2,1049+ (SNFS difficulty of 315.78)

[pinhodecarlos]

New SNFS work as been added to the 16e Lattice Sieve V5 application. The number is 2,1049+ and it is more difficulty than 2,1037- (SNFS difficulty of 315.78).
Early status: Last 16e Lattice Sieve V5 application wu received is at ~q=1,010M (start at 1,000M, goal unknown, maybe 1,500M)

As soon as I get more info I will post more details.
Both 16e Lattice Sieve V5 and 16e Lattice Sieve are still cleaning the remaining q ranges of 2,1037-.

Carlos Pinho

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=23M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1046M (goal to unknown, then backwards from 1000M until meet 16e in the middle)

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=33M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1050M (goal to unknown, then backwards from 1000M until meet 16e in the middle)

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=60M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1064M (goal to unknown, then backwards from 1000M until meet 16e in the middle)

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=95M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1076M (goal to unknown, then backwards from 1000M until meet 16e in the middle)

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=127M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1094M (goal to unknown, then backwards from 1000M until meet 16e in the middle)

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=144M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1106M (goal to unknown, then backwards from 1000M until meet 16e in the middle)

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=174M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1128M (goal to unknown, then backwards from 1000M until meet 16e in the middle)

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=181M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1133M (goal to unknown, then backwards from 1000M until meet 16e in the middle)

[Mumps]

Last 16e Lattice Sieve application wu received is at ~q=181M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1133M (goal to unknown, then backwards from 1000M until meet 16e in the middle)

Is there any idea when the upper end of the V5 range will be seen? Or is it one of those things they won't know until they get there? :)

[pinhodecarlos]

Is there any idea when the upper end of the V5 range will be seen? Or is it one of those things they won't know until they get there? :)

I still didn't ask Greg but you cna point to 1500M.

Carlos

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=220M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1160M (goal to unknown, then backwards from 1000M until meet 16e in the middle)

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=276M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1198M (goal to unknown, then backwards from 1000M until meet 16e in the middle)

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=323M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1233M (goal to unknown, then backwards from 1000M until meet 16e in the middle)

[pinhodecarlos]

We need more 16e Lattice Sieve crunchers.

[pinhodecarlos]

We need more 16e Lattice Sieve crunchers.

Mumps? zombie67? Just for a week after the current challenge taking place.

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=356M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1243M (goal to unknown, then backwards from 1000M until meet 16e in the middle)

L'Alliance Francophone team crunchers moved to 16e Lattice Sieve application wu so NFS@Home can quickly move faster on this application (lasievef).

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=410M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1261M (goal to unknown, then backwards from 1000M until meet 16e in the middle)

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=521M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1304M (goal to unknown, then backwards from 1000M until meet 16e in the middle)

[pinhodecarlos]

For 2,1049+:

Last 16e Lattice Sieve application wu received is at ~q=766M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1391M (goal to 1650M)

Better move all cores to 16e Lattice Sieve V5.

[pinhodecarlos]

Just received a warning...we need more V5 power. Linux users please help. Thank you.

Carlos

Edit: Also we are losing ground.

[Duke of Buckingham]

You are right Carlos is time to give a week to NFS but I will be in windows. Linux machine is doing another work to pay the bills, someone has to work for money and was her time to do so. $-)

I will start tonight since the Gerasim challenge finishes. Let us see what we can both do. Come on guys a couple of days of crunching on NFS can be fun.

CHARGE
http://www.paoyeomanry.co.uk/Yeomanry/LeicsYeo/Horse%20Furniture/YeomanryCharge1930.jpg

[Fire$torm]

I'll lend a hand after the next PG challenge. See you then.

F$

P.S. You may need to remind me.... :P

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=829M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1446M (goal to 1650M)

375,000 wu´s left.

[DrPop]

...
375,000 wu´s left.

OK so is that 375K WUs left for the whole project or just for a certain sub-set?:confused:

[pinhodecarlos]

Sub-set. Those remaining wu's are only related to 2,1049+ sieve space.

