CRUS LLRnet server discussion

[em99010pepe]

Originally Posted by Anonymous

Okay, then: Did anyone send in any results for the Sierp. server yet? If not, then would you please shut it down for the time being? That way, we can hold off until we can start it on port 443.

Not yet...Port 444 down but we need to think to use another port for Riesel server. I can easily run 10 LLRNet servers.....

[mdettweiler]

Originally Posted by em99010pepe

Not yet...Port 444 down but we need to think to use another port for Riesel server. I can easily run 10 LLRNet servers.....

Okay, good. Once the Riesel server has run dry, we can switch it to any other port easily.

We'll always want to have at least one server on port 443 or 80--thus, users behind corporate firewalls can access at least one of the servers. If you have port 80 available, that would be perfect--we could put the Riesel server there, and Sierpinski on 443. (Since nobody's touched the Sierp. server yet, we could easily switch it to 443 once the Riesel server isn't using that port any more.)

As for being able to run 10 LLRnet servers: well, I don't think we'll need that many, but that sure keeps our options open! Maybe for a future No Prime Left Behind server (let's hold off on that one until we've gotten the CRUS ones all set, though!).

[mdettweiler]

I've currently got one of my cores on my Core 2 Duo running on the LLRnet server (Riesel, of course, since that's the only one up right now). I'm going to move my other core over right after I post this, too, to help clean up the remaining candidates ASAP.

Carlos, roughly how much more do we have left of the 2nd Drive candidates--i.e. how much longer before the server starts handing out doublecheck tests?

Also, what time limit did you set the server to allot for clients to return tests?

[em99010pepe]

Soon the double check will start and the re-test of the rejected numbers...

[tnerual]

how many cores are you running as llrnet client ?

on my side, i run
- 2 cores of a dual athlon x2 4200
- 1 core of an athlon xp 2600 +
- 1 core of an P4 - 1600
- 1 core of a centrino-duo 1600

[mdettweiler]

Originally Posted by em99010pepe

Soon the double check will start and the re-test of the rejected numbers...

I've already got a few of the doublechecks showing up in my workfile.txt (set to cache 10 numbers).

[kar_bon]

1 core: P4 at 2GHz

now testing Riesel base 16 from n=100k (which i tested 2 weeks ago on Core2Quad)

[mdettweiler]

Originally Posted by tnerual

how many cores are you running as llrnet client ?

on my side, i run
- 2 cores of a dual athlon x2 4200
- 1 core of an athlon xp 2600 +
- 1 core of an P4 - 1600
- 1 core of a centrino-duo 1600

I've had one core of a Core 2 Duo running the llrnet client for the past 18 minutes; I just added the other core a few minutes ago.

[mdettweiler]

Originally Posted by kar_bon

1 core: P4 at 2GHz

now testing Riesel base 16 from n=100k (which i tested 2 weeks ago on Core2Quad)

From n=100K? Where'd you get those numbers? I didn't think the Riesel Base 16 sieve file was out yet for that high up. Did you mean Sierpinski Base 16?

[mdettweiler]

Well, it looks like the server has run out of new numbers to give out...Carlos, have you inserted the doublechecks that Karsten posted, too? (The ones that are all supposed to be prime)

[em99010pepe]

Originally Posted by Anonymous

Well, it looks like the server has run out of new numbers to give out...Carlos, have you inserted the doublechecks that Karsten posted, too? (The ones that are all supposed to be prime)

No....I'll do that in a minute...