Folding Forum

Posted: **Tue Dec 02, 2008 4:42 am**

Never saw this before (in my vast experience

), so I thought I'd post it. Got the successor WU to one of my 3798s.

[01:23:45] - Starting from initial work packet
[01:23:45] 
[01:23:45] Project: 3798 (Run 56, Clone 2, Gen 4)
[01:23:45] 
[01:23:45] Assembly optimizations on if available.
[01:23:45] Entering M.D.
[01:23:50] + New frame time estimate; Working...
[01:23:51] Protein: p3798
[01:23:51] 
[01:23:51] Writing local files
[01:23:55] + New frame time estimate; Working...
[01:25:06] Extra SSE boost OK.
[01:25:06] Writing local files
[01:25:06] Completed 0 out of 1500 steps  (0)
[01:25:53] Writing local files
[01:25:53] Completed 500 out of 1500 steps  (33)
[01:26:41] Writing local files
[01:26:41] Completed 1000 out of 1500 steps  (67)
[01:26:45] + Writing 'sec_per_frame = 2390.000000' to config
[01:26:45] + Working ...+ New frame time estimate; Working...
[01:26:55] + New frame time estimate; Working...
[01:27:00] + New frame time estimate; Working...
[01:27:05] + New frame time estimate; Working...
[01:27:10] + New frame time estimate; Working...
[01:27:15] + New frame time estimate; Working...
[01:27:20] + New frame time estimate; Working...
[01:27:25] + New frame time estimate; Working...
[01:27:28] Writing local files
[01:27:28] Completed 1500 out of 1500 steps  (100)
[01:27:28] Writing final coordinates.
[01:27:28] Past main M.D. loop
[01:27:30] + New frame time estimate; Working...
[01:27:35] + New frame time estimate; Working...
[01:27:40] + New frame time estimate; Working...
[01:27:45] + New frame time estimate; Working...
[01:27:50] + New frame time estimate; Working...
[01:27:55] + New frame time estimate; Working...
[01:28:00] + New frame time estimate; Working...
[01:28:05] + New frame time estimate; Working...
[01:28:10] + New frame time estimate; Working...
[01:28:15] + New frame time estimate; Working...
[01:28:20] + New frame time estimate; Working...
[01:28:25] + New frame time estimate; Working...
[01:28:28] 
[01:28:28] Finished Work Unit:
[01:28:28] - Reading up to 550656 from "work/wudata_09.arc": Read 550656
[01:28:28] - Reading up to 0 from "work/wudata_09.xtc": Read 0
[01:28:28] goefile size: 0
[01:28:28] Leaving Run
[01:28:30] + New frame time estimate; Working...
[01:28:30] - Writing 573212 bytes of core data to disk...
[01:28:30] Done: 572700 -> 519292 (compressed to 90.6 percent)
[01:28:30]   ... Done.
[01:28:30] - Shutting down core
[01:28:30] 
[01:28:30] Folding@home Core Shutdown: FINISHED_UNIT
[01:28:33] CoreStatus = 64 (100)
[01:28:33] Unit 9 finished with 100 percent of time to deadline remaining.
[01:28:33] Updated performance fraction: 0.994601
[01:28:33] Sending work to server


[01:28:33] + Attempting to send results
[01:28:33] - Reading file work/wuresults_09.dat from core
[01:28:33]   (Read 519804 bytes from disk)
[01:28:33] Connecting to http://171.64.122.139:8080/
[01:30:28] Posted data.
[01:30:28] Initial: 0000; - Uploaded at ~4 kB/s
[01:30:28] - Averaged speed for that direction ~23 kB/s
[01:30:28] + Results successfully sent
[01:30:28] Thank you for your contribution to Folding@Home.
[01:30:28] + Number of Units Completed: 391