[Al]

Almost done with the reconfigs. I have 1 more quad to do tonight or tomorrow and then I'll be all in.

[DrPop]

Just received a warning...we need more V5 power. Linux users please help. Thank you.

Carlos

Edit: Also we are losing ground.


Hi Carlos, I understand the losing ground part. :) However, what is meant by V5 power? Is this only crunched on Linux computers?:confused:

[pinhodecarlos]

Yes, V5 application is only available for linux 64.

[Duke of Buckingham]

Nice

1288

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=859M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1468M (goal to 1650M)

327k wu's left without taking into account the aborted, unfinished and left overs off the 2,1049+ number wu's.

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=921M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1502M (goal to 1650M)

Linux users should set to use

lasieve5f - used for huge factorizations, uses up to 1 GB memory: yes

at

http://escatter11.fullerton.edu/nfs/prefs.php?subset=project

That's very important. If we look at the sieve progress, and not taking into consideration the left overs, the 16e Lattice Sieve part will finish first than 16e Lattice Sieve V5.
Another question I would like to ask to SETI.USA members is if Greg (NFS administrator) can send to us the left overs of the 2,1049+ sieve. He can do it manually. That happened with me and Steve on last sieve project, the 2,1037-. Thank you.

Carlos

[Duke of Buckingham]

Last 16e Lattice Sieve application wu received is at ~q=921M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1502M (goal to 1650M)

Linux users should set to use

lasieve5f - used for huge factorizations, uses up to 1 GB memory: yes

at

http://escatter11.fullerton.edu/nfs/prefs.php?subset=project

That's very important. If we look at the sieve progress, and not taking into consideration the left overs, the 16e Lattice Sieve part will finish first than 16e Lattice Sieve V5.
Another question I would like to ask to SETI.USA members is if Greg (NFS administrator) can send to us the left overs of the 2,1049+ sieve. He can do it manually. That happened with me and Steve on last sieve project, the 2,1037-. Thank you.

Carlos

Very good Carlos, it is always good to have some help to Linux crunchers but we must not forget that most of the Boinc computers are Windows and that is what is missing the instructions for Windows users, thanks in advance for your help.

[pinhodecarlos]

Ricardo, soon a lasieve5f build will be available for windows.
The instructions for windows users are to set:

Run only the selected applications
lasieved - app for RSALS subproject, uses less than 0.5 GB memory: no
lasievee - work nearly always available, uses up to 0.5 GB memory: no
lasievef - used for huge factorizations, uses up to 1 GB memory: yes
lasieve5f - used for huge factorizations, uses up to 1 GB memory: no

at

http://escatter11.fullerton.edu/nfs/prefs.php?subset=project

Carlos

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at ~q=983M (goal to 1000M)
Last 16e Lattice Sieve V5 application wu received is at ~q=1535M (goal to 1650M)

Gap of 10M between "as received" and "done/valid".

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at q=1000M (goal to 1000M achieved, left overs being crunched, meanwhile 3,745+ started to be delivered)
Last 16e Lattice Sieve V5 application wu received is at ~q=1573M (goal to 1650M)

Gap of 10M between "as received" and "done/valid".

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at q=1000M (goal to 1000M achieved, left overs being crunched, meanwhile 3,745+ started to be delivered)
Last 16e Lattice Sieve V5 application wu received is at ~q=1595M (goal to 1650M)

Gap of 10M between "as received" and "done/valid".

To reduce load on the server side Greg doubled the length of all wu's as he doubled the credit, so they worth twice (as obvious).

[Duke of Buckingham]

Ricardo, soon a lasieve5f build will be available for windows.
The instructions for windows users are to set:

Run only the selected applications
lasieved - app for RSALS subproject, uses less than 0.5 GB memory: no
lasievee - work nearly always available, uses up to 0.5 GB memory: no
lasievef - used for huge factorizations, uses up to 1 GB memory: yes
lasieve5f - used for huge factorizations, uses up to 1 GB memory: no

at

http://escatter11.fullerton.edu/nfs/prefs.php?subset=project

Carlos

Thanks Carlos but at this moment I am crunching the lasieved to make 200 hours on that aplication, I have already more than enough on lasievee and lasievef. Is time to get a nice ring with a gem (or is it a star) at WUProp.

