Problems and Technical Issues with Rosetta@home
Message boards : Number crunching : Problems and Technical Issues with Rosetta@home
Previous · 1 . . . 144 · 145 · 146 · 147 · 148 · 149 · 150 . . . 311 · Next
| Author | Message |
|---|---|
|
Falconet Send message Joined: 9 Mar 09 Posts: 354 Credit: 1,276,393 RAC: 828 
|
Looks like another batch of 4.2 work is available. Looks like there are an extra 1.4 million queued tasks when compared to this morning's 2.2 million Pythons. |
 Greg_BE Send message Joined: 30 May 06 Posts: 5691 Credit: 5,859,226 RAC: 0 |
If you hurry, you can get a bit of the action in 4.2. 27,000+ queued up at the moment. 122,000 already processed. Only 1967 people on 4.2 at the moment. That leaves 13 tasks per person to crunch. As for the protein or bio science (non Vbox), then its just FAH, SiDock and WCG outside of the odds and ends here. TACC is a serious hit and miss program. You can go for weeks without getting any work from them. Not worth the time or electricity. It's just stuff they can't run on their supercomputer. I personally do not know of any other projects in protein folding that have work for PC's. |
|
Sid Celery Send message Joined: 11 Feb 08 Posts: 2146 Credit: 41,570,180 RAC: 8,210 
|
Looks like another batch of 4.2 work is available. Yes, but all seem limited to 100 decoys before stopping short, so only running between 1 & 2hrs here. They won't last long at all  
|
|
Jean-David Beyer Send message Joined: 2 Nov 05 Posts: 197 Credit: 6,619,262 RAC: 3,235 
|
Looks like another batch of 4.2 work is available. Yes it does: Tue 07 Dec 2021 03:10:29 PM EST | Rosetta@home | Sending scheduler request: To fetch work. Tue 07 Dec 2021 03:10:29 PM EST | Rosetta@home | Requesting new tasks for CPU Tue 07 Dec 2021 03:10:33 PM EST | Rosetta@home | Scheduler request completed: got 9 new tasks 1938180 Dec 7 15:10 epha2_site1_3skj_3h_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_2pa9fe3l.zip 1943095 Dec 7 15:10 epha2_site1_3skj_3h_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_6gv2zz0m.zip 1996781 Dec 7 15:10 epha2_site1_3skj_3h_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_8yf2ct8i.zip 1878391 Dec 7 15:10 epha2_site3_3c8x_3h_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_5rx2cb2v.zip 1990560 Dec 7 15:10 epha2_site3_3c8x_3h_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_6mw9rn0b.zip 2165492 Dec 7 15:10 niv_site2_6pd4_3h_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_3me3qk7w.zip 1953078 Dec 7 15:10 niv_site2_6pd4_3h_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_9wm2tx8g.zip 1923515 Dec 7 15:10 nkg2d_site1_4s0u_3h_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_1zj8yf9y.zip 2256771 Dec 7 15:10 nkg2d_site1_4s0u_3h_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_7lu6db8g.zip 
|
|
Greg_BE Send message Joined: 30 May 06 Posts: 5691 Credit: 5,859,226 RAC: 0 |
Looks like another batch of 4.2 work is available. An hour or so ago there was just 27K now its up to 29K but the total systems went up. So the average drops to 8 tasks per system. Run time to finish is just around 2hrs even though it shows 8 hrs when running. |
|
Greg_BE Send message Joined: 30 May 06 Posts: 5691 Credit: 5,859,226 RAC: 0 |
Looks like another batch of 4.2 work is available. I still have to keep my project_max in play for when I go back to python. I don't have time to enable app_config and remove it and put it back all the time. |
|
Jean-David Beyer Send message Joined: 2 Nov 05 Posts: 197 Credit: 6,619,262 RAC: 3,235
|
Yes, but all seem limited to 100 decoys before stopping short, so only running between 1 & 2hrs here. Here too. All mine hit 100 decoys.