[01:30:32] Trying to send all finished work units
[01:30:32] + No unsent completed units remaining.
[01:30:32] - Preparing to get new work unit...
[01:30:32] + Attempting to get work packet
[01:30:32] - Will indicate memory of 2046 MB.
[01:30:32] - Connecting to assignment server
[01:30:32] Connecting to http://assign.stanford.edu:8080/
[01:30:32] Posted data.
[01:30:32] Initial: 40AB; - Successful: assigned to (171.64.122.139).
[01:30:32] + News From Folding@Home: Welcome to Folding@Home
[01:30:32] Loaded queue successfully.
[01:30:32] Connecting to http://171.64.122.139:8080/
[01:30:36] Posted data.
[01:30:36] Initial: 0000; - Receiving payload (expected size: 238411)
[01:33:01] - Downloaded at ~1 kB/s
[01:33:01] - Averaged speed for that direction ~47 kB/s
[01:33:01] + Received work.
[01:33:01] Trying to send all finished work units
[01:33:01] + No unsent completed units remaining.
[01:33:01] + Closed connections
[01:33:01] 
[01:33:01] + Processing work unit
[01:33:01] Core required: FahCore_78.exe
[01:33:01] Core found.
[01:33:01] Working on Unit 00 [December 2 01:33:01]
[01:33:01] + Working ...
[01:33:01] - Calling 'FahCore_78.exe -dir work/ -suffix 00 -checkpoint 15 -verbose -lifeline 3908 -version 503'

[01:33:01] 
[01:33:01] *------------------------------*
[01:33:01] Folding@Home Gromacs Core
[01:33:01] Version 1.90 (March 8, 2006)
[01:33:01] 
[01:33:01] Preparing to commence simulation
[01:33:01] - Looking at optimizations...
[01:33:01] - Created dyn
[01:33:01] - Files status OK
[01:33:01] - Expanded 237899 -> 1167708 (decompressed 490.8 percent)
[01:33:01] - Starting from initial work packet
[01:33:01] 
[01:33:01] Project: 3798 (Run 56, Clone 2, Gen 5)
[01:33:01] 
[01:33:01] Assembly optimizations on if available.
[01:33:01] Entering M.D.
[01:33:03] + New frame time estimate; Working...
[01:33:07] Protein: p3798
[01:33:07] 
[01:33:07] Writing local files
[01:33:08] + New frame time estimate; Working...
[01:34:22] Extra SSE boost OK.
[01:34:22] Writing local files
[01:34:22] Completed 0 out of 1500 steps  (0)
[01:35:09] Writing local files
[01:35:09] Completed 500 out of 1500 steps  (33)
[01:35:57] Writing local files
[01:35:57] Completed 1000 out of 1500 steps  (67)
[01:35:58] + Writing 'sec_per_frame = 553.000000' to config
[01:35:58] + Working ...+ New frame time estimate; Working...
[01:36:08] + New frame time estimate; Working...
[01:36:13] + New frame time estimate; Working...
[01:36:18] + New frame time estimate; Working...
[01:36:23] + New frame time estimate; Working...
[01:36:28] + New frame time estimate; Working...
[01:36:33] + New frame time estimate; Working...
[01:36:38] + New frame time estimate; Working...
[01:36:43] + New frame time estimate; Working...
[01:36:44] Writing local files
[01:36:44] Completed 1500 out of 1500 steps  (100)
[01:36:44] Writing final coordinates.
[01:36:44] Past main M.D. loop
[01:36:48] + New frame time estimate; Working...
[01:36:53] + New frame time estimate; Working...
[01:36:58] + New frame time estimate; Working...
[01:37:03] + New frame time estimate; Working...
[01:37:08] + New frame time estimate; Working...
[01:37:13] + New frame time estimate; Working...
[01:37:18] + New frame time estimate; Working...
[01:37:23] + New frame time estimate; Working...
[01:37:28] + New frame time estimate; Working...
[01:37:33] + New frame time estimate; Working...
[01:37:38] + New frame time estimate; Working...
[01:37:43] + New frame time estimate; Working...
[01:37:44] 
[01:37:44] Finished Work Unit:
[01:37:44] - Reading up to 550656 from "work/wudata_00.arc": Read 550656
[01:37:44] - Reading up to 0 from "work/wudata_00.xtc": Read 0
[01:37:44] goefile size: 0
[01:37:44] Leaving Run
[01:37:46] - Writing 573212 bytes of core data to disk...
[01:37:47] Done: 572700 -> 519270 (compressed to 90.6 percent)
[01:37:47]   ... Done.
[01:37:47] - Shutting down core
[01:37:47] 
[01:37:47] Folding@home Core Shutdown: FINISHED_UNIT

Posted: **Wed Dec 03, 2008 3:45 am**

http://fah-web.stanford.edu/cgi-bin/fahproject?p=3798

It's a test WU.

