[QUOTE=EdH;283955]This is, of course, very rough, since I was doing other things for part of these runs and they are but a single test rather than an average of several. In fact, it is interesting that a 3000 digit certificate took less time than a 2900 digit one...[/QUOTE]But you really should average ~10 jobs or so, since some numbers might be "harder" to prove than others. An interesting metric might also be the size of the cert files. If a number took longer, more intermediate numbers would have been needed to be proved, so the certs should be (significantly?) larger....

[QUOTE=schickel;283970]But you really should average ~10 jobs or so, since some numbers might be "harder" to prove than others. An interesting metric might also be the size of the cert files. If a number took longer, more intermediate numbers would have been needed to be proved, so the certs should be (significantly?) larger....[/QUOTE]
Agreed I should do more, but for now, I'm on to other things. Here is a list which includes certificate sizes:
[code]
1500 - 945s - 293,423 bytes
1600 - 995s - 278,731 bytes
1700 - 1299s - 324,653 bytes
1800 - 1508s - 363,539 bytes
1900 - 2239s - 429,720 bytes
2000 - 2840s - 459,723 bytes
2100 - 3559s - 450,986 bytes
2200 - 4750s - 571,942 bytes
2300 - 4599s - 599,175 bytes
2400 - 5746s - 642,316 bytes
2500 - 8429s - 670,196 bytes
2600 - 8549s - 786,621 bytes
2700 - 9629s - 812,852 bytes
2800 - 12341s - 931,681 bytes
2900 - 15357s - 924,020 bytes
3000 - 13779s - 1,000,670 bytes
[/code]Not a direct correlation between time and certificate size...

Now, the db won't accept certificates from me. It times out and does a connection reset. I think I will give the certificate part of the db a rest and stop running them for a while...

[QUOTE=EdH;284318]Now, the db won't accept certificates from me. It times out and does a connection reset. I think I will give the certificate part of the db a rest and stop running them for a while...[/QUOTE]Actually, it looks like the whole site is not responding. The server is up, but only serving up blank pages....

[EDIT: Never mind--it's back up!]

I uploaded the rest of my certificates this morning, but I have my two certificate machines doing something else for now...

[QUOTE=akruppa;283933]Chalk me up as one of those who think the DB needs some kind of authentication to submit content. It's too valuable a tool to leave it at the mercy of everyone who knows of its existence not to crap-flood it, and the honor system tends to be short-lived on the Internet.[/QUOTE]Hmmm, that may be in the works.

There's a new line added to the status [URL="http://factordb.com/status.html"]page[/URL]: [B]N This number is not in database (and was not added due to your settings)[/B]

Congrats on new db hardware!

HEEEEEEELP!
 

Why on Earth did you put a 200 CPU-second limit on there? Remove the limit, make it bigger, or I won't ever use the DB again! I can't even get the status page to come up! I want more CPU time per Hour! Please. I can't use it like this! Raise it to at least 1000 seconds because 200 is not enough! I can't even get M42643801 to come up without it saying "you have reached 200 CPU-second limit". And please fix it so it lets me put M43112609 on it. It won't be fully up to date until it includes the largest known prime.

Anybody noticed the 'Show node graph' for Aliquot sequences yet?

Good work, Syd!

Also, Please fix the "Maximum of 2 parallel processing requests".

[QUOTE=kar_bon;287569]Anybody noticed the 'Show node graph' for Aliquot sequences yet?

Good work, Syd![/QUOTE]

well that'll save a lot of time especially if it uses the type of code the project uses. now to use them to make the trees.