We will change to 2_1137P Real Soon Now.
 

Hi everyone.

As many of you already know, the sieving of the 10_223P project is almost completed. The next project wil be to factor the 182-digit cofactor of 2^1137+1, aka 2_1137P. This will be a little harder than 10_223P but not markedly so.

The changeover will occur some time in the next couple of days, depending on how much work is done by the clients in that time. All of the pools have been loaded up with the new work but the precise time of the changeover for each particular pool will depend on the productivity of the clients working in each pool. That is, don't be surprised if some of your machines change before others.

The switch to the new project should be seamless and automatic. It has been in the past and we know of no reason why it should not be in the future. Making that statement, of course, is just asking for trouble. :wink:

The one thing of which you should beware is that the new project will create a new subdirectory in the projects directory and, in particular, a new roots.txt file there. This will be a few dozen megabytes in size so please make sure that you have enough free space on your disk(s) --- 100M is plenty and allows for a good safety margin.

Thanks for all your hard work so far, it is appreciated, and let's see a successful 2_1137P!


Paul

i've remark, the total WUs required for 10_223P increased of 1000 (passing from 12 000 to 13 000), so its about 2 more days for this project.

I hope seeing the switch during the week-end :)

i was taking a look into the nfsnetclient, and i saw:
[code]
14:45:43 Received assignment (Fenland 2_1137P_2 2072584-2073122)...
14:45:43 Getting project details...
14:45:43 Building .polys file...
14:45:43 Launching rootfinder...
[/code]

so i can confirm, the new project has just started :banana:

i've remark, my lines per minute speed drops at 3 (it was 6 for the 10_223P) on this machine. So we'll see in the stats tonight :)

The longhorn (texas server) is currently giving work on 10_223P

[code]
18:15:09 Requesting assignment...
18:15:09 Received assignment (Longhorn 10_223P_2 21639318-21639548)...
18:15:10 Building .in file...
18:15:10 Updating project .in file with assigned range...
18:15:10 Launching siever...
[/code]

Though the 2_1137P link isn't active yet (about 6 hours more), it should be

[url]http://www.nfsnet.org/stats2/statsreporter.cgi?template=index.html&project=2_1137P[/url]

if it is consistent with the 10_223P url.

Yep the new link is active.
Will update Nfsnet shortly

new project started
 

that's why 223 project has performance drop

27 march - 206 PC active
28 march - 62 PC active

Most systems have already switched to the new project. The stats on the main page will reflect the switch to 2_1137 soon.

Lars

[b]Project: 10_223P

Project Progress[/b]

96.35% complete (12526.5 of 13000 workunits)

But what with this? 3.65% should be done....

[QUOTE=Death][b]Project: 10_223P

Project Progress[/b]

96.35% complete (12526.5 of 13000 workunits)

But what with this? 3.65% should be done....[/QUOTE]

It will be done. NFSNet does not work on only one project at a time.
Although most of the machines have switched to 2,1137+, some are still working on 10,223+.

Actually we made a mistake and didn't put enough of the remaining work in the Fenland pool. The Fenlanders completed their assignments and switched to the new project. Once they switched, I didn't want to switch them back because that would confuse too many people.

So the machines in the Cantab and Longhorn pools will finish up 10,223+ while Fenland starts on 2,1137+. Overall, it won't matter. We still have plenty of time before the cluster will be available.

i've remark in the stats something odd.
even there's a lot of lines, there's no work units.

for this new project, whats the number of lines to do a WU ?

Plus, in the page [url]http://www.nfsnet.org/stats2/statsreporter.cgi?template=workunits.html&project=2_1137P&sortorder=d&sortcolumn=0&pagesize=&page=0[/url] , i've remark there's no graphic.

what's wrong exactly ?

thanks in advance.

Just FYI, the Longhorn pool switched over to 2,1137+ a short while ago.