mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Twin Prime Search (https://www.mersenneforum.org/forumdisplay.php?f=65)
-   -   LLRnet and PRPnet servers for automated LLR (https://www.mersenneforum.org/showthread.php?t=13805)

mdettweiler 2010-08-30 19:43

LLRnet and PRPnet servers for automated LLR
 
Hi all,

A few weeks ago, Oddball sent me a PM asking about how to set up an LLRnet server. To sum up a somewhat lengthy discussion, I offered to host a server on the existing NPLB/CRUS server equipment, and Oddball decided to go with side-by-side LLRnet and PRPnet servers to give users the option of using either. The sieve depth on the variable-n effort has advanced somewhat since then, and thus we have now set up the servers.

LLRnet and PRPnet are quite similar in concept. They have both a server and client part: the server keeps a queue of k/n pairs to be tested for potential primes, and clients connect to it over the internet to request pairs to work on. The client then feeds the number to LLR, which tests the number; then the client sends the result back to the server where it is stored.

LLRnet was the "original" automated LLR testing software. Years ago it was developed by the Riesel Sieve project and has gone through various changes over the years, during which it's been made use of by many projects in the prime search world. However, about 5 years ago (IIRC), Riesel Sieve ceased actively developing LLRnet due to their shifting more to a BOINC setup. At the time the LLRnet client had its LLR testing code built in, so it was stuck on LLR version 3.5.0 even as newer, faster, LLR versions came out.

In December 2008, Mark Rodenkirch (a.k.a. rogue on this forum) developed PRPnet, based on his existing ECMnet factoring software, as a replacement for LLRnet that would be modular, allowing for new LLR versions to be easily switched in as they came out. It also supported the use of other applications such as Phrot and PFGW, which at the time were faster for some types of numbers, and in the case of Phrot, supported non-x86 architectures. Since its initial release, PRPnet has developed into a solid, robust platform for automated primality testing and has addressed many of the shortcomings of LLRnet.

In March 2010, in response to the recent release of LLR v3.8, which improved speed an additional 4% or so over 3.7.1c (which was in turn 6% faster than the 3.5.0 used in LLRnet), Karsten Bonath (kar_bon on the forum), with the assistance of myself and Gary Barnes (gd_barnes), developed a new version of the LLRnet client which, like PRPnet, is modular and not tied to a particular LLR version. The server side of LLRnet was unmodified, but the new client remains backwards-compatible with the old one so that it can work on any LLRnet server. While this still does not address some of the other shortcomings of LLRnet (no native support for twin prime searches, no built-in stats, no email notifications of prime found, among others), it did provide another option at a time when PRPnet was not yet capable of supporting as high loads as LLRnet.

Since then, development of a PRPnet version capable of the same load capacity of LLRnet has completed, so they are now both solid options for high- and low-volume projects alike. While the servers are rather different in design, the PRPnet client and the "new" LLRnet client both work on a similar paradigm. Nonetheless, some users have a preference of one over the other, so we're therefore offering both at TPS.

Both servers are loaded with work from the n=480K-500K variable-n subproject. They are located at:

noprimeleftbehind.net port 12000 for PRPnet (currently working on n=480K-485K, k=100K-200K)
noprimeleftbehind.net port 12050 for LLRnet (currently working on n=485K-490K, k<100K)

A web page summarizing the current status of both servers can be found at [URL="http://www.noprimeleftbehind.net/tps/"][COLOR=#800080]http://www.noprimeleftbehind.net/tps/[/COLOR][/URL]. It contains links to the PRPnet server's built-in web pages for detailed information on its status, and to repositories where the results files from both servers will be copied off daily (and in the case of port 12000, converted to LLRnet format for consistency). The deadline for returning results on both servers is 3 days.

For instructions on how to set up and run each client, see:
[URL="http://www.mersenneforum.org/showthread.php?p=227767#post227767"]PRPnet instructions[/URL]
[URL="http://www.mersenneforum.org/showthread.php?p=227777#post227777"]LLRnet instructions[/URL]

If you have any questions about the server or need to report a problem with them, post in this thread--I'll be monitoring it. If you need to reach me directly, email me at [EMAIL="max@noprimeleftbehind.net"]max@noprimeleftbehind.net[/EMAIL] or send me a PM here at the forum. (Email is preferred so that I don't have to type my reply in a itty bitty text box, but PMs are OK too. :smile:)

Max :smile:

mdettweiler 2010-08-30 20:01

PRPnet client setup instructions
 
The PRPnet client is not difficult to set up and use. Those of you familiar with BOINC should get the hang of it quite easily; similar to how BOINC can juggle multiple projects and divvy up CPU time between them, PRPnet does the same with individual servers.



To set up the client:[LIST][*]Download the latest client for your operating system from [URL]http://pgllr.mine.nu/PRPNet/[/URL]. (Binaries provided courtesy of [URL="http://www.primegrid.com/"]PrimeGrid[/URL].) Windows clients are 32-bit (since LLR's calculations don't benefit from 64-bit, there's no point in making a 64-bit binary), and Linux is 64-bit (probably because the PrimeGrid people don't have a 32-bit Linux system on hand to make the binaries--let me know if you need one). Binaries are also available for Mac, both Intel and PPC, and even for Playstation 3's if you've got one of those. :smile:[*]Extract the zip file to a new folder.[*]Open the file prpclient.ini.[*]Plug in your email address and username near the top of the file. It is recommended that you replace "clientID" with a unique identifier for this particular computer; that way you can trace a particular result back to the computer it came from in the event that you submit a bad result.[*]If you wish to have your participation affiliated with a team (this does not affect individual stats, only the separate team stats), enter the team's name in the appropriate field, right under the client ID. Otherwise, leave it blank and your stats will be counted under team "None" in the team stats. Note: if your team name contains spaces, leave them out when entering it here. For instance, my team "Raiders of the Lost Primes" would be entered as "RaidersOfTheLostPrimes".[*]Below the area where you just put your user information, you'll see a section where you can configure what servers the client gets work from. Here, you can configure the proportions and queue sizes for each individual server. The instructions provided in the prpclient.ini file are rather self-explanatory, so I won't go into detail about that here. Those familiar with BOINC will feel right at home here. :smile: Add the following entry to the list of servers:
[I]TPS12000:100:1:noprimeleftbehind.net:12000[/I]
You may want to keep one or more of the PrimeGrid servers around as backup in case the TPS server goes offline or runs out of work. They should be already configured with "0" in the second field (indicating that they should only be used as backup), but you'll want to check to make sure this is the case; unlike BOINC, PRPnet's resource share numbers must total 100 for all servers. For a list of what kind of work is in each of the PrimeGrid servers, see [URL="http://www.primegrid.com/forum_thread.php?id=1215"]here[/URL]. Also, you may want to consider adding servers from the NPLB and CRUS projects as backup too (yeah, I know, shameless plug): see [URL="http://www.mersenneforum.org/showthread.php?t=12224"]here[/URL] and [URL="http://www.mersenneforum.org/showthread.php?t=12225"]here[/URL] respectively for their currently active servers.[*]Farther down the file, you'll see two options, "startoption" and "stopoption". By default, each time the client is stopped or started, it asks you what to do with any work left in its queue. This can become rather redundant during normal operation, and can be especially annoying on Windows systems where PRPnet will hold up a system shutdown until you answer its prompt. :smile: I recommend configuring the client to startoption=2 (complete assigned work units), and either stopoption=2 (return completed workunits, keep incomplete workunits in queue, and shut down) or stopoption=3 (shut down and hold everything in queue, to be picked up where it left off next time you run the program), depending on your preference. Note that stopoption=2 can delay a shutdown a bit if there are a lot of workunits to return, or if you're on dial-up; if this is important in your scenario, I'd recommend using stopoption=3.[*]Once you're satisfied with the configuration, save the file.[*]Now, run prpclient.exe (or ./prpclient from a terminal on Linux). The client will run, fetch its first batch of work from whatever server has the highest work proportion, and start working. When that batch is done, it will return the results and grab a new batch to work on. Et cetera, et cetera, ad infinitum.[*]To stop the client, press Ctrl-C. The client will stop according to your specified stopoption.[/LIST]That's all there is to it! :smile: If some of the prpclient.ini options look confusing, ignore them--they're quite safely set at their defaults. If you've got any questions, feel free to post them in this thread and either I or someone else knowledgeable of PRPnet will answer your question promptly.

