
gd_barnes's Avatar
 
May 2007
Kansas; USA

1020010 Posts
Default

Ian,

Since you started this effort, these are totally your choice but I'll throw out a couple of suggestions:
1. You might consider changing the 1st post to trillions vs. billions. It might be a little clearer.
2. If you can Email the latest sieve file to Max, he could put a link to it in the 1st post here. That'd make it easier for people to get started on it. (including me, lol)

Max, if Ian sends you a file, would you mind loading it to the noprimeleftbehind web pages and putting a link to it in the 1st post here?

Ian, I hadn't checked the status of this effort in a couple of months. Wow, you've really made some serious effort in the sieving. Thanks! Sometime after I get back, someone remind to compute an optimum sieve depth.

My initial thought is that we will break off the sieving and testing efforts in n=1M-1.3M, 1-3M-1.6M, and 1.6M-2M pieces. Even breaking off the pieces, I'm sure the initial optimum will be P>=100T for n=1M-1.3M. If anyone has another suggestion on breaking it up, let us know.


Gary

Last fiddled with by gd_barnes on 2009-12-15 at 21:48
gd_barnes is offline   Reply With Quote
Old 2009-12-15, 23:53   #5
mdettweiler
A Sunny Moo
 
mdettweiler's Avatar
 
Aug 2007
USA (GMT-5)

3·2,083 Posts
Default

Quote:
Originally Posted by gd_barnes View Post
Max, if Ian sends you a file, would you mind loading it to the noprimeleftbehind web pages and putting a link to it in the 1st post here?
Sure, no problem.
mdettweiler is offline   Reply With Quote
Old 2009-12-16, 04:15   #6
MyDogBuster
 
MyDogBuster's Avatar
 
May 2008
Wilmington, DE

54448 Posts
Default

Quote:
1. You might consider changing the 1st post to trillions vs. billions. It might be a little clearer.
I actually like it in G's. Bigger numbers. LOL

I'll get the file ready (I need too remove the latest factors) and send it to Max sometime tomorrow.

I am a little sieved out so I'll be stopping after my current set up to 93T, oops 93000G.

Edited: Sent the file to Max with factors removed up to 85000G.

Last fiddled with by MyDogBuster on 2009-12-16 at 04:34
MyDogBuster is offline   Reply With Quote
Old 2009-12-16, 04:44   #7
mdettweiler
A Sunny Moo
 
mdettweiler's Avatar
 
Aug 2007
USA (GMT-5)

3·2,083 Posts
Default

Quote:
Originally Posted by MyDogBuster View Post
Edited: Sent the file to Max with factors removed up to 85000G.
Okay, I've uploaded it and put a link in the first post.
mdettweiler is offline   Reply With Quote
Old 2010-01-06, 00:54   #8
MyDogBuster
 
MyDogBuster's Avatar
 
May 2008
Wilmington, DE

22×23×31 Posts
Default

85T-93T complete

Who gets the factors to remove them from the main file?

I'm all sieved out for the moment.

Last fiddled with by MyDogBuster on 2010-01-06 at 00:56
MyDogBuster is offline   Reply With Quote
Old 2010-01-06, 02:03   #9
gd_barnes
 
gd_barnes's Avatar
 
May 2007
Kansas; USA

23·3·52·17 Posts
Default

Quote:
Originally Posted by MyDogBuster View Post
85T-93T complete

Who gets the factors to remove them from the main file?

I'm all sieved out for the moment.
I don't blame you. That was a tremendous effort. Thanks a bunch.

You can Email them to me. In the next few days, I'll put a couple of quads here. I've been pouring my machines on David's (Ironbits) port 2000 to dry it out and after that port 8000 to dry it out in advance of getting all servers on my machine.

A friend will be coming over sometime this week to build two more machines that I already have parts for, one of which will be the new server machine. With those built, I'll be able to throw them or a couple of my older machines on sieving here and maintain my load on David's servers.

Question: How long does it take a quad to do a P=1T range? I'm thinking I'll reserve P=93T-100T for two quads and I'm wondering how long that would take.


Gary

Last fiddled with by gd_barnes on 2010-01-06 at 02:05
gd_barnes is offline   Reply With Quote
Old 2010-01-06, 05:29   #10
MyDogBuster
 
MyDogBuster's Avatar
 
May 2008
Wilmington, DE

22×23×31 Posts
Default

It take a core about 12 days to do 1T.
MyDogBuster is offline   Reply With Quote
Old 2010-01-14, 11:15   #11
gd_barnes
 
gd_barnes's Avatar
 
May 2007
Kansas; USA

1020010 Posts
Default

I think it probably makes the most sense to break this sieving effort up into 3 n-ranges: n=1M-1.3M, 1.3M-1.6M, and 1.6M-2M while continuing to sieve the higher n-ranges at each level. I just did an optimal sieve depth calculation for breaking off n=1M-1.3M:

On a 64-bit 2.4 Ghz Intel:

Test time of an n=1.21M candidate: 1984 secs.
Removal rate at P=93T: 1526 secs.

Optimum sieve depth:
1984 / 1526 * 93T = ~121T

So, we'll make it a nice round P=120T, which means we have just P=27T to go.

Anyone have any sieving machines available? If you don't have the file with all factors removed up to P=93T, let me know and I'll Email it to you.

In the mean time, reserving P=93T-100T.


Gary

Last fiddled with by gd_barnes on 2010-01-14 at 13:10
gd_barnes is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
GPU sieving drive part III: k<10000 n=3M-6M mdettweiler No Prime Left Behind 19 2011-02-17 21:13
GPU sieving drive part II: k<10000 n=2M-3M mdettweiler No Prime Left Behind 44 2010-11-28 10:59
Bigger and better GPU sieving drive: Discussion henryzz No Prime Left Behind 75 2010-10-31 16:51
GPU sieving drive for k<=1001 n=1M-2M mdettweiler No Prime Left Behind 11 2010-10-04 22:45
Sieving drive for k=2000-3400 n=50K-1M gd_barnes No Prime Left Behind 145 2009-06-23 18:28

All times are UTC. The time now is 16:45.

Sat Sep 19 16:45:31 UTC 2020 up 9 days, 13:56, 1 user, load averages: 1.33, 1.30, 1.35

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.