Posts by Azmodes

1) Message boards : Number crunching : Um, Got a Couple of Whoppers? (Message 1582)
Posted 3 Mar 2022 by Azmodes
Post:
I'm seeing a lot of multi-hour-long GPU WUs lately. I have one that's been running for 17 hours on an RTX 2060 (three tasks in tandem, admittedly) and it's barely at 50%. Is this to be expected more and more going forward? Any plans to update the credit system to address the task size disparity?
2) Message boards : Bug tracker : Supermassive WUs all ending in computation errors (Message 1359)
Posted 4 Mar 2020 by Azmodes
Post:
oh, I assumed it had because the server status recently switched to "part 2". Guess that's not entirely the same thing?
3) Message boards : Bug tracker : Supermassive WUs all ending in computation errors (Message 1357)
Posted 4 Mar 2020 by Azmodes
Post:
Are we sure this was fixed with the new batch? I'm seeing WUs like this again...

https://sech.me/boinc/Amicable/result.php?resultid=29983425
https://sech.me/boinc/Amicable/result.php?resultid=30017485
https://sech.me/boinc/Amicable/result.php?resultid=30017486
4) Message boards : Bug tracker : Supermassive WUs all ending in computation errors (Message 1354)
Posted 16 Feb 2020 by Azmodes
Post:
I see, thanks.
5) Message boards : Bug tracker : Supermassive WUs all ending in computation errors (Message 1352)
Posted 14 Feb 2020 by Azmodes
Post:
See here.

I checked some other users' hosts and I can't find a single one of these large tasks getting successfully finished. It seems they either end in errors or are aborted because of their bizarre runtime.

It would be nice not having to choose between constant micromanaging my crunchers and a lot of wasted time and energy. Any developments of comments on this?
6) Message boards : Number crunching : GPU version: kernel size tuning and less UI lags (Message 1344)
Posted 1 Feb 2020 by Azmodes
Post:
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?
7) Message boards : Number crunching : GPU version: kernel size tuning and less UI lags (Message 1318)
Posted 6 Dec 2019 by Azmodes
Post:
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)
8) Message boards : Number crunching : GPU version: kernel size tuning and less UI lags (Message 1312)
Posted 1 Dec 2019 by Azmodes
Post:
Ah well, both ended in a compute error after half a day...
9) Message boards : Number crunching : GPU version: kernel size tuning and less UI lags (Message 1311)
Posted 30 Nov 2019 by Azmodes
Post:
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?
10) Message boards : Number crunching : GPU version: kernel size tuning and less UI lags (Message 1310)
Posted 30 Nov 2019 by Azmodes
Post:
Probably because task size is 68*1012, 100 times more than usual. Task generator was definitely not perfect.

I guess I got another one of these. One of my GPU tasks has been running for almost three hours now and it's 9.5% done. I'll leave it running.
11) Message boards : Number crunching : GPU version: kernel size tuning and less UI lags (Message 1304)
Posted 14 Nov 2019 by Azmodes
Post:
Huh. Why are these so exceptionally long?
12) Message boards : Number crunching : GPU version: kernel size tuning and less UI lags (Message 1301)
Posted 14 Nov 2019 by Azmodes
Post:
I guess so.

Another thing, I'm running two tasks at once on this RTX 2080 and although it runs well (and throughput improves) every once in a while a task gets stuck and runs endlessly. Trying to reduce kernel size to 22 and see if it happens again.
13) Message boards : Number crunching : GPU version: kernel size tuning and less UI lags (Message 1298)
Posted 12 Nov 2019 by Azmodes
Post:
When looking at averaged task times, I noticed that my GTX 1650 appears to be at least as productive as my 1080 Ti, if not even a tad better (both 1.1-1.2 million cr/day). I have both running one task at a time, kernel size 23. Is Turing really that superior or am I doing something wrong with the 1080 Ti? The Ti is even running on a DDR4 Linux system, the 1650 only DDR3 and Windows. Both have ample RAM available and core load is close to 100%.
14) Message boards : Bug tracker : Constant Computation errors (Message 1297)
Posted 10 Nov 2019 by Azmodes
Post:
I put another 8 gigs of RAM in there and now it seems to run without a hitch.
15) Message boards : Bug tracker : Constant Computation errors (Message 1277)
Posted 30 Oct 2019 by Azmodes
Post:
I increased the size to 17 GB (I have two GPUs), still getting the same error.

EDIT: Nevermind, forgot to reboot. However, the system is now nigh-on unusable with lags, even though the screen is connected to the onboard graphics and not any of the cards (??). Kernel size at 21, GPU usage low to non-existent.

EDIT2: And still ending in computation errors.

Amicable/OpenCL.cpp, line 215: Error: boinc_get_opencl_ids() failed with error -1
16) Message boards : Bug tracker : Constant Computation errors (Message 1275)
Posted 29 Oct 2019 by Azmodes
Post:
What's the reason for this error?

Amicable/OpenCL.cpp, line 323: clCreateContext returned error -6
17) Message boards : Number crunching : Countermeasures for Increased CPU Time (Message 1117)
Posted 7 Apr 2019 by Azmodes
Post:
Anyone got a recommended configuration for RTX cards, number of tasks- and kernel-wise? I'm currently trying 10 tasks at once, with the lowest kernel size, but it's not even at 50% utilization.
18) Message boards : Number crunching : Why are the RTX Graphics Cards now really Poor at running this Project. (Message 1064)
Posted 20 Jan 2019 by Azmodes
Post:
I tried 8 tasks on my 2080, each with a full CPU core allocated, but core load was barely over 50% IIRC. Adding more caused a BSOD, so I assume I should reduce the kernel size, but I can't experiment with that right now. Anyone have a working configuration they'd like to share?
19) Message boards : Number crunching : Countermeasures for Increased CPU Time (Message 1004)
Posted 28 Nov 2018 by Azmodes
Post:
Hi Azmodes,

As far as I know there is no way to tell BOINC to send different numbers of tasks to different GPUs, but this may be possible with the cc_config file, which you can use to specify which GPUs are used for specific jobs or at all. I am not familiar with the cc_config files though. Here is the link to the cc_config page for BOINC though in case you want to try things out; [url]https://boinc.berkeley.edu/wiki/Client_configuration [/url]

Good luck!

I'm aware of that link, but thanks all the same. The workaround I did earlier today is simply to exclude the GTX from AN and use it for another project with no lag (Collatz).
20) Message boards : Number crunching : Countermeasures for Increased CPU Time (Message 997)
Posted 27 Nov 2018 by Azmodes
Post:
There is no way to tell BOINC to use different app_configs for different cards on the same host, right? For instance, one of my machines has an RTX 2070 and two GTX 1060 3GB and I'm getting screenlag with two tasks at the same time (screen connected to one of the GTXs), so it would be ideal to have the RTX crunch two and the rest only one.

Is it really necessary to assign a full CPU core to each task even with the decrease in usage more tasks entail?



©2024 Sergei Chernykh