<edit> corrected link

viewtopic.php?f=19&t=7207#p71999

Posted: **Wed Dec 03, 2008 7:35 am**

Thanks for the info. I was referring to the fact that I got the consecutive gen to one unit.

Project: 3798 (Run 56, Clone 2, Gen 4)
Project: 3798 (Run 56, Clone 2, Gen 5)

Posted: **Wed Dec 03, 2008 9:40 am**

Did you buy a lottery ticket this week?

Posted: **Wed Dec 03, 2008 4:08 pm**

hopefully not, as he wasted all the luck on getting a continuous gen WU.

Posted: **Thu Dec 04, 2008 5:50 am**

Good thing I saved my money.

Turns out it may not be such an oddity after all, or at least it occurred one other time:

Code: Select all

[08:54:07] - Starting from initial work packet
[08:54:07] 
[08:54:07] Project: 3798 (Run 37, Clone 1, Gen 10)
[08:54:07] 
[08:54:07] Assembly optimizations on if available.
    {snip}
[08:58:27] - Starting from initial work packet
[08:58:27] 
[08:58:27] Project: 3798 (Run 37, Clone 1, Gen 11)
[08:58:27] 
[08:58:27] Assembly optimizations on if available.
[08:58:27] Entering M.D.
[08:58:34] Protein: p3798

Posted: **Thu Dec 04, 2008 1:46 pm**

Yep, I've got these back-to-back gens too. I thought the extremely quick turnaround time for this project was a factor. But then I realized, the next Gen isn't supposed to be generated until Gen n-1 is submitted and the results analyzed. Does the server do this in the few seconds between result upload and WU download?

Posted: **Thu Dec 04, 2008 11:38 pm**

anandhanju wrote:Yep, I've got these back-to-back gens too. I thought the extremely quick turnaround time for this project was a factor. But then I realized, the next Gen isn't supposed to be generated until Gen n-1 is submitted and the results analyzed. Does the server do this in the few seconds between result upload and WU download?

I'm not sure what in included under your "results analyzed" category. As a minimum there are certain validation tests but it's likely that they're really fast. In a few projects, there's some coordination between various clones which certainly would take time, but that's an unusual case. I suspect that for most projects, the heavy analysis is done by the researcher off-line and perhaps only after the trajectory is completed, so there's going to be a lot of projects that take almost zero time to create a new WU from result number n-1. If the final checkpoint is turned into the zeroth checkpoint, a new WU can start immediately (though it's not quite that simple).

Posted: **Fri Dec 05, 2008 12:10 am**

bruce wrote:...there's going to be a lot of projects that take almost zero time to create a new WU from result number n-1.

Thanks bruce, that answers my question.

I was under the [wrong] impression that the reseacher needs to step in to make this happen.

Posted: **Fri Dec 05, 2008 4:29 am**

If a trajectory consists of (say) 50 Gens, and the average turn-around time for a high performance client like SMP is 3 days, then a trajectory can take 5 months to complete. If every Gen required the intervention of the researcher, it might take a year or two. It makes more sense to me for a project to run for 5 months and the researcher to spend several months studying the data. If that produces publishable results, you'd have to add time to prepare a paper, get it peer-reviews, and eventually get it published. If it does not produce publishable results, (s)he may go back to the drawing board to design a new experiment.

Clearly during that first 5 months, partial results may be analyzed leading to early termination of a project or to the design of new projects that can run in parallel.

These are only my guesses for project timing so they may or may not be typical. Certainly the average turn-around time per Gen is MUCH longer for the CPU client and shorter for the PS3/GPU clients but the number of Gens may also be different.

Folding Forum

Oddity - back to back units

Oddity - back to back units

Re: Oddity - back to back units

Re: Oddity - back to back units

Re: Oddity - back to back units

Re: Oddity - back to back units

Re: Oddity - back to back units

Re: Oddity - back to back units

Re: Oddity - back to back units

Re: Oddity - back to back units

Re: Oddity - back to back units