![]() |
|
|
#188 | |
|
A Sunny Moo
Aug 2007
USA (GMT-5)
3·2,083 Posts |
Quote:
|
|
|
|
|
|
|
#189 |
|
Mar 2006
Germany
1011010110112 Posts |
what about the PRPnet resultfiles?
i hope there're scripts to convert those files in LLRnet-format! if not, serveral people have to change their scripts for processing them (inserting in LLRnet-Database by IronBits or checking for completeness by me)! |
|
|
|
|
|
#190 | |
|
May 2007
Kansas; USA
101·103 Posts |
Quote:
Max, You're up to bat here. We definitely need this to be fed into the stats database and converted to "regular" results files for Karsten and my purposes. Can that be done fairly quickly? I'd like to get the mini-drive started within the next 1-2 days. An option is to load up LLRnet port G8000 with n=600K-700K while you work on this. After you know you have the interface working with our stats and a conversion process to "regular" results in place, the next higher n-range then could then be loaded into a PRPnet server. We have to make sure that we think of all of the impacts of having a PRPnet server. Gary |
|
|
|
|
|
|
#191 |
|
May 2008
Wilmington, DE
22·23·31 Posts |
k=395 700 -1000 complete 1 prime previously reported
results emailed |
|
|
|
|
|
#192 | |
|
A Sunny Moo
Aug 2007
USA (GMT-5)
3·2,083 Posts |
Quote:
I already developed the PRPnet/LLRnet results conversion script a while ago for Conjectures 'R Us. So, yes, the results can still be handled just as always.The only thing I *don't* have yet is a system to automatically copy off the PRPnet results files and convert them to LLRnet format, as well as a system for generating status webpages like what we've got for LLRnet. At least the results-file part of that should be somewhat easy to implement. Max
|
|
|
|
|
|
|
#193 | |
|
May 2007
Kansas; USA
101×103 Posts |
Quote:
OK, what is your ETA for converting the results to LLRnet format so that it can be fed into our stats system? The status web page would be nice but not 100% necessary at this point. FYI, let's put the JobMaxTime at 3 days for this. Gary |
|
|
|
|
|
|
#194 | |
|
A Sunny Moo
Aug 2007
USA (GMT-5)
3×2,083 Posts |
Quote:
After that's all set up (which shouldn't take long), I'll see if I can set up a "helper" LLRnet server (which would not be accessible from outside, only locally--say, on port 8001) which would always be loaded with the same workload as the PRPnet server. Then, I'll write a script that, when run, takes that day's results file, and processes it into a file formatted as an LLRnet client's tosend.txt. The script will then feed that tosend.txt file to a dedicated LLRnet client which it will use to submit the PRPnet results to the "helper" LLRnet server. The helper server will then keep track of all these in its knpairs.txt as alyways, and thus we can tie the status page into the helper server, which should in the long run be much easier than trying to wrest comparable information directly from PRPnet's vastly different file formats. This probably sounds like a huge, jumbled nightmare of a setup. In reality, if all goes well it shouldn't be too bad, though. Theoretically, this type of setup would keep the PRPnet server in perfect sync with its "helper" server, which we could then track using our normal LLRnet status scripts. The point is to make it so that on the outside, it integrates with our existing systems as easily and cleanly as an LLRnet server, with the only difference being that a slightly different piece of software is used on the client end.Hope this clears things up a bit. ![]() Max ![]() P.S.: lol, I just realized that I still didn't answer your question about what my ETA for all of this is. I should be able to get the server up and running for clients, and copying its results off to the usual results folder, either later today or sometime tomorrow. Getting it tied into the http://nplb-gb1.no-ip.org/llrnet/ status page for progress reports may take up to a few days; though I should have it up by the end of the week.
Last fiddled with by mdettweiler on 2009-02-10 at 22:22 |
|
|
|
|
|
|
#195 | |
|
May 2007
Kansas; USA
101×103 Posts |
Quote:
Wow; bunch of mumbo-jumbo there. lol OK, I'll throw out specific targets: 1. You'll have everything ready to go by Thurs. Feb. 12th so that we could start the effort. I'll make it an extra day so you aren't feeling rushed and so that we have time for some checking to make sure that everything is working as it should. 2. Let's set a target for tieing it into the web page with progress reports for a week, i.e. Feb. 17th. Does that sound good? Although I'm "pushing" this along, please don't feel rushed. We definitely want to have time to check stuff and I can be around to run tests if needed. It's better that it's delayed for a bit and that we know that it is 100% correct. This 6k effort is a good effort to do this on since it's not a huge drive like most of ours are. Thanks for your work on this. Having a working PRPnet server that LLR's as fast as manual LLR will be very exciting! ![]() Gary Last fiddled with by gd_barnes on 2009-02-11 at 00:14 |
|
|
|
|
|
|
#196 | |
|
A Sunny Moo
Aug 2007
USA (GMT-5)
141518 Posts |
Quote:
If we can get this whole setup to work smoothly and accurately for the mini-drive, then we can start gradually switching over other efforts to PRPnet. Eventually, we'll probably want to have everything running on PRPnet since LLRnet will only continue to grow older and more outdated as time goes on. Max
|
|
|
|
|
|
|
#197 | |
|
May 2007
Kansas; USA
101×103 Posts |
Quote:
|
|
|
|
|
|
|
#198 |
|
Mar 2006
Germany
32×17×19 Posts |
Max, your conversion is quite good but to much complicated for the purpose!
think about this: because Rogue(Mark) wrote the PRPnet-client/server, i think he's able to include an option like -L make LLRnet-server ouput-format (3 lines per k/n-pair) so you won't need any script or new ghost-LLRnet-server to convert them. with those there would be more chances of fails like ghost-server crashes or script won't started correctly! my opinion to the use of a new program! Last fiddled with by kar_bon on 2009-02-11 at 09:12 |
|
|
|
![]() |
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| Individual-k reservations k=300-400 n=1M-3M | gd_barnes | No Prime Left Behind | 273 | 2021-01-26 12:08 |
| Individual overall statistics graphs | Chuck | GPU to 72 | 4 | 2012-08-20 23:58 |
| Distinquishing Team from Individual accounts | strates | Teams | 5 | 2006-12-17 20:50 |
| What is the largest individual effort currently under way? | Gary Edstrom | Lounge | 10 | 2004-04-20 19:55 |
| Individual Account Report queries | shu_the_genius | PrimeNet | 5 | 2003-12-19 18:00 |