Welcome to MilkyWay@home

Posts by boinc127

1) Message boards : Number crunching : 2 NBody workunits stalled out so far. (Message 69529)
Posted 6 Feb 2020 by boinc127
Post:
I've had 2 out of 3 NBody mt workunits stall out so far. I'm not sure what the issue is at my end or if there is a problem with the workunits.

https://milkyway.cs.rpi.edu/milkyway/result.php?resultid=477905636
https://milkyway.cs.rpi.edu/milkyway/result.php?resultid=478037717

Both workunits stop reporting progress and the CPU usage drops to zero (they are supposed to use 3 cores). I aborted both workunits. Incidentally, both of these workunits are from today's run. The workunit that validated was from the 02/03/2020 run.
2) Questions and Answers : Web site : How do I choose Nbody workunits? (Message 68553)
Posted 16 Apr 2019 by boinc127
Post:
There is no option to choose between separation runs and Nbody workunits in the project preferences page. How do I choose Nbody workunits?
3) Questions and Answers : Web site : Website bugs since server transition (Message 68411)
Posted 27 Mar 2019 by boinc127
Post:
I've noticed in my account settings I can't choose between Nbody workunits or separation workunits. Right now it's a moot point since there are no Nbody workunits as I'm posting this.
4) Message boards : News : Validation Inconclusive Errors (Message 67014)
Posted 30 Jan 2018 by boinc127
Post:
I've also been having the issue with nbody multiple CPU tasks. I usually use 3 CPUs and will leave the tasks to compute on their own but sometimes I find the tasks get "stuck" and won't finish (they would be running for hours with no progress). I really haven't found a pattern of why they are behaving that way. I've also stopped computing those tasks because I find tying up 3 processors only for the task to get stuck is quite wasteful.
5) Message boards : Application Code Discussion : Source Code and Optimized FreeBSD Binaries (Message 66214)
Posted 23 Feb 2017 by boinc127
Post:
I tried compiling nbody a while back... basically FreeBSD doesn't implement cpow() in its msun library, and I was never sure how to write the function into the nbody code properly, so that it would compute accurately.

BTW, the Linux 32 bit version of separation does work on FreeBSD with the linuxulator (and it uses the SSE 4.1 path as well). I'm not sure about 64 bit Linux though, I've tried 64 bit Linux binaries before and have never been able to get ANY 64 bit Linux BOINC projects to work with the linuxulator. They all segmentation fault and error out after a few seconds.
6) Message boards : Application Code Discussion : Source Code and Optimized FreeBSD Binaries (Message 63315)
Posted 2 Apr 2015 by boinc127
Post:
Ya i've had some that validate also. I don't see any kind of recognizable pattern in it either. Its always hard to tell with Milkyway@home because so many workunits come back validation inconclusive, only to be validated later on. Of course it could also be my setup. One of the many risks following a stable FreeBSD branch.
7) Message boards : Application Code Discussion : Source Code and Optimized FreeBSD Binaries (Message 63313)
Posted 2 Apr 2015 by boinc127
Post:
Has anyone else had any issues running the FreeBSD Nbody 1.48 binary? I've gotten 8 validate errors with them so far... they aren't invalids, but they aren't validating. I've switched back over to the linux 32 bit (mt) binary and the issue goes away.

Perhaps there's some weird regression with LLVM or clang thats botching the workunits up?

EDIT: I've noticed some of these same issues with Darwin hosts when the workunit is sent to another host to check my inconclusives.
8) Message boards : Application Code Discussion : Source Code and Optimized FreeBSD Binaries (Message 63272)
Posted 26 Mar 2015 by boinc127
Post:
Thanks so much for the binaries. It works like a charm.
9) Message boards : Application Code Discussion : Source Code and Optimized FreeBSD Binaries (Message 62783)
Posted 5 Dec 2014 by boinc127
Post:
Wow those binaries are much faster, possibly faster than running the linux ones. I'm still trying to tackle compiling the binaries myself, as a good learning tool, but have hit a brick wall with undefined references galore while trying to compile separation. What a headache.
10) Message boards : Application Code Discussion : Source Code and Optimized FreeBSD Binaries (Message 62744)
Posted 27 Nov 2014 by boinc127
Post:
Hey quick question for you, what did you compile the NBody app with, clang or gcc? I haven't had a chance to test the new programs yet, I've been stuck doing work with Windows (bleck), but I wasn't sure if the NBody app is able to multithread with more than one core or thread. Just curious.
11) Message boards : Application Code Discussion : Source Code and Optimized FreeBSD Binaries (Message 62729)
Posted 23 Nov 2014 by boinc127
Post:
That's really great. I'm looking forward to using the new apps as soon as I restart my computer into FreeBSD. I was getting frustrated trying to understand all of the cmake errors and I got stuck when it reported the BOINC library files were too old. I'm glad to see you got them working.

