Message boards :
Number crunching :
AP V7
Message board moderation
Previous · 1 . . . 16 · 17 · 18 · 19 · 20 · Next
Author | Message |
---|---|
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14653 Credit: 200,643,578 RAC: 874 |
Need to say last few dozens of posts made this thread absolutely unworthy for monitoring AP7 issues. So if someone has something to say (and wanna be heared, of course) please start new thread or post in APv7 issues & errors one. 26-Oct-2014 22:16:29 [SETI@home] Scheduler request completed: got 2 new tasks 26-Oct-2014 22:16:29 [SETI@home] [sched_op] estimated total CPU task duration: 7843 seconds 26-Oct-2014 22:16:29 [SETI@home] [sched_op] estimated total Intel GPU task duration: 10242 seconds 26-Oct-2014 22:16:31 [SETI@home] Started download of 16jn14aa.25013.686396.438086664199.12.195.vlar 26-Oct-2014 22:16:31 [SETI@home] Started download of ap_30oc11ab_B3_P1_00206_20141026_21229.wu If you want to find a discussion of AP v7 applications, you'll mostly find it in my installer thread :) |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
If GPU queue would be filled first You could do project reset and get CPU tasks resent to GPU. Nothing wrong with that and I think some people have done that purposefully. All you have to do is set your Preferences to GPU only, then do a Project Reset. All the Tasks will be reassigned to the GPU. There are other ways without doing a reset, but, the reset is the easiest. I'm supposing the reset still works that way, I haven't tried it in a while. In about another day my one Host will be very low and I'll give it a try. I'm going to leave the Preferences set for both first, then change it again if they just go to the CPU, as I suppose they will. APs Are Not MBs. Any one notice APv7s are going much faster than APv6? Since the APs stopped late on the 26th/early 27th I'm down to In progress (186). Over half gone already. I'll take as many Resends as possible no matter where they go. Like I said, I can Compensate...as long as I have something to Compensate with. It would be really nice if all the resends went to the GPUs to begin with, save a little work that way. I count 26 resends since the well went dry, with 2 going to the GPU, http://setiathome.berkeley.edu/results.php?hostid=6796479 Down to 183. |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
I was just sent another 6 AP resends that all went to the CPUs even though my 3 GPUs are getting really Low. I'm really starting to like Claggy's old suggestion about the Server sending tasks without any device restrictions. APs are APs, they are the same. There isn't any need to have device restrictions. The Manager should be allowed to just send tasks to the device that needs them. It would make things much less complicated. |
Claggy Send message Joined: 5 Jul 99 Posts: 4654 Credit: 47,537,079 RAC: 4 |
Until your GPU cache is full, moderate your cache size until Boinc no longer asks for CPU work, once you have some GPU work then increase it in steps. Claggy |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
Until your GPU cache is full, moderate your cache size until Boinc no longer asks for CPU work, once you have some GPU work then increase it in steps. I'm just about out of GPU tasks. Take a Good look at the results page. In a couple hours they will probably all turn into CPU tasks. Shortly thereafter most of them will go back to GPU tasks. Did anyone ever submit a reasonable reason for not just having Plain APs as you suggested a while back? I can't think of any reason to have device restrictions. |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
A new High! Last week I thought I had seen something strange when after the Outage the Completed Ready to Reports on my Computer were already showing up on the Results page on the web even though my computer said it hadn't contacted the Mothership in Hours. That was fixed by setting NNT and updating. This week I decided to test the Project reset. All 79 AP tasks were resent to the computer and listed as AstroPulse 701 CPU tasks and the Manager showed 3 running on the CPUs. But the Web said they were ATI tasks! I should have taken a picture. I did the only thing I could, I restarted BOINC. When it came back it listed them All as ATI tasks, running on the GPUs. You can't make this stuff up! I checked the slots and there isn't any indication they had been run on the CPUs. This has to be the strangest thing I've seen to date. Oh well, it's working anyway... |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
What do you mean??? There is no difference between CPU and GPU AP tasks at all. Not even some scheduler rules as for VLAR MB... |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
The VLARs are the only problem, since those are kept from the GPUs. What I mean is why do I have AstroPulse v7 Anonymous platform (ATI GPU) And AstroPulse v7 Anonymous platform (CPU). Why not just AstroPulse v7 Anonymous platform. Then the Manager wouldn't have to limit the tasks to certain devices. That way you wouldn't get stuck with 79 CPU tasks that would take over 10 days to complete on the CPUs or only 11 hours to complete on the GPUs. The tasks could be sent to the GPUs or CPUs depending on which device needed them. I'm pretty sure that's what Claggy was suggesting back in one of those old CreditFew threads. Hmmm, seems I have more CPU resends. |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
Ah you just want a general pool of tasks on the client end and then let the client process them on the hardware as it sees fit. Other than the issues that the BOINC devs would have to overcome for that to happen, of which I imagine there would be a few, it seems like a logical way to do things. I think several people have suggested BOINC work like this since GPU crunched started. How other projects would handle such a change is really the issue. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
qbit Send message Joined: 19 Sep 04 Posts: 630 Credit: 6,868,528 RAC: 0 |
Only APs on my GPU, nothing else. Occasionally vLHC on CPU, but I don't think that could make APs on my GPU run slower, could it?