|
|
Falconet Send message Joined: 9 Mar 09 Posts: 354 Credit: 1,276,393 RAC: 828
|
Yes, 100 decoys. Mine just took much longer because they are running on my AMD 2500u laptop. |
|
Greg_BE Send message Joined: 30 May 06 Posts: 5691 Credit: 5,859,226 RAC: 0 |
Yes, 100 decoys. Mine just took much longer because they are running on my AMD 2500u laptop. DONE :: 100 starting structures 7075.52 cpu seconds This process generated 100 decoys from 100 attempts 196 minutes used |
|
MStenholm Send message Joined: 18 Apr 20 Posts: 19 Credit: 26,729,489 RAC: 18,574
|
Most countries have 60 minutes per hour and 60 seconds per minute but according to your calculation Belgium have 36 minutes per hour :). I know what you are doing but you have a long live ahead of you and someone ought to snap you out of that misconception. 7075/(60*60) = 1,96 hours, not 196 minutes. 1,96 hour = 1 hour and 0,96*60/100 = 57.6 minutes 60+57.6 =117,6 minutes |
|
Greg_BE Send message Joined: 30 May 06 Posts: 5691 Credit: 5,859,226 RAC: 0 |
Most countries have 60 minutes per hour and 60 seconds per minute but according to your calculation Belgium have 36 minutes per hour :). I know what you are doing but you have a long live ahead of you and someone ought to snap you out of that misconception. Middle of the night math! LMAO. Thanks for the correction. |
|
Greg_BE Send message Joined: 30 May 06 Posts: 5691 Credit: 5,859,226 RAC: 0 |
Something different...I have been getting some 4.2 stuff and now I am back to Python. Anyone figure out the scheduler routine yet? If I was to up the cores in the app_config (project_max_concurrent) to 3 or 4, would I get all Python or a blend of 4.2 and Python? |
|
robertmiles Send message Joined: 16 Jun 08 Posts: 1234 Credit: 14,338,560 RAC: 826
|
Something different...I have been getting some 4.2 stuff and now I am back to Python. It seems to be a random mix of whatever of available, rather than a schedule. I suspect that it is whatever they are currently teaching their students to create workunits for. |
|
Jean-David Beyer Send message Joined: 2 Nov 05 Posts: 197 Credit: 6,619,262 RAC: 3,235
|
I have the following in my app_config file: [/var/lib/boinc/projects/boinc.bakerlab.org_rosetta]$ cat app_config.xml <app_config> <project_max_concurrent>3</project_max_concurrent> </app_config> and the boinc-cllient runs up to three of these work units at a time. I am getting plenty of work units. I have a 16-core machine but the boinc-client is instructed to run no more than 8 Boinc tasks at a time. The various app_config.xml files allow up to 4 ClimatePrediction tasks at a time, 5 WCG at a time, three Rosetta at a time, and 2 Universe tasks at a time -- though the priority of the Universe tasks is so low that they hardly ever run, and then only one at a time. Currently I have these Rosetta tasks on my machine and none of them are running. 3443919 Dec 9 13:09 fcgr3a_site1_5mn2_3h_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_7ts1jp9r.zip 3453466 Dec 9 13:09 fcgr3a_site1_5mn2_3h_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_9zi4uu3j.zip 2646638 Dec 9 13:09 her2_site4_3h_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_0ic4nn1s.zip 2544075 Dec 9 13:09 her2_site4_3h_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_6kw0cf1j.zip 2804583 Dec 9 13:09 niv_site3_6vy5_jhr_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_2fs1jz6r.zip 2585084 Dec 9 13:09 niv_site3_6vy6_jhr_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_2cm7fe4w.zip 2748992 Dec 9 13:09 niv_site3_6vy6_jhr_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_2jg7fo9z.zip 2640314 Dec 9 13:09 niv_site3_6vy6_jhr_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_7jp7ej6d.zip 2575060 Dec 9 13:09 niv_site3_6vy6_jhr_ggraft_1_SAVE_ALL_OUT_IGNORE_THE_REST_8yo1qr6j.zip This is when I have been receiving work units for Rosetta: Tue 07 Dec 2021 03:10:29 PM EST | Rosetta@home | Sending scheduler request: To fetch work. Tue 07 Dec 2021 03:10:29 PM EST | Rosetta@home | Requesting new tasks for CPU Tue 07 Dec 2021 03:10:33 PM EST | Rosetta@home | Scheduler request completed: got 9 new tasks Wed 08 Dec 2021 12:30:55 AM EST | Rosetta@home | Sending scheduler request: To fetch work. Wed 08 Dec 2021 12:30:55 AM EST | Rosetta@home | Requesting new tasks for CPU Wed 08 Dec 2021 12:30:59 AM EST | Rosetta@home | Scheduler request completed: got 8 new tasks Wed 08 Dec 2021 09:52:40 AM EST | Rosetta@home | Sending scheduler request: To fetch work. Wed 08 Dec 2021 09:52:40 AM EST | Rosetta@home | Reporting 2 completed tasks Wed 08 Dec 2021 09:52:40 AM EST | Rosetta@home | Requesting new tasks for CPU Wed 08 Dec 2021 09:52:42 AM EST | Rosetta@home | Scheduler request completed: got 8 new tasks Wed 08 Dec 2021 10:55:08 PM EST | Rosetta@home | Sending scheduler request: To fetch work. Wed 08 Dec 2021 10:55:08 PM EST | Rosetta@home | Reporting 3 completed tasks Wed 08 Dec 2021 10:55:08 PM EST | Rosetta@home | Requesting new tasks for CPU Wed 08 Dec 2021 10:55:16 PM EST | Rosetta@home | Scheduler request completed: got 9 new tasks I do not know about python tasks because I do not get them because I cannot run them.