Max :smile:

mdettweiler 2010-08-30 21:41

LLRnet client setup instructions
 
The main difference to note between PRPnet and the "new" LLRnet client is that the LLRnet client only can be configured with one server, whereas PRPnet can split its time between any number of servers and have backup servers in case its primary servers go down. Thus LLRnet can be slightly simpler to configure, at the expense of some flexibility. (IMHO, they're both just about as easy to configure, but opinions on this vary both ways.)


To set up the client:[LIST][*]Download the latest client for your operating system [URL="http://www.mersenneforum.org/showthread.php?t=13165"]here[/URL]. There are a few different versions available there, but for Windows you want v0.73 and for Linux you want v0.71. (Ignore the disparity in version numbers, they're really completely separate programs.)[/LIST]At this point the instructions differ depending on your operating system. For [B]Windows[/B]:[LIST][*]First of all, note that the Windows client package includes a preconfigured LLRnet [I]server[/I] setup alongside the client. For most users this is not needed and you should ignore it entirely to avoid confusion. :smile: Extract just the contents of the "LLRnet_client" subdirectory to a new folder on your computer. You can also extract LICENSE.txt and README.txt from the root directory to the same place if you'd like.[*]Right-click the file do.bat and click Edit. Do NOT double-click on it; this will run the batch file and start the program (which won't work yet since you haven't configured it).[*]At the top of the file you'll see the following lines:
[I]set op_Iterations=10000[/I]
[I]set op_Timeout=60[/I]
[I]set op_Beep=TRUE[/I]
[I]set op_indPrimeLog=FALSE[/I]
[I]set op_connect=FALSE[/I]
These are your configuration options. They're (sort of) self-explanatory: op_Iterations controls LLR's OutputIterations option (how often it should update the % complete display on the screen); op_Timeout adjusts how long the client should wait before retrying if it can't connect to the server; op_Beep sets whether the computer will make a "BEEP!" sound out of the system speaker when a prime is found (FYI, this often happens even if your volume is muted); op_indPrimeLog sets whether the program should log primes to a file in the client's own directory (TRUE) or in the directory above it (FALSE--useful for multicore systems); op_connect sets whether the client should keep waiting for an unresponsive server to come back online (TRUE), checking every op_Timeout seconds, or shut down after the first failure (FALSE). I recommend leaving op_Iterations and op_Timeout be, setting op_indPrimeLog to true if your system has only one core (but leave it at false if otherwise), and setting op_connect to true. As for op_Beep, that's truly a matter of personal preference. :smile:[*]Save and close the file. Now open llr-clientconfig.txt in a text editor (you can double-click this one).[*]Change the server and port lines as follows:
[I]server = "noprimeleftbehind.net"[/I]
[I]port = 12050[/I]
Set username to your name here on the forum and WUCacheSize to the amount of pairs you'd like the client to do in each batch. If in doubt here, 1 is a safe option.[*]Save and close the file.[*]Double-click on do.bat to run it. It will fetch the number of pairs specified in WUCacheSize, process them, return them, grab another batch and do it all over again until you hit Ctrl-C to stop the client.[*]If you shut down the client and don't plan to restart it for more than 3 days, run _cancel.bat to tell the server that you won't be processing them (otherwise, it will assume that after 3 days has elapsed). Then run _new.bat to clean out any junk left in the folder. The client will now be emptied out and ready to start afresh on new work when you run do.bat in the future.[/LIST]For [B]Linux[/B]:[LIST][*]Note: you must have Perl installed on your system to run this. (Specifically, Perl 5.8 or greater, though it may work with earlier versions and I just don't know it.) Most Linux systems have Perl pre-installed.[*]Extract the zip file to a new directory on your computer.[*]Open do.pl in a text editor. Starting at line 15 you'll see the following:
[I]# Configuration variables, set as desired[/I]
[I]$llrPath = './llr';[/I]
[I]$llrnetPath = './llrnet';[/I]
[I]$beepOnPrime = "false";[/I]
[I]$individualPrimeLog = "false";[/I]
[I]$iniOptions = "OutputIterations=10000\n";[/I]
[I]$errorTimeout = 60;[/I]
llrPath and llrnetPath can be safely left alone. beepOnPrime specifies whether the client should beep the PC speaker when it finds a prime. (Note that this often will occur even if your speakers are muted.) individualPrimeLog tells it whether to log primes to a file in the client's own directory, or in the next directory up (useful for multicore systems). iniOptions lets you specify custom options for LLR, such as OutputIterations (how often LLR should update the % complete figure on the screen); if you want to add more, be sure to affix each with "\n". errorTimeout is how long the client should wait before attempting to reconnect to a downed server.[*]After editing the configuration variables however you'd like, save and close the file. Open llr-clientconfig.txt in a text editor.[*]Set server and port as follows:
[I]server = "noprimeleftbehind.net"[/I]
[I]port = 12050[/I]
Set username to your name here on the forum and WUCacheSize to the amount of pairs you'd like the client to do in each batch. If in doubt here, 1 is a safe option. The other options can be safely ignored.[*]Save and close the file.[*]Open a terminal and run the command "perl do.pl". Alternately, you can run "chmod +x do.pl" and thereafter "./do.pl" will work the same. The client will fetch WUCacheSize k/n pairs, process them, return them, and do another batch until you press Ctrl-C to stop the client.[*]To cancel any unfinished pairs in queue, run "perl do.pl -c" (or, "./do.pl -c" if you've done the chmod thing).[/LIST]Note that the Linux client package currently contains LLR v3.8.0. The latest version is 3.8.1, which has a few minor bugfixes. The more major ones don't really affect the kind of numbers TPS does, so you should be OK with 3.8.0, but if you want to upgrade to the latest version you can grab the 3.8.1 for Linux binary at [URL]http://jpenne.free.fr/index2.html[/URL] and replace the one in your LLRnet directory with it.

If you have any questions about the LLRnet client, post them here and I or someone else knowledgeable of this stuff will answer them. :smile:

Max :smile:

Oddball 2010-09-03 07:08

The PRPnet server seems to be down, and I can't get LLRnet to work either :sad:

I'm going to do some sieving for the time being.

mdettweiler 2010-09-03 07:14

[quote=Oddball;228239]The PRPnet server seems to be down, and I can't get LLRnet to work either :sad:

I'm going to do some sieving for the time being.[/quote]
Yes, as I just posted in the NPLB forum a moment ago, Gary's having a power and internet outage over on his end. (The server and his router equipment are on UPS backup, so since the internet doesn't work the cable line must be out too.)

Gary first notified me of the outage an hour ago via text message (though I didn't receive it until just now), and (obviously) it hasn't been fixed yet. Based on how long it usually takes my power company to repair a downed line, I'm optimistic that it will be fixed by morning CDT but that's not necessarily guaranteed. And of course there's the cable company who's got to fix their end as well...so it might be mid-morning or noon CDT by the time it's back up.

Oddball 2010-09-03 16:29

[quote=mdettweiler;228242]Yes, as I just posted in the NPLB forum a moment ago, Gary's having a power and internet outage over on his end.[/quote]
It seems that you posted that message in the NPLB forum a minute or two after I posted my message, so I didn't get to see your post.

[quote]
it might be mid-morning or noon CDT by the time it's back up.[/quote]
The server is back up :smile: and we're running low on work. Could you load n=480301-480500 into the PRPnet server?

mdettweiler 2010-09-03 17:19

[quote=Oddball;228297]It seems that you posted that message in the NPLB forum a minute or two after I posted my message, so I didn't get to see your post.[/quote]Yeah, I was going to post here next but you beat me to it. :smile:[quote]The server is back up :smile: and we're running low on work. Could you load n=480301-480500 into the PRPnet server?[/quote]Sure thing--it's now loaded and ready to go again. :smile:

BTW @Lennart: if you're planning to do another big dump on the TPS PRPnet server, be sure to post about it here so we know to load some more work in advance! :smile:

Lennart 2010-09-03 17:50

[quote=mdettweiler;228304]Yeah, I was going to post here next but you beat me to it. :smile:Sure thing--it's now loaded and ready to go again. :smile:

BTW @Lennart: if you're planning to do another big dump on the TPS PRPnet server, be sure to post about it here so we know to load some more work in advance! :smile:[/quote]

I am sorry ! I did not check how much was left :smile:

Lennart

mdettweiler 2010-09-03 18:53

[QUOTE=Lennart;228311]I am sorry ! I did not check how much was left :smile:

Lennart[/QUOTE]
No problem--that's the nice thing about PRPnet, you can dry out one server and nobody goes idle. :smile:

Do you by chance know how long you'll be staying on the TPS server with this number of cores? I can then calculate an approximate figure for how much work will need to be loaded in to cover that time frame.

