Message boards :
Number crunching :
GUPPI Rescheduler for Linux and Windows - Move GUPPI work to CPU and non-GUPPI to GPU
Message board moderation
Previous · 1 . . . 11 · 12 · 13 · 14 · 15 · 16 · 17 . . . 37 · Next
Author | Message |
---|---|
Luigi R. Send message Joined: 26 Nov 13 Posts: 10 Credit: 1,608,382 RAC: 0 |
Hello, I found that GUPPI Rescheduler doesn't work for me cause of different version_num between CPU (v8.05) and GPU (v8.01) apps. |
Shaggie76 Send message Joined: 9 Oct 09 Posts: 282 Credit: 271,858,118 RAC: 196 |
@Raistmer - I tried to create an account on that forum but got "An Error Has Occurred! Sorry, registration is currently disabled." Might be better to talk code over there than this thread. |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
I tried to create an account on that forum but got "An Error Has Occurred! Sorry, registration is currently disabled." It's not possible to create account there AFAIK. But that thread should be visible to anonymous guests too, not? If not please report back I'll re-post elsewhere. SETI apps news We're not gonna fight them. We're gonna transcend them. |
Jimbocous Send message Joined: 1 Apr 13 Posts: 1849 Credit: 268,616,081 RAC: 1,349 |
Anything I can look at or send your way to help swat this? Besides an extra "d" for overridden, that is? :p It occurs to me that another place this software concept could be of use would be in moving AP tasks from CPU to GPU, perhaps only by command line option flag. Just seems silly to spend 12 hours doing an AP on CPU that will run in an hour on a 750ti. so: -a Move Astropulse tasks to GPU(s) -b Operate in batch mode (no inputs) -g Move GUPPIs to CPU -v Move Aricebo VLARs to CPU and, of course, -f I'd like fries with that :) |
Mike Send message Joined: 17 Feb 01 Posts: 34253 Credit: 79,922,639 RAC: 80 |
I tried to create an account on that forum but got "An Error Has Occurred! Sorry, registration is currently disabled." This is in the dev area so he couldn`t read it anyways. He needs at least alpha tester status. With each crime and every kindness we birth our future. |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
I tried to create an account on that forum but got "An Error Has Occurred! Sorry, registration is currently disabled." Nope. Discussion thread, I just checked with private anonymous mode under Chrome. SETI apps news We're not gonna fight them. We're gonna transcend them. |
Shaggie76 Send message Joined: 9 Oct 09 Posts: 282 Credit: 271,858,118 RAC: 196 |
I tried to create an account on that forum but got "An Error Has Occurred! Sorry, registration is currently disabled." I can read it -- it just seems wrong to talk about windows threading model in "GUPPI Rescheduler for Linux and Windows - Move GUPPI work to CPU and non-GUPPI to GPU" |
Jimbocous Send message Joined: 1 Apr 13 Posts: 1849 Credit: 268,616,081 RAC: 1,349 |
@Raistmer - I tried to create an account on that forum but got "An Error Has Occurred! Sorry, registration is currently disabled." Yeah, good luck. I tried to get validated over there for almost two years before I wrote it off as of a waste of time. That's when I dumped my team membership in KWSN. No one is minding the store, and the new user process is broken ... |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
@Raistmer - I tried to create an account on that forum but got "An Error Has Occurred! Sorry, registration is currently disabled." Ok, I'll create separate thread. SETI apps news We're not gonna fight them. We're gonna transcend them. |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
http://setiathome.berkeley.edu/forum_thread.php?id=80173 SETI apps news We're not gonna fight them. We're gonna transcend them. |
Jimbocous Send message Joined: 1 Apr 13 Posts: 1849 Credit: 268,616,081 RAC: 1,349 |
Mr. Kevvy's GUPPI Rescheduler v0.51 - (c)2016 Kevin Dorner One other bit of info. This only happens on the host linked above, my other 4 boxes seem fine, at this point. Just ran it, all was good. ??? |
Keith Myers Send message Joined: 29 Apr 01 Posts: 13161 Credit: 1,160,866,277 RAC: 1,873 |
When I ran the rescheduler last night I got the "Error: could not determine CPU version_num from client_state. Nothing changed." on two machines. The third ran it OK. Still believe if the the nonVLAR count is only 1 or 2, then the command file fails with that error message. No harm, no foul. It just exits cleanly and doesn't accomplish anything. Ran it again a half hour ago across all 3 machines and it was successful moving 15-30 tasks. Plenty of AP on board too to show that it handled them well also. Seti@Home classic workunits:20,676 CPU time:74,226 hours A proud member of the OFA (Old Farts Association) |
rcthardcore Send message Joined: 23 Nov 08 Posts: 48 Credit: 1,306,006 RAC: 0 |
I have decided to run ALL guppy work units on CPU only. I am putting my GPU to better use. |
Stephen "Heretic" Send message Joined: 20 Sep 12 Posts: 5557 Credit: 192,787,363 RAC: 628 |
I have decided to run ALL guppy work units on CPU only. I am putting my GPU to better use. . . Welcome to the dark side brother! LOL . . Just don't mention it to Zalster or Grant :) |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
I have decided to run ALL guppy work units on CPU only. I am putting my GPU to better use. Yes, cause you can see how lousy my GPUs handle them, bahahahaha edit.. No disrespect to Mr. Kevvy, his app works great for what it's intended to do. |
Jimbocous Send message Joined: 1 Apr 13 Posts: 1849 Credit: 268,616,081 RAC: 1,349 |