[pinhodecarlos]

16e:
Query: select * from workunit where ( name like 'S2p1049_%' and assimilate_state!=2 and appid=8 ) order by id desc limit 20
5135 records match the query.

16e V5:
Query: select * from workunit where ( name like 'S2p1049_%' and assimilate_state!=2 and appid=9 ) order by id desc limit 20
31630 records match the query.

So that's the status of the wu's left for 2,1049+ sieve.

Carlos

[pinhodecarlos]

16e:
Query: select * from workunit where ( name like 'S2p1049_%' and assimilate_state!=2 and appid=8 ) order by id desc limit 20
3713 records match the query.

16e V5:
Query: select * from workunit where ( name like 'S2p1049_%' and assimilate_state!=2 and appid=9 ) order by id desc limit 20
25934 records match the query.

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at q=1000M (goal to 1000M achieved, left overs being crunched, less than 3000 wu's available)
Last 16e Lattice Sieve V5 application wu received is at ~q=1620M (goal to 1650M)

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at q=1000M (goal to 1000M achieved, left overs being crunched, less than 2000 wu's available)
Last 16e Lattice Sieve V5 application wu received is at ~q=1626M (goal to 1650M)

We are going slower on 16e V5 progress, I suspect someone called Mumps (lol) lifted his foot and now it is decelerating. More 5 days and I am free to help. Yesterday night I found that using linear algebra stage with 4 threads instead of eight on an i7 is more beneficial. I managed to reduce by 2.5 days that stage.

Carlos

[pinhodecarlos]

Last 16e Lattice Sieve application wu received is at q=1000M (goal to 1000M achieved, left overs being crunched, less than 396 wu's available)

Last 16e Lattice Sieve V5 application wu received is at q=1650M (goal to 1650M achieved, 8362 wu's available)

[pinhodecarlos]

Query: select * from workunit where ( name like 'S2p1049_%' and assimilate_state!=2 and appid=8 ) order by id desc
0 records match the query. Yay!

Query: select * from workunit where ( name like 'S2p1049_%' and assimilate_state!=2 and appid=9 ) order by id desc limit 20
2530 records match the query.

Last 16e Lattice Sieve application wu received is at q=1000M (no wu's left)

Last 16e Lattice Sieve V5 application wu received is at q=1650M (2530 wu's available)

[pinhodecarlos]

Query: select * from workunit where ( name like 'S2p1049_%' and assimilate_state!=2 and appid=8 ) order by id desc
0 records match the query. Yay!

Query: select * from workunit where ( name like 'S2p1049_%' and assimilate_state!=2 and appid=9 ) order by id desc limit 20
0 records match the query.

Last 16e Lattice Sieve application wu received is at q=1000M (no wu's left)

Last 16e Lattice Sieve V5 application wu received is at q=1650M (no wu's available)

From Greg:


I'm trying a filtering run now. The rate of duplicates was a bit higher than I anticipated, so I ended up with a bit fewer relations than expected. I expected 670-680M unique relations and ended up with 651M. I am expecting a matrix around 75M, but if it ends up over 77M we will do a bit more sieving.

Meaning that he might need our help to crunch a few more 16e V5 wu's with new wu's of 2,1049+. He might need to stretch the sieve space of 16e V5 to 1700M. I'll keep you guys in touch.

Carlos

[pinhodecarlos]

matrix is 78020347 x 78020524 (34401.1 MB) with weight 9860353277 (126.38/col)
sparse part has weight 8159808549 (104.59/col)

linear algebra at 0.1%, ETA 651h51m

[pinhodecarlos]

2,1049+ c218 26200588389252777147795585370499300827948074837175 23305103578833836846874399003773251362379490533169 8179. p115 NFS@Home snfs

Thread can be closed. Thank you all the sievers.