Ive got 2 machines stuck at 1%. One of them dedicated 3 hours to the workunit before I aborted it. The other is still at 1.604% and is 2 1/2 hours into the workunit....

Info on the workunit...
<result_name>FRA_2rio_134E_hom001_1_2rio_1_1zy4A_IGNORE_THE_REST_214_1303_79_0</result_name>
<active_task_state>1</active_task_state>
<app_version_num>534</app_version_num>
<slot>0</slot>
<scheduler_state>2</scheduler_state>
<checkpoint_cpu_time>0.000000</checkpoint_cpu_time>
<fraction_done>0.016030</fraction_done>
<current_cpu_time>7077.093750</current_cpu_time>
<vm_bytes>0.000000</vm_bytes>
<rss_bytes>0.000000</rss_bytes>

Do I abort it? Im hoping this isnt going to be a regular thing as I really dont want to have to babysit my computers on this project.

Thanks,
Spectre

no, the FRA 2rio wus take a long time to do a model. What you're seeing it it doing that ONE model (all WUs will run atleast ONE model without regard to "cpu run time" preference settings). I've had my older machines exceed the cpu run time pref just to complete that one wu.

tony

no, the FRA 2rio wus take a long time to do a model. What you're seeing it it doing that ONE model (all WUs will run atleast ONE model without regard to "cpu run time" preference settings). I've had my older machines exceed the cpu run time pref just to complete that one wu.

tony

Are you sure this is supposed to be this long? Generally this machine takes only a few hours to complete a job. Im now into 3 hours and it hasnt budged past 1.604%. As a matter of fact, as each sec passes on CPU TIME, COMPLETION TIME also goes up a sec....

What kind of puter? I'll post a pic of my closest one so you can see. I keep track of decoy/wu info on mine.

My AMD64 2800 takes 6000+ seconds to do ONE FRA_2rio_CASP7_hom001_1_2rio_1_1a06__IGNORE_THE_REST_100_1281_37_0 model.

It takes 39,000 seconds(10.83 hours) on my Celeron 500 to do ONE model.

my AMD64 mobile 3700 takes 5,500-6,100 to do one model

my AMD64 3700 sandiego takes from 4,200-4,800 to do one, and my AMD64 X2 4800 takes about the same.

Here's a small sampling of work done by my AMD64 2800, It's sorted by WU name. I use the "not selected" (default) run time preference.

600 pixel wide chart of AMD64 2800 WUs including decoy info

Tony,

I'd appreciate it if you didn't post graphics in normal threads please, even tho the content is useful and relevant (as it is here).

Your stats postings are very useful in the various dedicated threads (like when we were looking at the effects of a certain change to the credit system).

When you post them into a normal thread it makes the page wider than a screen (I view at 1024 pixels wide) and it means I have to scroll the whole screen sideways to read the normal postings.

And I don't want to use a higher pixel setting as the monitor's native resolution is 1024 and the text looks annoyingly messy on any other setting (one of the issues with TFT monitors of course).

And someone else might be trying to read this thread on a display that is even narrower, as they are entitled to.

River~~

deleted

deleted

Imageshack has a handy thumbnail utility. Just plug in the picture and it spits out the code for a thumbnail that links to the full-size picture.

Thanks for all your help, mmciastro. Just came back from the store and it appears it finished up in 3 hrs 28 mins...must have really taken off like a rocket after I left :) Much appreciated!

Spectre

you're welcome

Here is one that I don't understand... I am seeing a lot of these with 5.34
This was run on a XEON 2.8ghz w/hyperthread on.. (Linux 2.6) This code stayed at 1% for 59,397.01 seconds.. and granted 9.8 credits!!! This machine normally gets between 30 and 75 credits for half that amount of run time.. These have only been happening with 5.34.. It's also happening on a pair of IBM x440 XEON MP's... Lot's of run time for a single structure and practically no credit.

Result ID	43988024
Name	1hz6A_BOINC_NATIVEJUMPS_CLOSE_CHAINBREAKS_VARY_ALL_BOND_ANGLES_ALL_BOND_DISTANCES_SAVE_ALL_OUT__1306_1466_0
Workunit	38798622
Created	26 Oct 2006 1:02:24 UTC
Sent	26 Oct 2006 3:52:26 UTC
Received	27 Oct 2006 9:32:10 UTC
Server state	Over
Outcome	Success
Client state	Done
Exit status	0 (0x0)
Computer ID	299336
Report deadline	5 Nov 2006 3:52:26 UTC
CPU time	59397.007282
stderr out	

<core_client_version>5.4.9</core_client_version>
<stderr_txt>
Graphics are disabled due to configuration...
# random seed: 1265230
# cpu_run_time_pref: 43200
======================================================
DONE ::     1 starting structures built         0 (nstruct) times
This process generated      1 decoys from       1 attempts
======================================================


BOINC :: Watchdog shutting down...
BOINC :: BOINC support services shutting down...

</stderr_txt>

Validate state	Valid
Claimed credit	44.3006363200113
Granted credit	9.80524894894768
application version	5.34

---

Looking for a team ??? Join BoincSynergy!!

The workunits does give less credit,.

Here are 2 of them.

https://boinc.bakerlab.org/rosetta/result.php?resultid=44141198

https://boinc.bakerlab.org/rosetta/result.php?resultid=44114981

Is there any pattern in this or is it just this type of wu?

Anders n

---

--

Well, they do grant lower credit... Seems like a lot of horsepower being used to little result... (I know... it could be great result... it's just a lot sound and smoke and little movement!!)...

As an aside, here is a snap of my recent since 5.32 was released.. It started down right at that point...

http://web.hotiron.net/pics/johng/sta2.gif

---

Looking for a team ??? Join BoincSynergy!!