Lennart 2010-09-03 23:08

[quote=mdettweiler;228319]No problem--that's the nice thing about PRPnet, you can dry out one server and nobody goes idle. :smile:

Do you by chance know how long you'll be staying on the TPS server with this number of cores? I can then calculate an approximate figure for how much work will need to be loaded in to cover that time frame.[/quote]

I am not sure but I have not that many on now. I have some work to do on SR5 again I will not be on with that much core after today. I shall post here if I put many on the server.:smile:

Lennart

Lennart 2010-09-04 14:42

I have emailed Max that I will put some more on the server. I see now that it will soon dry out so I may have to move them for a while. :smile:

Lennart

mdettweiler 2010-09-04 17:14

[QUOTE=Lennart;228401]I have emailed Max that I will put some more on the server. I see now that it will soon dry out so I may have to move them for a while. :smile:

Lennart[/QUOTE]
Okay, let 'er rip. :smile: I've loaded an n=500 range this time (480501-480999) so it should last longer--though I'll watch it closely. (I'll need more sieve file to add anything more, though, as I mentioned in the LLR reservation thread.)

mdettweiler 2010-09-05 02:28

Apologies for the recent ~3 hour downtime on PRPnet--I had forgotten to set maxclients=1000 in the prpserver.ini file, so Lennart's hordes overloaded the server on the default of 10. It appears that we [i]may[/i] have also run into the dreaded "too many connections" bug (exacerbated by the anemic maxclients= figure), but I'm not entirely positive on that. Either way, though, the server was down for 3 hours and I probably could have avoided it had I paid more attention when initially setting up the server. :rolleyes:

I've now fixed the maxclients= setting and restarted the server. Hopefully we won't run into any more problems now. :smile:

mdettweiler 2010-09-05 22:50

There's another power outage at Gary's house right now, so the whole noprimeleftbehind.net site and all the LLRnet/PRPnet servers on it are down.

Man, we're getting them outages right and left now...usually Gary will go for months without any power outages, and even then they're mostly just <1 minute flickers, which the UPS can handle fine. Actually, the UPS should be able to handle the big ones as well--it can keep the computer and router going for up to ~1.25 hours. The problem seems to be on the ISP end: the internet goes kapooey every time the power cuts out. :sick:

mdettweiler 2010-09-05 23:00

[QUOTE=mdettweiler;228595]There's another power outage at Gary's house right now, so the whole noprimeleftbehind.net site and all the LLRnet/PRPnet servers on it are down.

Man, we're getting them outages right and left now...usually Gary will go for months without any power outages, and even then they're mostly just <1 minute flickers, which the UPS can handle fine. Actually, the UPS should be able to handle the big ones as well--it can keep the computer and router going for up to ~1.25 hours. The problem seems to be on the ISP end: the internet goes kapooey every time the power cuts out. :sick:[/QUOTE]
...and we're back! :big grin: That was fast, fortunately. And thanks to the UPS, even though the server was inaccessible for a little while in there, it picked up right where it left off when the internet came back online.

Lennart 2010-09-06 01:48

[quote=mdettweiler;228319]No problem--that's the nice thing about PRPnet, you can dry out one server and nobody goes idle. :smile:

Do you by chance know how long you'll be staying on the TPS server with this number of cores? I can then calculate an approximate figure for how much work will need to be loaded in to cover that time frame.[/quote]

I will not have many left now before next weekend :smile:

Lennart

mdettweiler 2010-09-06 04:40

Oddball, I see the LLRnet server is going to run out of work in a few minutes. Do you want me to load in a little more work (say, n=481501-481600)?

Oddball 2010-09-06 05:12

[quote=mdettweiler;228623]Oddball, I see the LLRnet server is going to run out of work in a few minutes.[/quote]
I just finished drying it up :wink:

[quote]Do you want me to load in a little more work (say, n=481501-481600)?[/quote]
It looks like LLRnet is less popular than PRPnet, so you can load n=484900-484999 instead of n=481501-481600.

mdettweiler 2010-09-06 13:20

[quote=Oddball;228626]It looks like LLRnet is less popular than PRPnet, so you can load n=484900-484999 instead of n=481501-481600.[/quote]
Done.

mdettweiler 2010-09-13 06:12

The servers are down right now--this time it's actually not a power outage, but rather a strange double outage of phone and internet. (Gary's internet is cable-based, so one wouldn't expect them to necessarily go out together.) Hopefully it will be back up soon.

Oddball 2010-10-01 06:33

Max, could you replace n=481400-481500 on the PRPnet server with these candidates?

[URL]http://www.sendspace.com/file/4asdih[/URL]

gribozavr just posted that file a short time ago, and it has a greater sieve depth.

Also, the LLRNet server has dried out, so it'll be nice if you load n=484000-484899 into port 12050.

mdettweiler 2010-10-01 07:39

[QUOTE=Oddball;232189]Max, could you replace n=481400-481500 on the PRPnet server with these candidates?

[URL]http://www.sendspace.com/file/4asdih[/URL]

gribozavr just posted that file a short time ago, and it has a greater sieve depth.

Also, the LLRNet server has dried out, so it'll be nice if you load n=484000-484899 into port 12050.[/QUOTE]
Done. The current load of each server is now:

[b]Port 12000 (PRPnet)[/b]
n=481000-481399 from the 5 September sieve file (in progress)
n=481400-481500 from the 1 October sieve file

[b]Port 12050 (LLRnet)[/b]
n=484900-484999 from the 5 September sieve file (in progress--1 pair left)
n=484000-484899 from the 1 October sieve file

Note that for LLRnet I listed the ranges seemingly "out of order". This reflects the exact order in which the server "sees" them and is due to a slight difference in how PRPnet and LLRnet handle these. PRPnet is set to sort by n, so it looks in the database for the lowest available n (and the lowest available k within that). LLRnet, on the other hand, reads pairs in the exact order they're listed in knpairs.txt. In this case, when adding the new range I concatenated the new pairs onto the existing knpairs.txt, so the one last pair from the old range actually comes first (even though it's higher n-wise). This will be reflected on the web page's first/last pair display (at least until that one last pair from the old range is completed).

Confused yet? :wink: (It's 3:30 AM here so please pardon me if this wasn't explained as well as it could be.)

Oddball 2010-10-06 06:45

The server seems to be down again :sad:

mdettweiler 2010-10-06 06:51

[QUOTE=Oddball;232685]The server seems to be down again :sad:[/QUOTE]
Hmm, it works for me. Additionally, I've been navigating the noprimeleftbehind.net web pages for the last 5 minutes or so before you posted this, and didn't notice any problems...strange. :huh:

Is it still down for you, or is it working again now?

Oddball 2010-10-06 07:07

[QUOTE=mdettweiler;232687]Hmm, it works for me. Additionally, I've been navigating the noprimeleftbehind.net web pages for the last 5 minutes or so before you posted this, and didn't notice any problems...strange. :huh:

Is it still down for you, or is it working again now?[/QUOTE]
It's still down for me, and I don't know why. Neither the LLRnet server or the PRPnet server works (the error code for the PRPnet server is 10060). The [URL]http://www.noprimeleftbehind.net/[/URL] website and the [URL]http://www.noprimeleftbehind.net/tps[/URL] sites both don't work, but I can get to other sites (Yahoo, Google, Wikipedia, etc).

I'll restart the computer and delete browsing history and cookies to see whether that solves anything.

[B]edit: Didn't need to do that, everything's working fine now. I wonder what happened earlier :confused:[/B]

mdettweiler 2010-10-06 07:11

[QUOTE=Oddball;232691][B]edit: Didn't need to do that, everything's working fine now. I wonder what happened earlier :confused:[/B][/QUOTE]
Hmm, would you know. :huh: The best explanation I can come up with for this would be DNS troubles at your ISP; that can cause outages to some sites but not others. ISP DNS issues are usually resolved pretty quickly (compared to other types of outages), which would be consistent with what happened here.

MooMoo2 2010-11-24 06:48

The PRPnet server has dried out.