When I ran the rescheduler last night I got the "Error: could not determine CPU version_num from client_state. Nothing changed." on two machines. The third ran it OK. Still believe if the the nonVLAR count is only 1 or 2, then the command file fails with that error message. No harm, no foul. It just exits cleanly and doesn't accomplish anything. Ran it again a half hour ago across all 3 machines and it was successful moving 15-30 tasks. Plenty of AP on board too to show that it handled them well also. Just rerun. Approximate status after run: Box #1 E5450 Xeon CPU - AP - 1 in process, 1 ready to run - BLC - 1 in process, 40 ready to run - VLAR- 0 in process, 44 ready to run - MB - 0 in process, 0 ready to run GPU - AP - 1 in process, 50 ready to run - BLC - 0 in process, 7 ready to run - VLAR- 0 in process, 0 ready to run - MB - 0 in process, 12 ready to run RESULT - Invalid ... v0.51 ... Reading configuration files... Error: could not determine CPU version_num from client_state. Nothing changed. Box #2 W3550 Xeon CPU - AP - 1 in process, 2 ready to run - BLC - 6 in process, 2 ready to run - VLAR- 0 in process, 16 ready to run - MB - 0 in process, 0 ready to run GPU - AP - 4 in process, 144 ready to run - BLC - 0 in process, 0 ready to run - VLAR- 0 in process, 0 ready to run - MB - 0 in process, 0 ready to run RESULT - Valid ... v0.51 ... Reading configuration files... Error: There must be at least one GPU workunit in sched_request_setiathome.berkeley.edu.xml to function. Box #3 W3550 Xeon CPU - AP - 0 in process, 27 ready to run - BLC - 6 in process, 11 ready to run - VLAR- 1 in process, 0 ready to run - MB - 0 in process, 0 ready to run GPU - AP - 6 in process, 34 ready to run - BLC - 0 in process, 0 ready to run - VLAR- 0 in process, 0 ready to run - MB - 0 in process, 218 ready to run RESULT - Valid ... v0.51 ... Reading configuration files... Found sched_request GPU platform=windows_intelx86 app_version=26 version_num=800 CPU platform=windows_x86_64 app_version=14 version_num=800 and GPU plan_class=cuda50 Searching for and moving workunits in client state... Writing updated configuration client_state.xml... Done: 60 non-GUPPI workunits moved to GPU and 0 GUPPI workunits moved to CPU. Box #4 E5620 Xeon x 2 CPU - AP - 0 in process, 18 ready to run - BLC - 7 in process, 16 ready to run - VLAR- 0 in process, 1 ready to run - MB - 7 in process, 0 ready to run GPU - AP - 6 in process, 49 ready to run - BLC - 0 in process, 0 ready to run - VLAR- 0 in process, 0 ready to run - MB - 0 in process, 171 ready to run RESULT - Valid ... v0.51 ... Reading configuration files... Found sched_request GPU platform=windows_intelx86 app_version=28 version_num=800 CPU platform=windows_x86_64 app_version=16 version_num=800 and GPU plan_class=cuda50 Searching for and moving workunits in client state... Writing updated configuration client_state.xml... Done: 49 non-GUPPI workunits moved to GPU and 9 GUPPI workunits moved to CPU. 7 workunits not moved as they were in progress. Box #5 Q9650 CPU - AP - 0 in process, 1 ready to run - BLC - 2 in process, 54 ready to run - VLAR- 0 in process, 29 ready to run - MB - 0 in process, 0 ready to run GPU - AP - 0 in process, 91 ready to run - BLC - 2 in process, 5 ready to run - VLAR- 0 in process, 0 ready to run - MB - 2 in process, 11 ready to run RESULT - Valid ... v0.51 ... Reading configuration files... Found sched_request GPU platform=windows_intelx86 app_version=28 version_num=800 CPU platform=windows_x86_64 app_version=16 version_num=800 and GPU plan_class=cuda50 Searching for and moving workunits in client state... No non-GUPPI workunits are assigned to CPU to move to GPU; no changes made. Apologies if this is too much, better that than too little? |
Keith Myers Send message Joined: 29 Apr 01 Posts: 13161 Credit: 1,160,866,277 RAC: 1,873 |
Box #1 did what was expected, no nonVLAR on the CPU to move. Box #2 same as above Box #3 moved 60 phantom nonVLAR????? when none actually on board Box #4 moved 48 phantom tasks confused as AP or something?? moved phantom guppis?? Box #5 worked as expected The only outliers are Box #3 and #4. How did it supposedly move non-existent tasks. The DQO is obviously getting confused with AP tasks still but at least not ghosting tasks as it did before. Seti@Home classic workunits:20,676 CPU time:74,226 hours A proud member of the OFA (Old Farts Association) |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13720 Credit: 208,696,464 RAC: 304 |
The only outliers are Box #3 and #4. How did it supposedly move non-existent tasks. Existing ghosts? Grant Darwin NT |
Jimbocous Send message Joined: 1 Apr 13 Posts: 1849 Credit: 268,616,081 RAC: 1,349 |
Box #1 did what was expected, no nonVLAR on the CPU to move. No, I was afraid of creating that confusion. Note that the inventory was taken AFTER the run. I mentioned that in the message, but probably should have added CAPS, bold or italic rather than just underlining. Proper methodology would have been to measure THEN run. Just didn't think of doing this for Kevin until after I had run. Next time I will. But box 1 results show an entry condition for a failure, and 2-5 can have the initial entry state determined by looking at the after results and adjusting for what the program did as indicated in its log. Convoluted, I know, but there it is ... l8r That's why I noted at the end of each whether it was valid (all but #1) or not. |
Keith Myers Send message Joined: 29 Apr 01 Posts: 13161 Credit: 1,160,866,277 RAC: 1,873 |
Yes, I missed that after preface. Just wasn't expecting a summation after running. Seti@Home classic workunits:20,676 CPU time:74,226 hours A proud member of the OFA (Old Farts Association) |
©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.