Discussion in 'World Community Grid (WCG)' started by Chicken Patty, Feb 20, 2009.
What's with all the high priority?
And haven't seen any of that on my machines.
Cool, I haven't seen anything like that!
Let us know how those do!
All of the GPU WUs I get say "Running, High Priority", but I rarely get that on the CPU WUs
I get high priority jobs a lot- the priority is related to the due date on the wu and other running jobs get suspended until the high priority jobs are finished.
I'm not sure if WCG sends them out to machines to get finished on time or if I already have them downloaded and they jump up in priority???
I'm pretty sure that WCG sends them out with a shorter deadline. I've seen some WUs get downloaded and instantly start running, even when the computer then has to pause some of the ones it's working on. I don't really understand why it does this, but I don't really mind. Even most of the High Priority ones still have a 7-day deadline (except for the GPU ones).
EDIT: Wooh, I got my 5th star
Except that the deadlines show as the 15th. How's that urgent?
(At least I assume it's the deadline column.)
That would the ~week deadline I was talking about. I dunno if BOINC is just in the head, but it seems to think that is urgent (consistently).
Running WCG suggested BOINC version? (6.10.58 or whatever it is?)
Remind reading somewhere that newer version were better with prioritizing, but aren't entirely sure. :S
That said, the beta had my laptop in total chaos, jumping all over the place with high priority.
(And that's what I get for skimming comments. Didn't see your mention of week away urgent Ion. )
I also get high priority jobs from time to time. Look at the dates of the other jobs. They will be later then that job. What happens atleast on my end is that the lets say today's date is the firstand I would have jobs due on lets say the tenth and one new job will come in with a date of the fifth so it will push that to the top and ran as high priority. to make sure it gets done before the later jobs.
WCG makes the call... maybe it's to fill a quorum on a job and the other wu's are closer to the deadline.
WCG is just in the head
Its similar to F@H deadlines that are about a week. You have to figure the majority of machines are not 24/7 dedicated and probably are set to run only when the computer is idle (idle time is really the bread and butter of donations, I imagine). So they project maybe 30-90min of run time a day and figure that should take about a week.
I've never seen a high priority outside betas.
And currently have a WU slated for the 10th happily waiting in line.
Also, are those estimates even close to correct?
I get high priority WUs across nearly all projects on a regular basis.
Hey guys, sorry if my PPD is down a bit for a while. I'm working on my P55 system again and it ALWAYS goes offline right when I leave work. Seriously, I even stayed at work late and watched it working fine, then I get home and it's offline. At least I am making some greater headway with overclocking it, but as you can see it is still rebelling at the worst times.
Unrelated to WCG, but I also need to move a pair of GTX 460s off of my A8N32-SLI Deluxe because the CMOS loses the settings on the slightest power loss and it's like the battery is useless. It also won't boot with dual channel RAM set above 266MHz, but single channel in any adjacent slots works fine. Yes, I tried a new jumper cap and bending the pins but that didn't help, so I'm chalking it up to ASUS A8N series design flaws as the ECS board from the same time works flawlessly, but the 2 ASUS boards I have either eat batteries or refuse to use them.
Long story short, I can't keep having the thing reset to year 2002 or be unreliable in general considering it houses 30% of my GPU power, as well as my P55 system (although the hardware seems to work fine there, but it is a b***h to actually get stable). Hopefully I get called in tomorrow so I can go add more volts to the P55 system as I'm tired of it being offline all the time.
Damn man that sucks. I though that one board was the only one to have battery issues. Hell My A8n-Sli premium is working flawlessly right now.
I don't know if I should take a chance and buy a new EEPROM from the ASUS eStore. I love the board considering it is loaded with features like dual GBE NICs and has an 8 phase VRM, but I can't stand it when the wind blows wrong or something and it loses all its settings again.
I'm going to find some datasheets on 8 pin CMOS microcontrollers and see what the voltage input and ground pins are (I believe they are pins 1 and 8). Maybe I can test those pins on the board and see if there is any voltage output to determine if the board has a bad trace or the CMOS chip is bad. I should also check the part numbers on both and see if they happen to be the same one, as maybe they both have a design flaw.
Have you tried simply replacing the CMOS battery? You can find them anywhere. Those are CR2032 batteries, walmart should have them.
I didn't bother uninstalling it at all. It was on my storage drive anyway, so I just thought I'd run right over it with the new install and it hit the ground running.
Yep, the A8N32-SLI has had several brand new batteries installed and they still read as fresh on my Micronta battery tester that applies 1 mA load, yet CMOS always gets wiped even if the PSU is not supplying current for less than 5 seconds.
The A8N-E on the other hand eats batteries once a week and can't shut itself off, so I wired the battery holder to the power LED header and it runs like that.
I uninstall it before I sell the system--and it makes it easier to transfer it to a new system
Oh the A8n-e when I was investigating the battery issue the problem with it was leading to the cmos chip draining it down. But on your other one I believe you are on the right track with narrowing it down by seeing if the power is even getting to the chip. Considering the batteries aren't drained that either like you said the power isn't getting to the chip due to a bad trace or the cmos chip could be bad. I don't think it's a a design flaw because the 2 A8n boards I have here don't suffer from any problems with batteries or loss of cmos settings. One is the A8n-Sli premium and the other board is a A8n-vm csm.
may get my hands on a cheap i5 2400S 65W
will prolly put it in my primary, then put the 3570K in the sabertooth i got laying here for dedicated
KieX has delayed shipping out the 2700k combo until Tuesday due to when FedEX picks up, but I should have it within a week and a half from then
It appears I'm still running 9 wu's at a time on my 8 core FX-8150 rig
I checked WCG forum and they are claiming it's an error but I'm doubting this as I'm getting credit for the wu and I have more waiting?
The WCG forum thread discussing it:
I replied on that thread a few minutes ago....
What is the identifier that is separating that wu from the other Gfam unit's? Is it the tetramer part? Just wondering so I can cheak my rigs and see if any are on those and see if they are running them different then you rig.
The "tetramer" is what I'm seeing that is different in the wu #
I got one of those in the waiting area on this system and the other 1055t and my 555 is running 2 at the moment while running 2 other wu's. I am not getting the extra wu running as you are. I will check the other 3 systems later when I do my "rounds" and see what going on with them.
Here is a screenie from the 555.
Edit: I see where yours is also running on the gpu.
And I just received a beta for the gpu.
Separate names with a comma.