I'll be back when more candidates from the most recent sieve file are loaded onto port 12000.

mdettweiler 2010-11-24 07:23

[QUOTE=MooMoo2;238471]The PRPnet server has dried out.

I'll be back when more candidates from the most recent sieve file are loaded onto port 12000.[/QUOTE]
Ah, thanks for the heads-up--I had been monitoring the server and was aware that it was going to dry out soon but it's good to know that it will be OK for you (currently the only active PRPnet contributor) to wait for a new sieve file to reload it. I was initially going to just load more from the current file, inefficient as that may be.

Oddball 2010-11-28 04:12

The new small k file is now available:
[URL]http://www.sendspace.com/file/ul26bb[/URL]

Max, could you load n=481501-482000 into the PRPnet server?

mdettweiler 2010-11-28 05:23

[QUOTE=Oddball;238993]The new small k file is now available:
[URL]http://www.sendspace.com/file/ul26bb[/URL]

Max, could you load n=481501-482000 into the PRPnet server?[/QUOTE]
Done.

cipher 2011-01-08 17:24

Just a thought. We should be tracked on NPLB's home page, with the point system. Also we should be tracked or get credit on
[url]http://stats.free-dc.org/stats.php?page=proj&proj=nplb[/url]

Just a thought. I know twin prime is not an NPLB project how hard would it be to convert the existing points to NPLB and maybe being on NPLB home page we can attract new participants.

Cipher.

mdettweiler 2011-01-08 17:46

[QUOTE=cipher;245144]Just a thought. We should be tracked on NPLB's home page, with the point system. Also we should be tracked or get credit on
[url]http://stats.free-dc.org/stats.php?page=proj&proj=nplb[/url]

Just a thought. I know twin prime is not an NPLB project how hard would it be to convert the existing points to NPLB and maybe being on NPLB home page we can attract new participants.

Cipher.[/QUOTE]
Actually, I have given some thought to this idea before (in a PM discussion with Oddball)...what I was thinking is that TPS could have their own distinct stats system based on the same underlying code as NPLB's. I've been meaning to talk to Dave, our stats admin, about this but haven't yet gotten the chance...been kind of busy in "real life". It's on my to-do list... :smile:

mdettweiler 2011-01-16 06:34

upcoming planned downtime on PRPnet port 12000
 
Hi all,

Over in the LLR reservation thread, I just posted the reservation of the last n=480K-485, k<100K range, which has been loaded into PRPnet. This should last us until the end of the rally and probably won't take too long to polish off after that.

At that point, [b]I will need to dry out the server[/b] so I can upgrade it to the latest PRPnet server version (4.1.3 as of this writing). The "normal" upgrade process going from 3.3.x to 4.0.x is rather arduous (I tried it on some personal servers and it was rather messy), so it will be much easier for me (and with less chance of error) to simply dry out the server, then start up a completely fresh one under version 4.0.x. Not to worry, though, [b]all stats will be copied into the "new" database when I do this[/b]. Once I've done that, I'll upgrade the 4.0.x server to 4.1.x (a much simpler process that only consists of a couple quick SQL commands). The total downtime shouldn't be too long; the main thing for the end user will be that the server will be out of work for a day or two.

That done, I will load the next range (n=480K-485K, k=100K-200K) into PRPnet as discussed in the "What's next?" thread. As also discussed there, LLRnet will proceed on to n=485K-490K, k<100K. It will not be affected by the brief downtime on PRPnet.

Hopefully this will not cause any problems for anyone. On the plus side, 4.1.x will give us a number of cool new features:

-Built-in support for teams on the server stats page (should make things much simpler for Oddball in the next rally :smile:)
-Automatic testing of +1 counterparts for -1 primes (this is technically present already in 3.3.6, but it didn't work well for TPS since it tested +1 first instead of -1 and didn't save secondary residues)
-And a host of other features that don't really affect the way TPS will be using the server :smile:

Max :smile:

mdettweiler 2011-01-17 19:18

Now that the rally is concluded and just about everyone's pulled their cores off, it looks like we got as far as n=483335. We need to get it up to n=483999 to dry out the server. Anybody able to spare a computer to help with this? :smile: (My own resources, unfortunately, are booked until the end of January.)

Oddball 2011-01-29 18:20

[QUOTE=mdettweiler;246690]That done, I will load the next range (n=480K-485K, k=100K-200K) into PRPnet as discussed in the "What's next?" thread. [/QUOTE]
For the k=100K-200K part, could you load n=483K-484K instead of n=480K-481K? I'm beginning to suspect that something is wrong with that sieve file (n=483K-485K was originally sieved by pschoefer before being turned over to the distributed effort). It's not the large prime gap from n=482995-483634 that makes me suspicious, it's the fact that there might be three of them in the n=483K-485K range (one from n=482995-483634, one from 483634-484107, and one from n=484406-484996). Even if we do find a prime or two before the PRPnet and LLRnet servers dry out, there will still be an unusually small amount of primes in the n=483K-485K range.

mdettweiler 2011-01-29 21:01

[QUOTE=Oddball;250315]For the k=100K-200K part, could you load n=483K-484K instead of n=480K-481K? I'm beginning to suspect that something is wrong with that sieve file (n=483K-485K was originally sieved by pschoefer before being turned over to the distributed effort). It's not the large prime gap from n=482995-483634 that makes me suspicious, it's the fact that there might be three of them in the n=483K-485K range (one from n=482995-483634, one from 483634-484107, and one from n=484406-484996). Even if we do find a prime or two before the PRPnet and LLRnet servers dry out, there will still be an unusually small amount of primes in the n=483K-485K range.[/QUOTE]
Okay, will do.

Shouldn't be too long now before the PRPnet server dries out and I can upgrade/reload it...should be sometime on Sunday assuming you can stay on until the end. (Otherwise it will be a few more days with just my one quad working on it, plus the core or two from Robert47.)

Lennart 2011-01-31 18:52

[QUOTE=mdettweiler;250349]Okay, will do.

Shouldn't be too long now before the PRPnet server dries out and I can upgrade/reload it...should be sometime on Sunday assuming you can stay on until the end. (Otherwise it will be a few more days with just my one quad working on it, plus the core or two from Robert47.)[/QUOTE]


I'll help you dry it out.:smile:


Lennart

mdettweiler 2011-01-31 18:52

[QUOTE=Lennart;250658]I'll help you dry it out.:smile:


Lennart[/QUOTE]
:bow:

mdettweiler 2011-01-31 21:58

The PRPnet server has now finished drying out. Much thanks to everyone who helped us get it cleaned up! :smile:

Now all we have left of the n=480K-485K, k<100K range is what's currently in the LLRnet server (<24 hours worth of work on a quad).

I'll get to work on upgrading the PRPnet server to version 4.1.4 sometime today. I will, however, need some more sieve files before I can load it with work, and the same goes for LLRnet when it imminently runs out. Grivozavr, can you please post the following files?

n=480K-485K, k=100K-200K (for PRPnet)
n=485K-490K, k<100K (for LLRnet)

Thanks,
Max :smile:

mdettweiler 2011-02-01 06:47

PRPnet server upgrade successful!
 
Hi all,

I am pleased to report that the PRPnet server has been successfully upgraded from version 3.3.6 to 4.1.4 (the latest as of this writing)! :big grin: The upgrade went smoothly and all data is intact.

A couple reminders about what's new and what to keep in mind with the new release:

-4.x has better support for the "Twin" worktype, so we are now using that instead of just "Fixed b, n, c" as before. This means that all -1 primes will have their +1 counterparts tested [b]automatically[/b]. Oddball, you should be able to find the residuals for these supplemental tests in the PRPnet-format daily results files posted at [url]http://www.noprimeleftbehind.net/tps/results/prpnet/[/url] for posting in the primes thread.

Note that I'm not entirely positive yet whether the server will display the candidates on the web page grouped by k or by n (since as of yet I have no sieve file to load into the server for its next range). For the type of work TPS is doing, displaying one web page line for each k is a recipe for disaster whenever anyone tries to load the web page, since there are 50,000 k's in there at a time for only a few n's each. If the server ends up displaying it this way, I'll have to switch it back to "Fixed b, n, c" and put in a feature request for twin-servertype grouping options in a future release. :smile:

-If you are using a 4.1.x client, you can specify a team name under which to be counted in the team stats. You can, of course, leave this blank, in which case your stats will go under team "None". Note that pre-4.1.x clients, which do not support teams, simply show a null entry in the team column (as opposed to even "None") and their results aren't even counted at all in the team stats.

