GPU version: kernel size tuning and less UI lags

Message boards : Number crunching : GPU version: kernel size tuning and less UI lags

To post messages, you must log in.

Previous · 1 · 2 · 3

AuthorMessage
Azmodes

Send message
Joined: 5 Mar 17
Posts: 20
Credit: 3,017,841,792
RAC: 0
   
Message 1311 - Posted: 30 Nov 2019, 21:02:15 UTC
Last modified: 30 Nov 2019, 21:03:24 UTC

And another.

https://sech.me/boinc/Amicable/workunit.php?wuid=12033885
https://sech.me/boinc/Amicable/workunit.php?wuid=12033891

So shouldn't these colossal things yield more than the usual amount of credit?
ID: 1311 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Azmodes

Send message
Joined: 5 Mar 17
Posts: 20
Credit: 3,017,841,792
RAC: 0
   
Message 1312 - Posted: 1 Dec 2019, 10:16:46 UTC

Ah well, both ended in a compute error after half a day...
ID: 1312 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
pututu

Send message
Joined: 6 Feb 17
Posts: 6
Credit: 107,241,610
RAC: 0
   
Message 1313 - Posted: 1 Dec 2019, 20:21:08 UTC - in response to Message 1305.  
Last modified: 1 Dec 2019, 20:23:12 UTC

I randomly sampled the tasks that I had completed over the past 3 days with varying run time and task size. Though not perfect, it seems to indicate that there is a good correlation between task size and run time. All tasks were completed on the same GPU. BTW, I"m running two tasks per GPU and the GPU run time shown in the graph is not normalized to one task. There is one outlier but if we ignore this single point, the regression-squared or R-squared increased from 52% to 95%.

Wondering if there is any chance to allocate higher credit points for larger task size? Some folks don't care about credits but just my one cent to keep things within reasonable perspective. Perhaps allocating task size of 1011, 1012 and 1013 with different fixed credit buckets?

https://postimg.cc/gx5J8Rg1
ID: 1313 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Sergei Chernykh
Project administrator
Project developer

Send message
Joined: 5 Jan 17
Posts: 506
Credit: 72,451,573
RAC: 0
   
Message 1314 - Posted: 2 Dec 2019, 10:13:52 UTC - in response to Message 1313.  

Yes, I'll adjust credit for these tasks.
ID: 1314 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
pututu

Send message
Joined: 6 Feb 17
Posts: 6
Credit: 107,241,610
RAC: 0
   
Message 1315 - Posted: 2 Dec 2019, 15:29:29 UTC - in response to Message 1314.  

Thank you sir!
ID: 1315 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Azmodes

Send message
Joined: 5 Mar 17
Posts: 20
Credit: 3,017,841,792
RAC: 0
   
Message 1318 - Posted: 6 Dec 2019, 10:07:25 UTC
Last modified: 6 Dec 2019, 10:35:05 UTC

I'm only getting compute errors for the longer tasks. Is kernel size to blame? Quite a bit of crunching time wasted.

exceeded elapsed time limit 28237.96 (10000000.00G/341.91G)
ID: 1318 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Jozef J

Send message
Joined: 24 Jan 17
Posts: 20
Credit: 1,180,716,010
RAC: 447
   
Message 1319 - Posted: 8 Dec 2019, 10:25:29 UTC

Hi, i see ,amicable pairs discovering are bit stall, or go slowly now.. We get to another phase" like before?
ID: 1319 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Jozef J

Send message
Joined: 24 Jan 17
Posts: 20
Credit: 1,180,716,010
RAC: 447
   
Message 1323 - Posted: 11 Dec 2019, 12:51:09 UTC - in response to Message 1319.  

Hi, i see ,amicable pairs discovering are bit stall, or go slowly now.. We get to another phase" like before?


Wow , now is it going good...))) I like also this project app on memory using , becouse many boinc projest dont use memory and leave memory on computers just resting..in not use.. or is not optimised to use memory on computers ,just like another factor/buffer to improve computing on tasks. . not like "only cpu cores" and end..) we need more boinc projects where willl ram size another great key factor to do tasks .
ID: 1323 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Azmodes

Send message
Joined: 5 Mar 17
Posts: 20
Credit: 3,017,841,792
RAC: 0
   
Message 1344 - Posted: 1 Feb 2020, 9:40:28 UTC - in response to Message 1318.  

I'm only getting compute errors for the longer tasks. Is kernel size to blame? Quite a bit of crunching time wasted.

exceeded elapsed time limit 28237.96 (10000000.00G/341.91G)

This is still happening. Every once in a while I get one of the much longer tasks and by now I simply abort them whenever I see them, since a) they always end in an error for me and b) credit doesn't seem to have been adjusted anyway (but feel free to prove me wrong). Any official comment on those?
ID: 1344 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Previous · 1 · 2 · 3

Message boards : Number crunching : GPU version: kernel size tuning and less UI lags


©2024 Sergei Chernykh