I was using the Linux 32 bit versions of nbody (1.44) and separation modified fit (1.36) and they ran fine. I know that separation found the sse4.1 code path with no issues. And I just put a copy of the libgomp library file in my milkyway project directory to get nbody working ok. But native FreeBSD is much better. I'm always happy to represent FreeBSD while crunching BOINC projects, and this will go even farther.

Thanks a lot!
Jeremy
12) Message boards : Application Code Discussion : Source Code and Optimized FreeBSD Binaries (Message 62727)
Posted 23 Nov 2014 by boinc127
Post:
If you could do that that would be totally awesome! I was just about ready to attempt to compile from some source code, I believe from

https://github.com/Milkyway-at-home/milkywayathome_client

but honestly, I don't have much experience compiling from source code for FreeBSD. Using cmake makes it tremendously easier to figure out how to fix issues I have come across, but my lack of experience has made my endeavor substantially slower. I was actually about to give up and switch over to the linux binary, which I believe would also be much faster. I know when running the linux binary separation (modified fit) program through FreeBSD it does pick up the SSE4.1 intrinsics.

Thanks
Jeremy
13) Message boards : News : New Version of Separation Modified Fit (1.32) (Message 62315)
Posted 12 Sep 2014 by boinc127
Post:
I had some workunits error out. I suppose the server queue still has some workunits that haven't reached their max error #...

ps_modfit_16TestStars_1s_132_wrap_2_1405680903_9442119_1
ps_modfit_16TestStars_1s_132_wrap_2_1405680903_9442124_1
de_modfit_16TestStars_1s_132_wrap_2_1405680903_9441944_1
ps_modfit_16TestStars_1s_132_wrap_2_1405680903_9442123_1
ps_modfit_16TestStars_1s_132_wrap_2_1405680903_9442197_1
ps_modfit_16TestStars_1s_132_wrap_2_1405680903_9442196_1
ps_modfit_16TestStars_1s_132_wrap_2_1405680903_9442122_1
de_modfit_16TestStars_1s_132_wrap_2_1405680903_9418202_1

bad astronomy_parameters.txt file.

Also... is the ver. 1.32 Nvidia OpenCL GPU app for Windows 64 bit the same as the ver. 1.30 app? There doesn't seem to be a ver 1.32 app for Windows 64 bit Nvidia.
14) Message boards : Number crunching : hey quick question (Message 46619)
Posted 18 Mar 2011 by boinc127
Post:
i'm new here, i got a dual core intel 2.66GHz [Family 6 Model 15 Stepping 6] processors. i just ran the work unit # 255014325 called de_separation_10_3s_fix_5_1610326_1300391361_0. i don't have a double precision gpu, otherwise i would use it. i'm running on pure cpu power here.

my goodness it seemed this work unit kicked my rigs ass. the last couple of work units got crunched relatively quickly. but the initial "to completion" time for this work unit was 4 hours, but it ended up taking over 11 hours. and then it returned "verified but no consensus yet". are many of these work units suited to work with double precision gpu's? or are these computations very complex and take hours to accomplish? it seemed like as i was watching the "to completion" time, it kept increasing, not decreasing. i already tell boinc to use 100% of my cpus and 85 to 90% cpu time. i know the computer's working because i monitor the temperature carefully (i don't wanna burn my rig out running these programs). opinions anyone?




©2021 Astroinformatics Group