-There is a small bug in this server version that causes the "Team" and "Client" columns to be switched in the User Primes section on the [url=http://noprimeleftbehind.net:12000/all.html]web page[/url]. This is a cosmetic bug only; [b]all the data is still in the right place on the back end[/b]. It will be fixed in the next PRPnet release.

That's pretty much all. Now I just need to wait for gribozavr to send me the new sieve files and we'll be of! :grin:

Max :smile:

gribozavr 2011-02-01 18:35

[url]http://www.sendspace.com/file/owhcei[/url] n_480000-484999__k_100000_200000__01feb2010.txt
[url]http://www.sendspace.com/file/e5f4m2[/url] n_485000-489999__k_1_100000__01feb2010.txt

mdettweiler 2011-02-01 19:35

LLRnet and PRPnet have now been loaded with their new reservations as noted in the LLR Reservation Thread.

To update on what I mentioned yesterday regarding how PRPnet displays Twin-servertype group stats on the web page: I talked with Mark (the PRPnet developer) and he mentioned that while it does group by k by default, only a small code modification is needed to make it group by n. The modification seems to have worked successfully and as such, we will indeed be able to use the Twin servertype (i.e. automatic testing of +1 counterparts to -1 primes) from now on. :smile:

Oddball 2011-02-05 20:48

[QUOTE=mdettweiler;250754]-4.x has better support for the "Twin" worktype, so we are now using that instead of just "Fixed b, n, c" as before. This means that all -1 primes will have their +1 counterparts tested [B]automatically[/B]. Oddball, you should be able to find the residuals for these supplemental tests in the PRPnet-format daily results files posted at [URL]http://www.noprimeleftbehind.net/tps/results/prpnet/[/URL] for posting in the primes thread.
[/QUOTE]
The +1 testing feature is acting rather strangely. There are no residues in the result files, and usernames are missing. Here's what comes out:

[code]
user=mdettweiler
[2011-02-05 07:26:18]
121677*2^483018-1 is prime! Time : 0.0 sec.
user=
[2011-02-05 07:26:18]
121677*2^483018+1 is not prime. Res64: Time : 0.0 sec.
[/code]

The server stats on [URL]http://noprimeleftbehind.net/tps/[/URL] also doesn't display the info properly. Right now, I'm getting:
[code]
Results submitted today by:
mdettweiler : 173
: 1
Total results submitted today: 174
[/code]

mdettweiler 2011-02-05 21:09

Okay, thanks for the heads-up. I need to tweak my LLRnet-format conversion script to handle the different output format used for supplemental tests.

mdettweiler 2011-02-05 21:31

[QUOTE=mdettweiler;251443]Okay, thanks for the heads-up. I need to tweak my LLRnet-format conversion script to handle the different output format used for supplemental tests.[/QUOTE]
Fixed now. Should be OK for all future such supplemental tests. :smile: (Well, there is a small chance that it might barf on an actual twin since I wasn't able to test it on actual data from such an occurrence...but I'll cross that bridge if and when we come to it.)

Oddball 2011-03-27 18:32

The LLRnet server has dried out.

Max, could you load n=486001-486300 into port 12050? Usually, I would ask for a bigger range (n=486001-487000), but I'm waiting until April's sieve rally is over before loading n=486301 and up.

mdettweiler 2011-03-27 19:39

[QUOTE=Oddball;256739]The LLRnet server has dried out.

Max, could you load n=486001-486300 into port 12050? Usually, I would ask for a bigger range (n=486001-487000), but I'm waiting until April's sieve rally is over before loading n=486301 and up.[/QUOTE]
Okay, I'll do that a little later today.

Oddball 2011-04-20 07:02

Max, could you load 100K<k<200K, n=480000-481000 into the PRPnet server and k<100K, n=486300-487000 into the LLRnet server? Port 12000 and 12050 have both dried out.

Use the newest sieve files, which have been posted by gribozavr here:

[URL]http://www.mersenneforum.org/showpost.php?p=259026&postcount=8[/URL]

Lennart 2011-04-21 02:14

[URL]http://prpnet.mine.nu:12098/all.html[/URL]

Here you can see a test I have done on Twin search.
This is on PRPNet 4.3.2

As you can see I added n= 333333 twin just to test that it works.

You also get all composite +1 test with residue on those -1 test that is prime.

+1 test will only be done if -1 test is Prime

It start test -1 ( if you load -1) then +1 and if that also is prime it does a GF test.

Lennart

Oddball 2011-04-24 05:18

It looks like there's a mini-rally going on now :smile: Today, 18217 k/n pairs were tested - 16458 on PRPnet and 1759 on LLRnet. This is TPS's second highest daily LLR rate, trailing only the 29,961 pairs tested on January 16th, which was the last day of the TPS Discovery Rally.

As you might have guessed, this means that the PRPnet server will dry out soon. So Max, could you load n=481000-482000 into the PRPnet server? There are only ~2500 candidates in port 12000, and that number will hit zero in a couple of hours at the rate we're going now.

ellipse 2011-04-24 06:12

[QUOTE=Oddball;259444]It looks like there's a mini-rally going on now :smile: [/quote]
The "mini-rally" may be turning into something big. 1316 pairs were LLR'ed in the past hour, which will be a record 31,584 pairs/day if it's sustained.

Anyone up for an impromptu rally?

mdettweiler 2011-04-24 14:14

[QUOTE=Oddball;259444]It looks like there's a mini-rally going on now :smile: Today, 18217 k/n pairs were tested - 16458 on PRPnet and 1759 on LLRnet. This is TPS's second highest daily LLR rate, trailing only the 29,961 pairs tested on January 16th, which was the last day of the TPS Discovery Rally.

As you might have guessed, this means that the PRPnet server will dry out soon. So Max, could you load n=481000-482000 into the PRPnet server? There are only ~2500 candidates in port 12000, and that number will hit zero in a couple of hours at the rate we're going now.[/QUOTE]
Looks like Lennart decided to give his newly expanded farm a whirl here. :smile:

My apologies if any record-breakings were prevented by the inconvenient drying of the server...it's reloaded now. I'll continue to keep a close eye on it as long as this period of high activity continues.

@ellipse: interesting idea--how high can we go? :smile: I can't join in myself at the moment (leaving for church in a few minutes and don't have time to switch anything over) but I may be able to finagle something when I get home.

ellipse 2011-04-24 20:33

[QUOTE=mdettweiler;259467]@ellipse: interesting idea--how high can we go? :smile: I can't join in myself at the moment (leaving for church in a few minutes and don't have time to switch anything over) but I may be able to finagle something when I get home.[/QUOTE]
Right now (15:30 CDT), we're at 9883 tests. If we continue at a rate of 1200 tests/hour or more for the rest of the day, we'll get above 20,000 tests. That's higher than yesterday's rate but lower than the all-time rate on January 16th.

What we can do is set a new 3-day record and a new 24 hour record tomorrow. The Jan 14-16 discovery rally had 53,179 pairs tested at an average of 17,726 pairs/day. That's lower than yesterday's rate and likely lower than today's rate as well. If we break the 24 hour record tomorrow, we'll be looking at ~70,000 pairs completed from April 23-25: ~18,000 pairs yesterday, ~20,000 pairs today, and possibly ~32,000 pairs tomorrow.

My internet connection is a bit spotty at the moment, but I'll put some cores on this as soon as the problem is resolved.

Oddball 2011-04-25 17:16

The party's over :sad: We finished 19,119 pairs yesterday (16989 on PRPnet and 2130 on LLRnet), which was higher than the day before but not enough to beat January's record. Today's halfway over, and we're only at 4135 candidates tested. That's a higher rate than usual, but far short of what's needed to set a new record. I'll be pulling off my cores soon to work on sieving and to verify the primes that have been found during the past few days.

One more thing: Max, could you remove the rows for n=480000-481500 and n=483001-484000 on the PRPnet stats page? All k/n pairs in those ranges have been tested, and it'll make this page load much faster:
[URL]http://noprimeleftbehind.net:12000/all.html[/URL]

Finally, I'd like to welcome Beyond to TPS. He put a core or two on the LLRnet server yesterday.

mdettweiler 2011-04-25 18:17

[QUOTE=Oddball;259562]One more thing: Max, could you remove the rows for n=480000-481500 and n=483001-484000 on the PRPnet stats page? All k/n pairs in those ranges have been tested, and it'll make this page load much faster:
[URL]http://noprimeleftbehind.net:12000/all.html[/URL][/QUOTE]
Okay, I've removed n=480000-480999 and n-483001-484000. The way the database is set up, I need to match ranges based on text strings (for instance, "%*2^480%-1" where % is any character) rather than on integer ranges; thus, it's somewhat inconvenient to remove chunks more granularly than n=1K at a time.

