I think it's time to reduce the jobmaxtime of IB400 to 2 days or even 1 day to rapidly clean the lower candidates.

[quote=em99010pepe;153906]I think it's time to reduce the jobmaxtime of IB400 to 2 days or even 1 day to rapidly clean the lower candidates.[/quote]


Yes, I was thinking the same.

Max and David, please temporarily reduce the JobMaxtime to 1 day on ports 400 and 4000. After we get all 1st drive candidates processed, then you can go back to 3 days on port 400 and 5 days on port 4000 like it is now. Current ETA on port 400 is Monday and on port 4000 is Sunday. Nugget has reported that his manual range will be done Monday and Mini-Geek has said that his will be done right around now...perfect! :-)

To all:

Max has now loaded k=800-1001/n=600K-605K into port 4000 and (I believe) has sent files to David for k=400-600/n=600K-605K to load into port 400 and k=600-800/n=600K-605K to load into port 5000.

We are attempting to make this as seemless as a transition as possible. As we move to n>600K, you need not move any of your machines to other servers unless you want to. Port 400 on this drive will just have a narrower k-range and like ports 5000 and 4000, will progress more quickly upwards by n-range if the same resources are left on it.

I am in the process of setting up the new drives now. Manual files will be posted after the 1st drive is done.


Gary

[quote=gd_barnes;154030]Yes, I was thinking the same.

Max and David, please temporarily reduce the JobMaxtime to 1 day on ports 400 and 4000. After we get all 1st drive candidates processed, then you can go back to 3 days on port 400 and 5 days on port 4000 like it is now. Current ETA on port 400 is Monday and on port 4000 is Sunday. Nugget has reported that his manual range will be done Monday and Mini-Geek has said that his will be done right around now...perfect! :-)[/quote]
Well, as of right now the lowest outstanding n on G4000 has been moving forward in lockstep with the leading edge (with Ian being the only one on the server, and none of his machines doing any funny stuff that would cause k/n pairs to be abandoned). Thus, we may be able to cruise right past n>600K without a problem, without even needing to adjust the jobMaxTime. However, yes, I will keep a close eye on it (I usually check the status page frequently throughout the day), and be sure to make any necessary adjustments. :smile:

1 day set on port 400
knpairs loaded on both 400 and 5000
port 5000 progress reports and email notifications setup
did I miss anything?

[quote=IronBits;154079]1 day set on port 400
knpairs loaded on both 400 and 5000
port 5000 progress reports and email notifications setup
did I miss anything?[/quote]
Looks good to me! :tu:

547.1-547.7 complete. One new prime and one confirmed prime. lresults attached. The lresults aren't in a perfectly linear order because of the way I added it all. (i.e. instead of one file having .1-.4 and the other .4-.7, it's .1-.2, .3-.4, and .5-.6 in one and the rest in the other) I hope that's not a problem. I'm running on IB400 now.

[quote=Mini-Geek;154086]547.1-547.7 complete. One new prime and one confirmed prime. lresults attached. The lresults aren't in a perfectly linear order because of the way I added it all. (i.e. instead of one file having .1-.4 and the other .4-.7, it's .1-.2, .3-.4, and .5-.6 in one and the rest in the other) I hope that's not a problem. I'm running on IB400 now.[/quote]
That shouldn't be too big a deal--my LLRnet processing tools can sort that out pretty quickly. The only problem...did you forget the attachment? :smile:

[quote=mdettweiler;154096]That shouldn't be too big a deal--my LLRnet processing tools can sort that out pretty quickly. The only problem...did you forget the attachment? :smile:[/quote]
Oops. I'm sure I clicked the thing to upload it, it must've just messed up somehow. It's attached now.
Perhaps it's because the first time I said to submit that post it stopped me because I had just submitted another post...I'm going to try that out real quick. Edit: nope, that doesn't seem to be it. Oh well, whatever.

[quote=Mini-Geek;154118]Oops. I'm sure I clicked the thing to upload it, it must've just messed up somehow. It's attached now.
Perhaps it's because the first time I said to submit that post it stopped me because I had just submitted another post...I'm going to try that out real quick. Edit: nope, that doesn't seem to be it. Oh well, whatever.[/quote]
Okay, thanks. I'll get the results sorted shortly, and edit my post here to attach them as soon as I'm done with that. :smile:

[quote=mdettweiler;154122]Okay, thanks. I'll get the results sorted shortly, and edit my post here to attach them as soon as I'm done with that. :smile:[/quote]
Hmm...interestingly enough, when I was processing/sorting the range, it turned out that there was one missing result! :shock: Since it was only one k/n pair, I just re-crunched it myself:

705*2^547488-1 is not prime. LLR Res64: A4BF10D574497E96 Time : 472.381 sec.

I checked in both of Mini-Geek's original results files, and this result is present in neither. Now I'm *really* glad that I re-sorted this range! :smile: (Of course, even if I hadn't found the missing result now, Gary would have seen it when he verifies the drive as a whole later on--but, it's still good that we caught it now rather than later. :smile:)

Anyway, I've attached the sorted, complete lresults file for 547.1K-547.7K. In case the timings look funny to anyone, that's because they *are* funny--for some completely unknown reason, my processing scripts have a bad habit of chewing up the times before spitting them out. :smile: This shouldn't be a big deal, though--it's not like we use the timings for anything.

[quote=mdettweiler;154124]Hmm...interestingly enough, when I was processing/sorting the range, it turned out that there was one missing result! :shock: Since it was only one k/n pair, I just re-crunched it myself:

705*2^547488-1 is not prime. LLR Res64: A4BF10D574497E96 Time : 472.381 sec.[/quote]
I just noticed an lprime.log file in one of my folders.
[code][Sat Dec 06 03:10:16 2008 - ver 3.7.1]
Error opening the results file ; see result below :
705*2^547488-1 is not prime. LLR Res64: A4BF10D574497E96 Time : 769.343 sec.
[/code]Not entirely sure why that happened, but there's the result (residues match). Perhaps a conflict with AES's CheckForPrime in opening the lresults.txt file. Normally I'd include the lprime.log file if there was one, but I didn't notice it this time.
Edit: This is my 2^10th post. :grin: