Message boards : Number crunching : Short-running WU ( 13 to 30 seconds) Is this OK???
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile jay

Send message
Joined: 10 Aug 07
Posts: 56
Credit: 831,474
RAC: 0
Message 19278 - Posted: 19 Mar 2008, 21:22:16 UTC

Greetings!

Here is a sample WU:
http://lhcathome.cern.ch/lhcathome/workunit.php?wuid=2444387

I was happy to see that I was not the only one that got the run time of less than one minute...

Was this intended?

The name of this WU is/was:
w2_lhc270_4__51__s__64.28_59.31__18_20__5__45_1_sixvf_boinc23480


Recently, I have seen other WUs with times of
13.3
0.86
0.83
3.48
3.44 seconds.

Just looked strange. Thought I would ask..

T H A N K Y O U
Jay
ID: 19278 · Report as offensive     Reply Quote
Profile BlueBird

Send message
Joined: 16 Mar 08
Posts: 1
Credit: 4,594
RAC: 0
Message 19279 - Posted: 19 Mar 2008, 22:49:22 UTC - in response to Message 19278.  

Greetings!

Here is a sample WU:
http://lhcathome.cern.ch/lhcathome/workunit.php?wuid=2444387

I was happy to see that I was not the only one that got the run time of less than one minute...

Was this intended?

The name of this WU is/was:
w2_lhc270_4__51__s__64.28_59.31__18_20__5__45_1_sixvf_boinc23480


Recently, I have seen other WUs with times of
13.3
0.86
0.83
3.48
3.44 seconds.

Just looked strange. Thought I would ask..

T H A N K Y O U
Jay



I just had several short ones and a couple that were about half an hour long.

mike
ID: 19279 · Report as offensive     Reply Quote
Michael Karlinsky
Avatar

Send message
Joined: 18 Sep 04
Posts: 163
Credit: 1,682,370
RAC: 0
Message 19280 - Posted: 19 Mar 2008, 23:03:13 UTC - in response to Message 19278.  


Just looked strange. Thought I would ask..

T H A N K Y O U
Jay



That is OK, see e.g. here.
Team Linux Users Everywhere
ID: 19280 · Report as offensive     Reply Quote
Profile ChertseyAl
Avatar

Send message
Joined: 28 Nov 05
Posts: 31
Credit: 115,957
RAC: 0
Message 19283 - Posted: 20 Mar 2008, 9:14:16 UTC - in response to Message 19278.  

Just looked strange. Thought I would ask..


As per Michaels link earlier, the short WUs are 'Wall-Crashers' ;)

WUs currently come in three flavours - Long (1million turns), Short (100k turns) and Wall-Crashers (Kaboom!).

It's frustrating to get a batch of WUs and all of them to turn out to be WCs, but it's all part of the science :)

Al.
ID: 19283 · Report as offensive     Reply Quote
Profile EB_Foster

Send message
Joined: 26 Feb 08
Posts: 3
Credit: 12,815
RAC: 0
Message 19285 - Posted: 20 Mar 2008, 18:21:43 UTC

I've been getting a lot of the short WUs also. But I think I'd rather have those than the 80 hour monster that uFluids sent me - I don't see how it can possibly finish before the report deadline...
ID: 19285 · Report as offensive     Reply Quote
EclipseHA

Send message
Joined: 18 Sep 04
Posts: 47
Credit: 1,886,234
RAC: 0
Message 19286 - Posted: 21 Mar 2008, 1:23:34 UTC
Last modified: 21 Mar 2008, 1:24:13 UTC

if your WU only ran a short time, the beam hit the wall..

If it claimed less than 0.01 credits, you'll be able to remember it for a long, long, long time, as it will stay in a "pending" state in your results list!

I've got pages of these 0.00x "pending" WU's from the last year or so.... :(
ID: 19286 · Report as offensive     Reply Quote
KAMasud

Send message
Joined: 7 Oct 06
Posts: 114
Credit: 23,192
RAC: 0
Message 19287 - Posted: 21 Mar 2008, 5:09:57 UTC


LoL'z, Alex and Neasan my pages also overflowing with these wall crashers :( can you chaps, do nothing about them?
Regards
Masud.
ID: 19287 · Report as offensive     Reply Quote
Profile EB_Foster

Send message
Joined: 26 Feb 08
Posts: 3
Credit: 12,815
RAC: 0
Message 19291 - Posted: 21 Mar 2008, 20:07:54 UTC - in response to Message 19286.  

if your WU only ran a short time, the beam hit the wall..

If it claimed less than 0.01 credits, you'll be able to remember it for a long, long, long time, as it will stay in a "pending" state in your results list!

I've got pages of these 0.00x "pending" WU's from the last year or so.... :(


There are a lot of good reasons to keep it from happening, so I'm not going to gripe about getting wall-crashers. My shop lost a Tektronix 7904A* High-speed Storage scope to a beam hitting a wall. It had been borrowed to use for magnet alignment work on the Los Alamos Meson Physics Facility and was beside the beam-line when the beam wandered. They told us it would be 2 years before it had cooled down enough to use again. $24k down the tubes and no compensation for us because it was still functional...

Is there any way to clean out the 0 credit claims so they're not cluttering up the lists, or does that involve too much work to be worth the resources?

(* At least I think the model was a 7904A - it's been 20 years...)
ID: 19291 · Report as offensive     Reply Quote
Profile Paul D. Buck

Send message
Joined: 2 Sep 04
Posts: 545
Credit: 148,912
RAC: 0
Message 19336 - Posted: 10 Apr 2008, 20:19:28 UTC

It is kind of the point of the work we are doing to test the system model to prevent the actual machine from experiencing these problems ...

In our computer it is just a short Task (and it can happen to any of the three lengths of assigned task) so a one million turn task only goes around 2 or 3 times ...

I just wish I could do more work ... :)
ID: 19336 · Report as offensive     Reply Quote
jaysback

Send message
Joined: 29 Sep 04
Posts: 3
Credit: 284,021
RAC: 0
Message 19347 - Posted: 11 Apr 2008, 9:39:46 UTC

Knocking these 0.00 credits straight out of the pending credit would be great.
I don't mind running the w/u's but the long list of pendings is a bit of a pain and must fill your (LHC's) server up a bit.


Phanteks Luxe 2, AMD 3700x, MSI 2070 super X trio, 32gb Corsiar vengance RGB pro, watercooled.
ID: 19347 · Report as offensive     Reply Quote

Message boards : Number crunching : Short-running WU ( 13 to 30 seconds) Is this OK???


©2024 CERN