Message boards :
Number crunching :
Shorties estimate up from three minutes to six hours after today's outage!
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 . . . 9 · Next
Author | Message |
---|---|
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
In switching a machines from a x64 OS to a x86 OS this morning. It would seem I left the <platform> tags out of the app_info.xml so it dumped all of the work. It is getting it all agian as resends. Only the MB work has extreamly large estimated run times. The AP tasks are estimated at 16 hrs which seems to be in line with the normal run time for the machine. On the plus side the downloads are finished very quickly instead of sticking liek they have been recently. This will probably help everyone finish up their pending downloads and let the d/l servers rest a bit. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
W-K 666 Send message Joined: 18 May 99 Posts: 19062 Credit: 40,757,560 RAC: 67 |
I got a few WU's from 25jn11ab that estimated at 133 hours on my CPU. Final times were around 3 hours. That is an unusually long time for even a VLAR on my PC. The estimates for all AR's were done a long time ago. See Estimates and deadlines This if I remember rightly was just after the Enhanced app was introduced, when there was only one receiver. But the same applies to MB (7 receivers with horizontal and vertical beams) which reduced the estimates to about 70% (probably it is 1/sqrt2%) of the enhanced estimates. |
kittyman Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004 |
I can already see the fubar/unfubar process starting on my top rig. I have a bunch of GPU work still in cache, so it works on them normally until the DCF or additional downloaded work with the fake completion times triggers it into doing tasks just downloaded. Then it crunches up a batch of the brand new work, readjusts the DCF, and goes back to normal crunching. It's kinda fun watching a task with an hour twenty estimated completion time zip through in a little over 2 minutes. I can only imagine what Boinc is thinking........ "Freedom is just Chaos, with better lighting." Alan Dean Foster |
perryjay Send message Joined: 20 Aug 02 Posts: 3377 Credit: 20,676,751 RAC: 0 |
Yeah, I was just gonna say it's fun watching time drop three or four minutes every second. :-) PROUD MEMBER OF Team Starfire World BOINC |
kittyman Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004 |
Yeah, I was just gonna say it's fun watching time drop three or four minutes every second. :-) I just hope Boinc figures it out before the ol' cache gets too thin. "Freedom is just Chaos, with better lighting." Alan Dean Foster |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13736 Credit: 208,696,464 RAC: 304 |
I notice that the new increased runtime WUs are maked as 6.09 (cuda23) whereas before they were all 6.10 (cuda_fermi). EDIT- *scratches head* At least that's the case on one machine, on the other they're still 6.10 (cuda_fermi). Will have a closer look later on when i'm awake. Grant Darwin NT |
MusicGod Send message Joined: 7 Dec 02 Posts: 97 Credit: 24,782,870 RAC: 0 |
....it must be an Einstein thingy.LOL |
soft^spirit Send message Joined: 18 May 99 Posts: 6497 Credit: 34,134,168 RAC: 0 |
In a strange way, this has worked out great.. It has cleared the backlog(at least temporarily) where I can actually GET some gpu work!! Janice |
Dimly Lit Lightbulb 😀 Send message Joined: 30 Aug 08 Posts: 15399 Credit: 7,423,413 RAC: 1 |
I've got an astropulse with a 175hr estimate and a multibeam with a 15hr estimate. A few more of those and there goes my cache :(. |
soft^spirit Send message Joined: 18 May 99 Posts: 6497 Credit: 34,134,168 RAC: 0 |
I've got an astropulse with a 175hr estimate and a multibeam with a 15hr estimate. A few more of those and there goes my cache :(. Cache? Still trying to get enough to keep my gpu wet. let alone stocked! Janice |
Dave Stegner Send message Joined: 20 Oct 04 Posts: 540 Credit: 65,583,328 RAC: 27 |
Cache? My AP jumped from ~17 hours to almost 9 days. That means where I used to have ~22 to 24 tasks I will now have 1 or 2. Not very promising. Dave |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
Cache? When did you guys down these AP tasks? The ones I got before 14:00 UTC had my normal estimates. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
Dave Stegner Send message Joined: 20 Oct 04 Posts: 540 Credit: 65,583,328 RAC: 27 |
Recent, today. Dave |
Dimly Lit Lightbulb 😀 Send message Joined: 30 Aug 08 Posts: 15399 Credit: 7,423,413 RAC: 1 |
I got the astropulse with the whacky time estimate 14 Sep 2011 | 9:49:30 UTC. The astropulse before it with a normal estimate I got 10 Sep 2011 | 11:49:59 UTC. |
Sutaru Tsureku Send message Joined: 6 Apr 07 Posts: 7105 Credit: 147,663,825 RAC: 5 |
(...) Dave was informed that the new changeset destroyed the well working estimated times of app_info.xml users and soon he will release a fix? - Best regards! - Sutaru Tsureku, team seti.international founder. - Optimize your PC for higher RAC. - SETI@home needs your help. - |
Dave Stegner Send message Joined: 20 Oct 04 Posts: 540 Credit: 65,583,328 RAC: 27 |
That is great news. Let's hope he fixes it sooner than later, before more machines get more screwed up. Dave |
Sutaru Tsureku Send message Joined: 6 Apr 07 Posts: 7105 Credit: 147,663,825 RAC: 5 |
No, this was a question of me.. ;-) BTW, I'm confused that this new changeset wasn't tested before at S@h Beta Test. [EDIT: Sorry, english is not my mother language. I had for a long time only a few years in school.. ;-)] - Best regards! - Sutaru Tsureku, team seti.international founder. - Optimize your PC for higher RAC. - SETI@home needs your help. - |
kittyman Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004 |
Well, I am seeing a bit of a chink in the Cricket graph already..... Looks like the bogus completion estimates may be fooling most of the Boincs on the project that their computers don't need any more work. "Freedom is just Chaos, with better lighting." Alan Dean Foster |
Blake Bonkofsky Send message Joined: 29 Dec 99 Posts: 617 Credit: 46,383,149 RAC: 0 |
Does anyone know if these estimates are going to fix themselves? Is this going to be like when we first start running an app and have to wait for 10 validations before times fix? |
kittyman Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004 |
Does anyone know if these estimates are going to fix themselves? Is this going to be like when we first start running an app and have to wait for 10 validations before times fix? Joe's post (the 2nd one in this thread) seems to indicate that Boinc will self adjust for it eventually, but I think the adjustment will be in the local Boinc installation, not server side. Which to me indicates that things will be screwed totally for any rig which still has work cached before the change was put into place. And then, it will be a slower process of adjustment. Not simply completing and reporting 10 valiated WUs. That's my take on it. I suspect that Joe will chime in again if my understanding of it is wrong. "Freedom is just Chaos, with better lighting." Alan Dean Foster |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.