Oddball 2011-04-30 21:37

The status of the PRPnet server here:

[URL]http://noprimeleftbehind.net:12000/all.html[/URL]

has been down the past day or two, but the results found on [URL]http://nplb-gb1.no-ip.org/tps/[/URL] indicates that port 12000 has dried out.

Max, could you load n=482000-483000 into the PRPnet server?

mdettweiler 2011-04-30 22:44

[QUOTE=Oddball;260063]The status of the PRPnet server here:

[URL]http://noprimeleftbehind.net:12000/all.html[/URL]

has been down the past day or two, but the results found on [URL]http://nplb-gb1.no-ip.org/tps/[/URL] indicates that port 12000 has dried out.

Max, could you load n=482000-483000 into the PRPnet server?[/QUOTE]
Indeed, it looks like Lennart got his clients switched over to the alternate server address* and cleaned out the server. :smile:

I'm loading up 482001-483000 right now. Note however that the 19 April 2011 sieve file for k=100K-200K only goes up to n=483000. Grivozavr, could you please send me a file for k=100K-200K, n=484001-484999? Thanks.

We've been working through the k=100K-200K range amazingly quickly...at this rate we'll be on to k=200K-300K very shortly. :smile: That just gave me an idea, though...perhaps we could try doing all of k=200K-400K at once this time (same n-range, twice the k-range)? We've been working quickly enough that we'd still proceed at a decently fast rate, but the admin management effort would be halved.

[size=1]*For those who haven't seen the notice over at NPLB, we've been having some problems with the noprimeleftbehind.net domain lately and it isn't pointing to the right IP address. However, our old domain, nplb-gb1.no-ip.org, still works and gets to exactly the same place, so you can just replace noprimeleftbehind.net with nplb-gb1.no-ip.org in your clients or web links to get through.[/size]

ellipse 2011-04-30 23:43

[QUOTE=mdettweiler;260066]We've been working through the k=100K-200K range amazingly quickly...at this rate we'll be on to k=200K-300K very shortly. :smile: That just gave me an idea, though...perhaps we could try doing all of k=200K-400K at once this time (same n-range, twice the k-range)? We've been working quickly enough that we'd still proceed at a decently fast rate, but the admin management effort would be halved.
[/QUOTE]
If we finish sieving the n=485K-490K range to 6002T, we could load n=487001-489999, k<100K into PRPnet instead of k=200K-400K. The sieve depth for the entire n=480K-490K range will be the same, so it'll make sense to test the lower k's first.

mdettweiler 2011-04-30 23:53

[QUOTE=ellipse;260069]If we finish sieving the n=485K-490K range to 6002T, we could load n=487001-489999, k<100K into PRPnet instead of k=200K-400K. The sieve depth for the entire n=480K-490K range will be the same, so it'll make sense to test the lower k's first.[/QUOTE]
Indeed. The question is, though, whether we will finish getting n=485K-490K to optimal depth in time; if Lennart continues his periodic bursts of high firepower on port 12000, we'll be needing more LLR work long before then. :smile:

Oddball 2011-05-04 06:26

[QUOTE=ellipse;259445]The "mini-rally" may be turning into something big. 1316 pairs were LLR'ed in the past hour, which will be a record 31,584 pairs/day if it's sustained.

Anyone up for an impromptu rally?[/QUOTE]
We're now completing tests at a rate of 1365 pairs/hour (32760 pairs/day if sustained) :smile:

At this rate, the PRPnet server will dry out in a few hours, so it's time to load the next range into the server. Max, could you load n=489K-490K for k=<100K if n=484K-485K from k=100K-200K is not available?

mdettweiler 2011-05-04 18:21

[QUOTE=Oddball;260457]We're now completing tests at a rate of 1365 pairs/hour (32760 pairs/day if sustained) :smile:

At this rate, the PRPnet server will dry out in a few hours, so it's time to load the next range into the server. Max, could you load n=489K-490K for k=<100K if n=484K-485K from k=100K-200K is not available?[/QUOTE]
Loading right now. :smile:

BTW @all: as recently posted over in the NPLB forum, the noprimeleftbehind.net DNS issues have been resolved. Everything should be working fine now for most people but if you're still having lingering issues, they should clear up within the next half-day or so as various DNS servers around the world update their cached info.

Also note that due to an incidental, but not directly related issue, mail sent to my [email]max@noprimeleftbehind.net[/email] address started bouncing late yesterday. I've fixed the problem, but again, it may take a day or so for the changes to be distributed. If you have problems sending mail to me, you can send mail to [email]max@noprimeleftbehind.net.test-google-a.com[/email], which is independent of our DNS records and will get to the same mailbox.

Oddball 2011-06-04 08:31

Max, could you load n=487001-487500 into port 12050 (LLRnet) and n=487501-489000 into port 12000 (PRPnet)? The LLRnet server has dried out, and the PRPnet server will dry out soon.

Shortly from now, gribozavr will probably be uploading the latest k<100K file in this thread:
[URL]http://www.mersenneforum.org/showthread.php?t=12260[/URL]
But if the file isn't there by Monday at 10PM CDT, just go with the latest version that you have.

mdettweiler 2011-06-04 17:00

[QUOTE=Oddball;262974]Max, could you load n=487001-487500 into port 12050 (LLRnet) and n=487501-489000 into port 12000 (PRPnet)? The LLRnet server has dried out, and the PRPnet server will dry out soon.

Shortly from now, gribozavr will probably be uploading the latest k<100K file in this thread:
[URL]http://www.mersenneforum.org/showthread.php?t=12260[/URL]
But if the file isn't there by Monday at 10PM CDT, just go with the latest version that you have.[/QUOTE]
Done. :smile:

It seems that with this, we have loaded the last remaining work for k<100K, n=480K-490K. However, as of right now, the 490K-495K range is hardly well-sieved; we're not quite even at 1P yet. So it would be rather inefficient to continue on to testing n>490K until sieving has progressed further. In the meantime, we do have plenty of well-sieved work in the k=100K-200K range; everything from n=484001-489999 is essentially done sieving and ready to go. I would recommend shifting the LLR focus (at least temporarily) to that range once the current server loads are done, until 490K-495K reaches a better sieve depth.

Oddball 2011-06-04 20:18

[QUOTE=mdettweiler;262997]Done. :smile:

It seems that with this, we have loaded the last remaining work for k<100K, n=480K-490K. However, as of right now, the 490K-495K range is hardly well-sieved; we're not quite even at 1P yet. So it would be rather inefficient to continue on to testing n>490K until sieving has progressed further. In the meantime, we do have plenty of well-sieved work in the k=100K-200K range; everything from n=484001-489999 is essentially done sieving and ready to go. I would recommend shifting the LLR focus (at least temporarily) to that range once the current server loads are done, until 490K-495K reaches a better sieve depth.[/QUOTE]
Sounds good!

You can also remove n=481K-483K (which was done quite a while ago) from the stats page here: [URL]http://noprimeleftbehind.net:12000/all.html[/URL]

That should make the page load a bit faster.

Oddball 2011-06-08 04:23

Max, could you transfer n=487100-487500 from LLRnet to PRPnet? The PRPnet server is going to dry out in the next 2-3 hours, but the LLRnet server doesn't have much activity on it now.

mdettweiler 2011-06-08 06:15

[QUOTE=Oddball;263247]Max, could you transfer n=487100-487500 from LLRnet to PRPnet? The PRPnet server is going to dry out in the next 2-3 hours, but the LLRnet server doesn't have much activity on it now.[/QUOTE]
Uh...you sure about those figures? By my reckoning (based on the ~3.5 days it took to cover 487.5K-489K), we have another 2.3 days left with the currently loaded work.

Anyway, though, no problem--at the rate we're going, we'll need the extra work in PRPnet anyway soon. :smile: I've moved n=487101-487500 from LLRnet to PRPnet. (Note that this is slightly different than the range you specified; I started it at 487101 instead of 487100 to maintain consistency with how I've generally been doing everything else.)

