1) Message boards : Number crunching : MT Nbody 1.62 workunit locked up (Message 65064) Posted 22 Aug 2016 by Kwartet! Post: Also got some of those on Linux (Ubuntu): task wu computer 1739366241 1277165096 698989 1739366264 1274864663 698989 1739367380 1277019122 698989 1737734321 1275223469 698989 1736835394 1259652472 698989 1736836444 1275252939 698989 1736835522 1274974287 698989 1736835532 1271470693 698989 1733987832 1271863723 698989 1730688175 1271147729 698989 1725242958 1267766875 698989 1723391303 1266509406 698989 Exit status 196 (0xc4) EXIT_DISK_LIMIT_EXCEEDED Stderr output: <core_client_version>7.6.31</core_client_version> <![CDATA[ <message> Disk usage limit exceeded </message> <stderr_txt> <search_application> milkyway_nbody 1.62 Linux x86_64 double OpenMP, Crlibm </search_application> Using OpenMP 7 max threads on a system with 8 processors </stderr_txt> ]]> Milkyway assumes a disk limit of 50.0 MB, exceeds this limit (two tasks still in my event log: 77.57, 117.54 MB) and aborts those tasks. There's still almost 20 GB available to BOINC.

2) Message boards : Number crunching : threads per gpu? (Message 64968) Posted 2 Aug 2016 by Kwartet! Post: Berkeley, the makers of BOINC software, have this to say about client configuration files https://boinc.berkeley.edu/wiki/Cc_config.xml I have a cc_config in my W7 installation, but none on my other PC running w10. I most definitely did not type that myriad of configuration options! But I did once edit a line, maybe two. No cc_config in W10? Didn't realize that, seems to run OK. Further down in the Berkeley Wiki, app_config.xml is described. I don't do anything with app versions. Best regards

3) Message boards : Number crunching : threads per gpu? (Message 64961) Posted 31 Jul 2016 by Kwartet! Post: Place a file called app_config.xml in MW's data directory (Windows 7~ C:\ProgramData\BOINC\projects\milkyway.cs.rpi.edu_milkyway). Contents as follows. If you want to edit, save as ANSI; other encodings may not be read correctly. <app_config> <app> <name>milkyway</name> <gpu_versions> <gpu_usage>0.33</gpu_usage> <cpu_usage>0.33</cpu_usage> </gpu_versions> </app> <app> <name>milkyway_separation__modified_fit</name> <gpu_versions> <gpu_usage>0.25</gpu_usage> <cpu_usage>0.25</cpu_usage> </gpu_versions> </app> </app_config> This directs MW to run 3 or 4 units in parallel, respectively.

4) Message boards : Number crunching : Run Multiple WU's on Your GPU (Message 64186) Posted 17 Dec 2015 by Kwartet! Post: What's wrong with rereading the configuration? Works for me. Boinc Manager > Options > Read configuration. I suppose that's the item in English, my Manager spells them in Dutch.

5) Message boards : Number crunching : New Benchmark Thread - times wanted for any hardware, CPU or GPU, old or new! (Message 64145) Posted 3 Dec 2015 by Kwartet! Post: Benchmark instructions: http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=3551&postid=63837 Basically: run some wu's, find their duration in the tasks overview in your account data, compute average runtime. Find more results here: http://forums.anandtech.com/showthread.php?t=2384985

6) Message boards : Number crunching : New Benchmark Thread - times wanted for any hardware, CPU or GPU, old or new! (Message 64144) Posted 3 Dec 2015 by Kwartet! Post: HD7950 Boost 900/1400 MHz will overclock to 1100/1400, but not when running MW 31,2 secs Now back to running 2 parallel, giving better throughput 58,3/2 = 29,1 secs Doesn't vary too much, average of 18 timings, I left out 2 suspiciously fast ones (52 and 44 secs). i5-3570 automagically overclocks to 3,6 GHz (MSI utility) Keeping one core free for GPU tasks Radeon 15.11.1 (Crimson) Win7 Pro

7) Message boards : Number crunching : Lowering required cpus (Message 62906) Posted 29 Dec 2014 by Kwartet! Post: 45 seconds is good for two wu's.

8) Message boards : Number crunching : Lowering required cpus (Message 62901) Posted 28 Dec 2014 by Kwartet! Post: Hi Astromancer, App_info is ye olde waye; these days it's possible to use an app_config.xml file in the project's data directory. Mine looks like this (under Windows, save as ANSI, not UTF): <app_config> <app> <name>milkyway</name> <gpu_versions> <gpu_usage>0.5</gpu_usage> <cpu_usage>0.05</cpu_usage> </gpu_versions> </app> <app> <name>milkyway_separation__modified_fit</name> <gpu_versions> <gpu_usage>0.5</gpu_usage> <cpu_usage>0.05</cpu_usage> </gpu_versions> </app> </app_config> On a HD7950, i5-3570, W7-64 bit, other BOINC tasks claiming the cpu, my run times are between 45 and 150 seconds, depending on application (modified fit or not).

©2024 Astroinformatics Group