|
|
Greg_BE Send message Joined: 30 May 06 Posts: 5691 Credit: 5,859,226 RAC: 0 |
Something different...I have been getting some 4.2 stuff and now I am back to Python. I think you missed out at what I am asking. I get both 4.2 and Python. However, I had to limit RAH to 2 tasks at a time because while 4.2 runs just fine and takes very little memory and processes very fast, Python is a memory hog and if it runs 3 it kills my other projects ability to run seamlessly. So I was wondering if anyone has seen a pattern from the two schedulers, does it do 4.2 for a time and then go back to Python or does it also go 1:1 or whatever with python and 4.2 at the same time? Or is it just a random mix of the two? |
|
Greg_BE Send message Joined: 30 May 06 Posts: 5691 Credit: 5,859,226 RAC: 0 |
Before I got interested in python I let RAH take whatever it wanted. Then when I got into Python and letting it do its thing, then it was 3 tasks and taking up over half the memory and stalling other tasks. Then I took your project_concurrent and limited it to 2. But I see "name" does not affect anything. Project is project. So RAH is limited to 2 tasks no matter which application is running. I am afraid to release it, because then Python will take over. Unless that was a matter of RAH debt when 4.2 ran out. All I know is that Python x3 is a system killer unless I buy some new memory sticks. Then we are talking an absurd amount of memory to keep BOINC running the way I like it to run or dropping Python which seems crazy after complaining about not getting it for so long. Why can't you run Python? I forgot. |
|
.clair. Send message Joined: 2 Jan 07 Posts: 274 Credit: 26,399,595 RAC: 0 |
All I know is that Python is a system killer unless I buy some new memory sticks. I don't know what motherboard you have, but don't scimp on it, you may have to max out what the board can take. ................ Recently because of going to some dodgy website an downloading a new version of boinc mangler 7,16.20 that had `virtual pox` in with it. {its not had VB before} I get my computer infected with pythons in cages, and they are horrible big things that realy take over the poor thing. I decided to upgrade one my systems from 32GB to 128GB {not its max} it`s `only` a twin xeon E5 2697 48 thread 128GB of ram and they can realy stuff it. even if actual memory use is 50GB and only run 16 at a time. it'll run plenty more R4.20, [ and other work] further up this thread I had problems with another system getting daft disk space messages that I now ignore, {and I had just upped that from 16 to 32 GB} now even this one gets them, Just now, rosetta is using 77GB of disk space and it has 246GB more free available for boinc to use. ............. 09-Dec-2021 06:40:46 [Rosetta@home] Rosetta needs 1907.35MB more disk space. You currently have 0.00 MB available and it needs 1907.35 MB. 09-Dec-2021 06:40:46 [Rosetta@home] rosetta python projects needs 19073.49MB more disk space. You currently have 0.00 MB available and it needs 19073.49 MB ............ crazy stuff I don't bother about any more And I have rosetta set to a zero resource share on that system to try and limit what funky stuff it can do |
|
Sid Celery Send message Joined: 11 Feb 08 Posts: 2146 Credit: 41,570,180 RAC: 8,210
|
Looks like another batch of 4.2 work is available. I did get a few Robetta tasks that ran the full 8hrs but it looks like all Rosetta 4.20 tasks have been downloaded now. Won't be long before I'm back running WCG My resource share is set to Rosetta 29 - WCG 1 but my RAC is 6.5k to 19.6k atm... <sigh>
|
|
Jean-David Beyer Send message Joined: 2 Nov 05 Posts: 197 Credit: 6,619,262 RAC: 3,235
|
Why can't you run Python? I forgot. I do not have VirtualBox, so I cannot run them.
|
|
Greg_BE Send message Joined: 30 May 06 Posts: 5691 Credit: 5,859,226 RAC: 0 |
Why can't you run Python? I forgot. Do you want to run them? Then download virtual box 5 not 6. And enable the option to get them on your profile on this page: https://boinc.bakerlab.org/rosetta/show_host_detail.php?hostid=5958977 (Thats your windows machine) But they are memory hogs (can only do 3 tasks) plus the VDI is 9 gigs. |
Message boards :
Number crunching :
Problems and Technical Issues with Rosetta@home
©2024 University of Washington
https://www.bakerlab.org