Oddball 2011-06-08 06:55

[QUOTE=mdettweiler;263250]Uh...you sure about those figures? By my reckoning (based on the ~3.5 days it took to cover 487.5K-489K), we have another 2.3 days left with the currently loaded work.
[/QUOTE]
Yup, I'm sure. Remember, n=489K-490K had already been done before.

If you didn't load the n=487101-487500 work, we would have been out of k/n pairs now. The "count untested" column now says that 12,667 candidates haven't been tested, and all but one has n's from 487101-487500.

mdettweiler 2011-06-08 16:35

[QUOTE=Oddball;263253]Yup, I'm sure. Remember, n=489K-490K had already been done before.

If you didn't load the n=487101-487500 work, we would have been out of k/n pairs now. The "count untested" column now says that 12,667 candidates haven't been tested, and all but one has n's from 487101-487500.[/QUOTE]
Ah, of course--I forgot that 489K-490K was already done. :rolleyes: That makes all the difference. :smile:

Oddball 2011-06-10 04:16

Max, could you load k=100K-200K, n=484K-490K into the PRPnet server and k<100K, n=490K-491K into the LLRnet server? Both servers have dried out, and the files should be posted by gribozavr in this thread soon:
[URL]http://www.mersenneforum.org/showthread.php?t=12260[/URL]

mdettweiler 2011-06-11 16:55

[QUOTE=Oddball;263444]Max, could you load k=100K-200K, n=484K-490K into the PRPnet server and k<100K, n=490K-491K into the LLRnet server? Both servers have dried out, and the files should be posted by gribozavr in this thread soon:
[URL]http://www.mersenneforum.org/showthread.php?t=12260[/URL][/QUOTE]
I don't see the files yet...am I missing something, or have they not actually been posted yet?

Oddball 2011-06-11 19:10

[QUOTE=mdettweiler;263559]I don't see the files yet...am I missing something, or have they not actually been posted yet?[/QUOTE]
They haven't been posted yet :sad:

Oddball 2011-06-14 02:24

[QUOTE=mdettweiler;263559]I don't see the files yet...am I missing something, or have they not actually been posted yet?[/QUOTE]
It's posted now. See here:
[URL]http://www.mersenneforum.org/showpost.php?p=263700&postcount=5[/URL]

mdettweiler 2011-06-14 13:36

[QUOTE=Oddball;263731]It's posted now. See here:
[URL]http://www.mersenneforum.org/showpost.php?p=263700&postcount=5[/URL][/QUOTE]
All righty--the files are loading right now. :smile: LLRnet is all ready to go, and 484K-485K has been loaded into PRPnet; 485K-490K is still loading (the file is rather big and takes a while to upload), but in the meantime the server should have work available from 484K-485K.

Oddball 2011-07-06 05:42

I checked the [URL]http://noprimeleftbehind.net/tps/[/URL] page and found something weird. It said,

[code]
Results submitted today by:
sm5ymt : 17774
odicin : 31
: 1
Total results submitted today: 17806
[/code]

Does anyone know what happened with that ": 1" result? :huh:

mdettweiler 2011-07-06 06:21

[QUOTE=Oddball;265632]I checked the [URL]http://noprimeleftbehind.net/tps/[/URL] page and found something weird. It said,

[code]
Results submitted today by:
sm5ymt : 17774
odicin : 31
: 1
Total results submitted today: 17806
[/code]

Does anyone know what happened with that ": 1" result? :huh:[/QUOTE]
The script that converts the results from PRPnet to LLRnet format sometimes barfs on slightly-malformed results lines. Since the (separate) script that does the daily stats reads the LLRnet format results, it is affected by this as well. The problem is here:
[code]
[2011-07-05 06:10:03 CDT] 161631*2^484489-1 received by Email: ****@**** User: sm5ymt Client: 161 Program: llr 27A376A2E3F90409
[2011-07-05 06:10:05 CDT] 128367*2^484490-1 received by Email: ****@**** User: sm5ymt Client: 60 Program: llr.exe 4417C4BA0584AB32
[2011-07-05 06:10:08 CDT] 151881*2^484489-1 received by Email: ****@**** User: sm5ymt Client: 161 Program: llr 5843637031083387
[2011-07-05 06:10:08] 155895*2^484489-1 received by Email: ****@**** User: sm5ymt Client: 161 Program: llr D0F1A5278A5D8425
[2011-07-05 06:10:10 CDT] 119667*2^484488-1 received by Email: ****@**** User: sm5ymt Client: 47 Program: llr AA5336CC0DB4C6CD
[2011-07-05 06:10:11 CDT] 134655*2^484490-1 received by Email: ****@**** User: sm5ymt Client: 60 Program: llr.exe BCC14265341684CD
[2011-07-05 06:10:25 CDT] 130917*2^484490-1 received by Email: ****@**** User: sm5ymt Client: 160 Program: llr AAC07F03BDB0CB77
[2011-07-05 06:10:28 CDT] 145845*2^484489-1 received by Email: ****@**** User: sm5ymt Client: 103 Program: llr F6579543046AA5E6
[/code]
As you may notice, one of the lines is missing the "CDT" at the end of the timestamp, which was enough to throw off the conversion script, resulting in this LLRnet-format output:
[code]
user=sm5ymt
[2011-07-05 06:10:05]
128367*2^484490-1 is not prime. Res64: 4417C4BA0584AB32 Time : 0.0 sec.
user=sm5ymt
[2011-07-05 06:10:08]
151881*2^484489-1 is not prime. Res64: 5843637031083387 Time : 0.0 sec.
user=
[2011-07-05 06:10:08]]
is not prime. Res64: Time : 0.0 sec.
user=sm5ymt
[2011-07-05 06:10:10]
119667*2^484488-1 is not prime. Res64: AA5336CC0DB4C6CD Time : 0.0 sec.
user=sm5ymt
[2011-07-05 06:10:11]
134655*2^484490-1 is not prime. Res64: BCC14265341684CD Time : 0.0 sec.
[/code]
This particular error tends to happen every once in a while; I think it's occurred only once or twice in the past here at TPS, but it's shown up a number of times at NPLB. (There, it manifests itself as a result with a blank username in the hourly stats tables.) So far I've been fixing them all manually as they come up; I'll do that shortly with this one.

I really should add some code to the conversion script to handle these errors properly...shouldn't be too hard. :smile:

lsoule 2011-08-25 15:48

Are the binaries or sources available for the server part of llrnet?

thanks,
Larry

rogue 2011-08-25 16:26

[QUOTE=lsoule;270081]Are the binaries or sources available for the server part of llrnet?[/QUOTE]

Out of curiosity, why? Does llrnet provide any functionality that is not already covered by PRPNet? Is it in some way easier to set up and use than PRPNet?

lsoule 2011-08-25 16:37

I was interested in distributing LLR over a bunch of local machines. I guess I never looked at PRPnet due to the name - does it also distribute LLR?

lsoule 2011-08-25 16:40

That is, running my own server locally to distribute LLR over a bunch of local machines...

amphoria 2011-08-25 16:56

[QUOTE=lsoule;270085]That is, running my own server locally to distribute LLR over a bunch of local machines...[/QUOTE]

It does. On windows it uses cllr.exe the command line version of LLR. I have been using it to distribute work over local machines for about 2 years now. The server is easy to set up once you have installed mysql.

rogue 2011-08-25 16:59

[QUOTE=lsoule;270085]That is, running my own server locally to distribute LLR over a bunch of local machines...[/QUOTE]

Yes, that and much more. There are a number of differences between the two, including:

1) The client program (prpclient) and llr program (cllr) are not linked together. This allows you to replace one or the other allowing you to upgrade more easily. So if Jean upgrades llr or I upgrade PRPNet, you can drop in the new executables without waiting for someone else to integrate the two together.
2) Lots of stats available via HTML. You can see CRUS stats here: [url]http://noprimeleftbehind.net:1400/all.html[/url]
3) I'm continually making enhancements and bug fixes. In other words I keep it actively developed.
4) Used by many more projects, both great and small. PrimeGrid has more than 10 projects, some of which have had hundreds of users. Many people on this forum (including myself) run PRPNet servers at home.
5) Support for many project types. Beyond the Sierpinski/Riesel searches (offshoots of CRUS), it can handle simple k*b^n+c searches and Twin Prime Searches. I'm open to adding new searches as people make requests.
6) Support for many client programs. It supports LLR (for k*b^n+c), genefer, PFGW, and phrot.
7) Support for many OSes and CPUs. It runs on x86 and ppc across all versions of Windows, Mac, and Linux.
8) The server can run on MySQL or PostgreSQL databases, both of which are free and easy to install and easy to use (presuming you know SQL).

