Any way to transfer WU's?

QuietIce
QuietIce
Joined: 9 Sep 06
Posts: 3
Credit: 1801717
RAC: 0
Topic 191887

I have 2 machines running Einstein - a 3-yr-old 2100 and a new Opteron. The older machine has a LOT of WU's on it and the new machine can't seem to get any new WU's. I'm assuming the lack of new WUs for the new machine is b/c of all the work backed up on the older machine.

So, is there any way to transfer those WU's from the old machine to the new one?

Pooh Bear 27
Pooh Bear 27
Joined: 20 Mar 05
Posts: 1376
Credit: 20312671
RAC: 0

Any way to transfer WU's?

Transfering units between computers is not possible the way you want to do it. The machine that downloads must upload the units.

As for the one getting a lot, the other getting few, I am thinking the old one is getting a lot of small units, the new one is getting a few large units? Since your computers are hidden, I can not see if there are other things going on, but I am betting this is what is going on.

You can tell if the unit is small or large by the number after the first 2 letters, and underscore. EG xx_0001.... If that number is 400 or more, it's a large unit, under 400 is a small unit.

If this is not what is going on, unhinding your computers helps us determine if something else might be going on.

QuietIce
QuietIce
Joined: 9 Sep 06
Posts: 3
Credit: 1801717
RAC: 0

Sorry it took so long to get

Sorry it took so long to get back. I didn't realize my computers were hidden, that's been changed.

Apparently it worked itself out one way or another. I was looking at 12-15 units and all of them were reading a 4 hour estimated CPU time with a deadline only a week away. This would have been impossible for the smaller machine to perform given my preferences. Either the CPU time estimate was adjusted or the server figured out the machine couldn't handle it in time or maybe it was late and the server gave it to someone else - who knows? Since the smaller machine is an extra I don't see it every day so I'm not sure what happened to resolve the issue. But everything looks fine, now - units enough to last the next few days as it should be.

On another note my daily computer (QuieIce) ran out of work yesterday and it's not downloading any new units. I've seen this happen before so I'm not worried about it - the machine will just crunch more SETI numbers instead :) - but it is kind of odd behavior.

Thanks for the help ...!->

Pooh Bear 27
Pooh Bear 27
Joined: 20 Mar 05
Posts: 1376
Credit: 20312671
RAC: 0

With shared projects, some

With shared projects, some faster machines get their debt a little in a weird back and forth type deal. I have one machine that does 3 projects, CPDN, SETI and Einstein. It always has CPDN (2 of them, 4 processors), and it almost always has just SETI or just Eisntein. BOINC calculates debt in an odd way on multiprocessor units, so it's how it goes. I do not think it's a big deal that it does this, and it happily is always crunching, so I do not worry about it.

As for the slower machine, it looks like it finished all the work. Time estimates are that, estimates. Because they were all smaller units, they got done in plenty of time, and you got your credit on every one of them. It looks like it took under 50 minutes for most of them to finish (with the 4 hour esitmate). I am betting that it has something to do with DCF, time online, etc. estimates.

Glad to see it all worked out, and that your machines continue to do good crunching.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.