Develop LLRnet version 3.7.1 for pay!

[mdettweiler]

Originally Posted by gd_barnes

<snip>
I know a remote desktop will help but not only do I not have it set up yet (hope to within a few weeks), I'm not sure I'd have much time to mess with my ranges while I'm out of town.

When you're ready to get the remote desktop thing set up just give me a holler, I can probably help with that.

I can see where the very heavy-hitters, and by that I mean much bigger than me, i.e. 50+ cores, are coming from. They want to run LLRnet for an extended period if necessary and not worry about it much. Otherwise, they'd be spending 4 hours+ per day determing what to reserve, loading files, concatenating on to the end of existing files, posting results files, etc.

And finally, what would it take to convert LLRnet to run version 3.7.1 of LLR?

Nobody's quite sure how much work it would take to make an LLRnet based on 3.7.1--a few people, including me, have tried simply dropping the 3.7.1 source code on top of the 3.5.0 code in the LLRnet source tree, but just got a bunch of complex error messages instead. My hunch is that it might be easier just to replace the code in LLRnet that makes it do the primality tests through a direct tie-in with the 3.5.0 code, with a simple command that runs a separate 3.7.1 executable in the same directory and passes it the number through the command line (using the -q switch, a new feature added in 3.7.1 for Linux and Windows command line versions). I would try it myself but that I know nothing about the Lua programming language (which LLRnet is written in). I also considered rewriting the entire thing, as a wrapper to an existing LLR 3.7.1 executable of course, in Perl, but I couldn't quite figure out how to deal with a raw network socket (so I could make it speak LLRnet protocol instead of just a "standard" protocol like HTTP).

I've tried bugging the people at RieselSieve (who originally invented/wrote LLRnet) to make a 3.7.1-based version, but the guy there who wrote LLRnet seems to have disappeared from their forum, and the admins themselves are usually tied up with running their BOINC setup and thus don't have time for making a new version of LLRnet.

If any of you programmer-types out there want to undertake the project, you have my hearty blessing. (Not that my blessing matters anyway...lol)

Anon

[gd_barnes]

Originally Posted by Anonymous

When you're ready to get the remote desktop thing set up just give me a holler, I can probably help with that.


Nobody's quite sure how much work it would take to make an LLRnet based on 3.7.1--a few people, including me, have tried simply dropping the 3.7.1 source code on top of the 3.5.0 code in the LLRnet source tree, but just got a bunch of complex error messages instead. My hunch is that it might be easier just to replace the code in LLRnet that makes it do the primality tests through a direct tie-in with the 3.5.0 code, with a simple command that runs a separate 3.7.1 executable in the same directory and passes it the number through the command line (using the -q switch, a new feature added in 3.7.1 for Linux and Windows command line versions). I would try it myself but that I know nothing about the Lua programming language (which LLRnet is written in). I also considered rewriting the entire thing, as a wrapper to an existing LLR 3.7.1 executable of course, in Perl, but I couldn't quite figure out how to deal with a raw network socket (so I could make it speak LLRnet protocol instead of just a "standard" protocol like HTTP).

I've tried bugging the people at RieselSieve (who originally invented/wrote LLRnet) to make a 3.7.1-based version, but the guy there who wrote LLRnet seems to have disappeared from their forum, and the admins themselves are usually tied up with running their BOINC setup and thus don't have time for making a new version of LLRnet.

If any of you programmer-types out there want to undertake the project, you have my hearty blessing. (Not that my blessing matters anyway...lol)

Anon

I would be willing to PAY someone to develop this due to its complexity. Either with money or with some LLR or sieving processing or a combination of both.

Anyone game? We can negotiate the price and how to do the payment by PM's or Emails. My Email is gbarnes017 at gmail dot com.

Edit: I'm making this a temporary sticky in a new thread for visibility. I'll unstick it in 3-4 days.


Gary

[em99010pepe]

You guys need to talk to Jean Penné and his son.
Jean told me he was busy with other tasks and so to be patient.

[gd_barnes]

Originally Posted by em99010pepe

You guys need to talk to Jean Penné and his son.
Jean told me he was busy with other tasks and so to be patient.

Jean has a son that is involved in prime searching? I didn't know that. I'll talk to Jean in the CRUS threads.

[em99010pepe]

Originally Posted by gd_barnes

Jean has a son that is involved in prime searching? I didn't know that. I'll talk to Jean in the CRUS threads.

Open client.lua and read the first lines...hehe

[em99010pepe]

Any news?

Just to note that every candidate tested under the current llrnet version is at least 10 % slower than tested on standard manual client (llr 3.7.1c).

[Tumo]

Originally Posted by em99010pepe

Any news?

Just to note that every candidate tested under the current llrnet version is at least 10 % slower than tested on standard manual client (llr 3.7.1c).

There is a rather simple fix for that in multiple ways and Jean's son can be of great help, reducing the development / verification time required.

Ty.