I could go on, but that gives you an idea of what it can do above and beyond llrnet.

lsoule 2011-08-25 17:24

Great - thanks for the reply. I'm on Ubuntu and it looks like MySql is easy to install there. I'll take a look!

-Larry

rogue 2011-08-25 17:47

[QUOTE=lsoule;270090]Great - thanks for the reply. I'm on Ubuntu and it looks like MySql is easy to install there. I'll take a look![/QUOTE]

If you run into issues, there are a couple of Ubuntu users on this forum who are running PRPNet servers. If you have any questions about PRPNet itself, please let me know.

lsoule 2011-08-25 18:26

So far so good. Is there a readme for the prpserver.ini? I'd like to search for primes of the form k*2^n-1 and not sure of the server type. Also, is there an example mysql.ini as well?

thanks,
Larry

rogue 2011-08-25 18:51

[QUOTE=lsoule;270096]So far so good. Is there a readme for the prpserver.ini? I'd like to search for primes of the form k*2^n-1 and not sure of the server type. Also, is there an example mysql.ini as well?[/QUOTE]

The prpserver.ini has a lot of comments regarding the options. You will most likely use one of these server types:

// 3 - Fixed b, k, and c, grouped by n
// 4 - Fixed b, n, and c, grouped by k

Which you use depends upon how you sieved. If you used a fixed k sieve, then use servertype 3. If you used a fixed n sieve use servertype 4. The servertype impacts a number of things. For example it affects how stats are stored and how HTML is generated. It also impacts the client as the client will use it to determine which software to use for the PRP test.

If the documentation references mysql.ini, then it is incorrect. I renamed the ini file to database.ini when I introduced PostgreSQL support. You will find that file in the server directory.

mdettweiler 2011-08-26 03:57

[QUOTE=lsoule;270081]Are the binaries or sources available for the server part of llrnet?

thanks,
Larry[/QUOTE]
FYI--though your ensuing discussion with Mark regarding PRPnet has somewhat obviated this request, we do nonetheless have a copy of the LLRnet source code (both client and server) posted on the noprimeleftbehind.net website for posterity. You can download it from:

[url]http://www.noprimeleftbehind.net/downloads/llrnet-sources.tgz[/url]

That said, I would definitely agree that PRPnet is a better choice these days. LLRnet does still work (in fact we use it alongside PRPnet at NPLB, with an updated client we developed that lets it use the latest LLR in a similar manner to PRPnet), though it is much less robust and full-featured. LLRnet does have the advantage of being very simple to set up (no database, just a straight-up NewPGen format text file for input); but once you get the hang of PRPnet, it's pretty easy to keep up. Personally, I find the server-end administration process much easier with PRPnet.

Anyway, just my $0.02... :smile:

odicin 2011-09-17 09:58

Is it possible, to update the prpnet-server from 4.1.4 to an newer version after the rally? During the race (and also before), sometimes the problem occurs, that the server is send a wu to my client but the client never reach them. Every time I checked this with the candidate in the prpclient.log and test_result.log on the client. The candidate never reached them. Thats why a pending zomie wu is now haunting 72h at the server (at pending tests). I try to remember, that this was a bug in this version of prpnet, besides some other like twisting client/team name.

Regards Odi

mdettweiler 2011-09-17 19:35

[QUOTE=odicin;271895]Is it possible, to update the prpnet-server from 4.1.4 to an newer version after the rally? During the race (and also before), sometimes the problem occurs, that the server is send a wu to my client but the client never reach them. Every time I checked this with the candidate in the prpclient.log and test_result.log on the client. The candidate never reached them. Thats why a pending zomie wu is now haunting 72h at the server (at pending tests). I try to remember, that this was a bug in this version of prpnet, besides some other like twisting client/team name.

Regards Odi[/QUOTE]
Yep, it's on my to-do list. :smile: I have hesitated to upgrade many of the NPLB/CRUS/TPS PRPnet servers from 4.1.4 since it has worked very well on the whole and I have been severely cramped for time; but it is definitely on my list. We've had CRUS's port 1400 running on the latest 4.3.5 for a couple of months now and it's proven to be quite stable.

odicin 2011-09-18 02:41

[QUOTE=mdettweiler;271916]running on the latest 4.3.5 for a couple of months now and it's proven to be quite stable.[/QUOTE]

And also some nice features like expiration time on pending list...

Regards Odi

mdettweiler 2011-09-18 06:37

[QUOTE=odicin;271930]And also some nice features like expiration time on pending list...

Regards Odi[/QUOTE]
Indeed, yes. I'm definitely planning to upgrade...just haven't gotten the time yet. (And yes, you all are allowed to bug me about it...doesn't hurt to have a little reminder "oh yes, and I still need to do that". :smile:)

Oddball 2011-09-19 06:48

Max, could you transfer n=490150-491K from LLRNet to PRPNet so port 12000 won't dry out before the end of the rally? There has hardly been any activity on LLRNet since the start of the rally, so drying out the LLRNet server isn't much of a concern.

mdettweiler 2011-09-19 13:34

Done. However, as I'm sure you realize, this isn't going to last us too long--maybe a half-day at the current processing rates; and we have two more days left in the rally. I'm assuming you're working on getting hold of gribozavr so we can refill on sieved-file material soon?

Oddball 2011-09-19 20:27

[QUOTE=mdettweiler;272040]Done. However, as I'm sure you realize, this isn't going to last us too long--maybe a half-day at the current processing rates; and we have two more days left in the rally. I'm assuming you're working on getting hold of gribozavr so we can refill on sieved-file material soon?[/QUOTE]
I PMed gribozavr and posted a message on the sieve thread a few days ago, but he hasn't responded. I'll send him another PM today.

If we don't get a new sieve file, we could make one on our own by taking the latest sieve file, removing the factors ourselves, and temporarily getting rid of k>100K. Of course, using this approach would miss gribozavr's factors (and possibly a few others), but it's better than nothing.

Lennart 2011-09-21 01:38

PRPNet is empty !! :smile:

mdettweiler 2011-09-21 01:54

Okay, I've made a temporary master sieve file for k<100k, n=491K-495K by taking the latest file posted in the sieving thread (28 May 2011) and paring it down to the desired k and n range with a script I have from over at NPLB. For reference, I've put the file up for download at:

[url]http://www.sendspace.com/file/crtab3[/url]

Meanwhile, I'll load n=491000-492500 into the PRPnet server momentarily. That should hold us over until the end of the rally, and hopefully until gribozavr can produce a more up-to-date file.

(Edit: Sorry that this is taking so long to get loaded...it's the top of the hour right now, so the NPLB hourly database refresh is running and MySQL is a little sluggish at getting the candidates in.)

(Edit #2: Actually, never mind that--I forgot, the server goes ahead and starts handing candidates out even as they're still being loaded, so we're good to go right now. :smile:)

Oddball 2011-10-02 18:18

Max, could you replace n=483800-484999 in the PRPNet server with these candidates:
[URL]http://www.sendspace.com/file/03u6od[/URL]

That's the latest sieve file, which is optimally sieved to p=6200T.

Also, it would be nice if you would remove the finished n=481K-493K range and load n=495K-500K into port 12000. k<100K can be extracted from gribozavr's new file here:
[URL]http://www.sendspace.com/file/739vcd[/URL]

mdettweiler 2011-10-05 01:32

[QUOTE=Oddball;273229]Max, could you replace n=483800-484999 in the PRPNet server with these candidates:
[URL]http://www.sendspace.com/file/03u6od[/URL]

That's the latest sieve file, which is optimally sieved to p=6200T.

Also, it would be nice if you would remove the finished n=481K-493K range and load n=495K-500K into port 12000. k<100K can be extracted from gribozavr's new file here:
[URL]http://www.sendspace.com/file/739vcd[/URL][/QUOTE]
Okay, I have the new candidates loading right now. My apologies for not completing this in a very timely manner--I was swamped with work and didn't get the chance until now. (Incidentally, we passed n=483800 before I got the chance to update the sieve file; thus I split it off at n=483900 instead.)


All times are UTC. The time now is 16:30.

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