Advanced search

Message boards : News : [wildlife] EXACT v0.25 and v0.26

Author Message
Travis Desell
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 16 Jan 12
Posts: 1793
Combined Credit: 2,265,607
DNA@Home: 293,563
SubsetSum@Home: 349,212
Wildlife@Home: 1,622,832
Wildlife@Home Watched: 212,926s
Wildlife@Home Events: 51
Climate Tweets: 21
Images Observed: 710

              
Message 7064 - Posted: 21 May 2017, 22:28:24 UTC

There was a bug I missed in the EXACT v0.25 client which will cause it to crash from over allocating memory. Please abort any WUs running with this version ASAP so they don't eat up all your machines memory. I really apologize, it ran correctly for my local tests but I had missed a call to convert one of the input filenames to it's proper location in the client.

That being said, v0.26 should be running fine. I made quite a few improvements to the code, so now it should require about 1/4th the memory as the v0.24 client (so the CIFAR work units will now take less memory that the MNIST work units from v0.24 and before). It's also running ~2x faster. That and I upgraded the back propagation algorithm to use batch normalization which also means we'll be getting better results from it. Pretty excited to see how it runs.

mmonnin
Send message
Joined: 31 May 16
Posts: 25
Combined Credit: 19,817,721
DNA@Home: 0
SubsetSum@Home: 1,023,200
Wildlife@Home: 18,794,520
Wildlife@Home Watched: 0s
Wildlife@Home Events: 0
Climate Tweets: 54
Images Observed: 0

      
Message 7070 - Posted: 22 May 2017, 22:23:44 UTC - in response to Message 7064.

Trying out the new app flavor of the week.

Travis Desell
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 16 Jan 12
Posts: 1793
Combined Credit: 2,265,607
DNA@Home: 293,563
SubsetSum@Home: 349,212
Wildlife@Home: 1,622,832
Wildlife@Home Watched: 212,926s
Wildlife@Home Events: 51
Climate Tweets: 21
Images Observed: 710

              
Message 7071 - Posted: 23 May 2017, 0:17:36 UTC - in response to Message 7070.

Trying out the new app flavor of the week.


Hope it tastes good. :) Should be quite a bit faster and use a lot less memory.

mmonnin
Send message
Joined: 31 May 16
Posts: 25
Combined Credit: 19,817,721
DNA@Home: 0
SubsetSum@Home: 1,023,200
Wildlife@Home: 18,794,520
Wildlife@Home Watched: 0s
Wildlife@Home Events: 0
Climate Tweets: 54
Images Observed: 0

      
Message 7072 - Posted: 23 May 2017, 0:47:24 UTC - in response to Message 7071.
Last modified: 23 May 2017, 0:49:07 UTC

Trying out the new app flavor of the week.


Hope it tastes good. :) Should be quite a bit faster and use a lot less memory.


1st 3 done/validated so far.

Comparing to MNIST CNN those were getting 0.096 points per CPU second across 124 tasks. This is getting 0.125. Small sample so far. The highest MNIST CNN was 0.122.

Tastes good. :)

145-150mb so far. 2P 2670v1 in Ubuntu.

ARCHspark
Send message
Joined: 22 May 17
Posts: 5
Combined Credit: 472,305
DNA@Home: 0
SubsetSum@Home: 0
Wildlife@Home: 472,305
Wildlife@Home Watched: 0s
Wildlife@Home Events: 0
Climate Tweets: 0
Images Observed: 0

  
Message 7075 - Posted: 24 May 2017, 19:53:20 UTC - in response to Message 7071.

Trying out the new app flavor of the week.


Hope it tastes good. :) Should be quite a bit faster and use a lot less memory.


it's WAY faster (thought it's just smaller WUs) and around ~150MB/WU

OT: how long does the validation take? my other account still has v0.24 WUs pending for validation that are up to 7 Days old...

Travis Desell
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 16 Jan 12
Posts: 1793
Combined Credit: 2,265,607
DNA@Home: 293,563
SubsetSum@Home: 349,212
Wildlife@Home: 1,622,832
Wildlife@Home Watched: 212,926s
Wildlife@Home Events: 51
Climate Tweets: 21
Images Observed: 710

              
Message 7077 - Posted: 25 May 2017, 15:12:59 UTC - in response to Message 7075.

Trying out the new app flavor of the week.


Hope it tastes good. :) Should be quite a bit faster and use a lot less memory.


it's WAY faster (thought it's just smaller WUs) and around ~150MB/WU

OT: how long does the validation take? my other account still has v0.24 WUs pending for validation that are up to 7 Days old...


Validation is kind of up to your wing(wo)men. Basically the validator waits for other users to report the same workunit then compares the results. So if the other users are slow reporting their WUs then validation is going to take longer.

GLeeM
Send message
Joined: 1 Jul 13
Posts: 118
Combined Credit: 33,720,144
DNA@Home: 28,994
SubsetSum@Home: 231,079
Wildlife@Home: 33,460,071
Wildlife@Home Watched: 3,888,714s
Wildlife@Home Events: 628
Climate Tweets: 0
Images Observed: 0

          
Message 7080 - Posted: 27 May 2017, 0:25:46 UTC - in response to Message 7077.

Don't think anything needs fixing, just an observation.

Your new calculation that extends the WU deadline to a nice date, I think, also will cause newly assigned WUs to run before a WU that was assigned many days ago.

I noticed this when an almost finished older WU (deadline was well into June) was paused so a newly downloaded WU with a deadline date of a couple days away could start.


Post to thread

Message boards : News : [wildlife] EXACT v0.25 and v0.26