Thx Hal & Mike, I will keep an eye on this! |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
Yes, seems if you could just have a pool of AP tasks that would solve the recently discussed problems. I'm sure there would be others. Speaking of which, New APs are being Split...just in time. I'm out of GPU tasks, I've already received a few New tasks, and they are All labeled CPU. Wed Oct 29 13:42:08 2014 | SETI@home | Sending scheduler request: To fetch work. Wed Oct 29 13:42:08 2014 | SETI@home | Reporting 1 completed tasks Wed Oct 29 13:42:08 2014 | SETI@home | Requesting new tasks for CPU and ATI Wed Oct 29 13:42:11 2014 | SETI@home | Scheduler request completed: got 4 new tasks Wed Oct 29 13:57:37 2014 | SETI@home | Sending scheduler request: To fetch work. Wed Oct 29 13:57:37 2014 | SETI@home | Reporting 1 completed tasks Wed Oct 29 13:57:37 2014 | SETI@home | Requesting new tasks for CPU and ATI Wed Oct 29 13:57:40 2014 | SETI@home | Scheduler request completed: got 1 new tasks Wed Oct 29 14:02:47 2014 | SETI@home | Sending scheduler request: To fetch work. Wed Oct 29 14:02:47 2014 | SETI@home | Reporting 1 completed tasks Wed Oct 29 14:02:47 2014 | SETI@home | Requesting new tasks for CPU and ATI Wed Oct 29 14:02:49 2014 | SETI@home | Scheduler request completed: got 2 new tasks Wed Oct 29 14:21:15 2014 | SETI@home | Sending scheduler request: To fetch work. Wed Oct 29 14:21:15 2014 | SETI@home | Requesting new tasks for CPU and ATI Wed Oct 29 14:21:17 2014 | SETI@home | Scheduler request completed: got 1 new tasks http://setiathome.berkeley.edu/results.php?hostid=6796479 |
Nigel Garvey Send message Joined: 18 May 99 Posts: 39 Credit: 2,228,452 RAC: 3 |
HAL9000 wrote: I have an old first gen Intel iMAC. I noticed it wasn't downloading and AP v7 work. I seem to recall hearing that 64-bit Intel was going to be a requirement for future releases? Looking at the list of apps I see "Mac OS/X 10.3+ 7.00" under AP v7. Is that an app for PPC and/or hardware with OS X.3? Seems to be for PPC anyway. My G5 (PPC, running OS X 10.5) downloaded it and a task this morning and seems to have got through the first hour OK. NG |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
HAL9000 wrote:I have an old first gen Intel iMAC. I noticed it wasn't downloading and AP v7 work. I seem to recall hearing that 64-bit Intel was going to be a requirement for future releases? Looking at the list of apps I see "Mac OS/X 10.3+ 7.00" under AP v7. Is that an app for PPC and/or hardware with OS X.3? I figured it would be for that. Good to know for sure. PPC & 64-bit Intel are much large groups than 32-bit Intel in MAC land. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
So I just noticed something new, at least to me. I had just readjusted my ffa values up to higher settings to test something with another user. I had left it there as my system has around 16 GB of Ram and I wasn't worried about those AP "huggers" previously mentioned. This is the AP unit ap_13jn10ab_B6_P1_00107_20141101_19799.wu_1 Anyway, I got one and was watching it for over the past hour. It progressed to 33 % complete and 1.7 GB of memory. What caught my eye was the CPU utilization went all the way up to 96% I continued to watch this and soon enough, the AP crashed and restarted. It did this 2 more times before I reset the values in the FFA to lower values. I switched it to the following -ffa_block 8192 -ffa_block_fetch 4096 Anyway, so Ram Usage is under 1 GB but CPU utilization continues to climb over 90% and when it hits some value in the upper 90% it crashes back and restarts again at around 2-3% progress. So it's done this 5 times total now, 3 times at the higher settings and 2 times at the lower now. Is this the same thing that Juan and Raistmer were talking about? Because I thought it was only the RAM they were talking about. I'm going to keep letting it run and see if it ever completes. Any ideas? Edit.. OK, so it's the same problem that others have seen and they are working on. What I get for not paying more attention the first time around. Sorry about that folks. I know they are working on a fix for it now. Everyone have a good night Zalster |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
Any ideas? Was that with r2721 or r2737? SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
Lunatic v 0.43 I haven't tried any of Raistmer's new apps yet. tomorrow maybe...it's going to be a long night.. |
Mike Send message Joined: 17 Feb 01 Posts: 34258 Credit: 79,922,639 RAC: 80 |
Thats the reason why r_2737 has been released. With each crime and every kindness we birth our future. |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14653 Credit: 200,643,578 RAC: 874 |
Thats the reason why r_2737 has been released. Well, it's been made available as a public Beta test - not a full release yet. Read the posts in the Installer v0.43 Release Notes thread about the issues involved in switching between Beta and Release applications. |
Mike Send message Joined: 17 Feb 01 Posts: 34258 Credit: 79,922,639 RAC: 80 |
Thats the reason why r_2737 has been released. I noticed that of course. With each crime and every kindness we birth our future. |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14653 Credit: 200,643,578 RAC: 874 |
Thats the reason why r_2737 has been released. Of course you did. Just making sure that all the other readers of this thread notice it as well. |
©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.