Posts by Jean-David Beyer

1) Message boards : Number crunching : NBody tasks taking much longer ... (Message 74134) Posted 9 Sep 2022 by Jean-David Beyer Post: On my machine, Nbody tasks usually have been taking 4 to 8 minutes. This morning one has about 2 1/2 hours on it and seems to be running normally. boinc 325569 16183 99 07:33 ? 02:49:19 ../../projects/milkyway.cs.rpi.edu_milkyway/milkyway_nbody_1.82_x86_64-pc-linux-gnu__mt -f nbody_parameters.lua -h histogram.txt --seed 196001641 -np 11 -p 3.58248 1 0.117433 0.148923 56.652 0.301723 1 1 1 1 1 This is not a complaint, but it sure is a surprise.
2) Message boards : Number crunching : Validation inconclusive (Message 73674) Posted 24 May 2022 by Jean-David Beyer Post: I do not know if anything is getting validated. This is my oldest. There are 184 others. Workunit 458806994 name de_modfit_86_bundle5_3s_south_pt2_2_1652888500_3384632 application Milkyway@home Separation created 21 May 2022, 6:48:25 UTC minimum quorum 1 initial replication 1 max # of error/total/success tasks 2, 9, 6 validation Pending Task Computer Sent Time reported explain Status Run time CPU time Credit Application 279025918 928280 22 May 2022, 8:24:29 UTC 22 May 2022, 23:13:27 UTC Completed, waiting for validation 2,791.94 2,777.26 pending Milkyway@home Separation v1.46 x86_64-pc-linux-gnu
3) Message boards : Number crunching : I do not understand waiting for validation (Message 73547) Posted 16 May 2022 by Jean-David Beyer Post: Now, the "Quorum 1, initial replication 1" is just the way a project using adaptive replication should launch new work units; when the initial task is returned and the validator has looked at it, if the result is not obviously invalid and if the system that returned it has a good run of consecutive validated returns there's a significant likelihood that it will validate without needing a second opinion - there's a random factor involved, and it'll allow most tasks to self-validate. If the system doesn't have a run of validated results, or if the random factor decrees that a second opinion is needed, the validator will request a second task (and the quorum and/or replication counts will increment.) ... Hope this helps... That is probably the reason then. I am a new user of Milkyway and while all my results seem to me to be correct, perhaps a day or two of results are not enough to go through self-validation even though they say a quorum of one is the requirement. I notice I now have no work units in this category.
4) Message boards : Number crunching : I do not understand waiting for validation (Message 73539) Posted 16 May 2022 by Jean-David Beyer Post: Workunit 452113339 name de_modfit_84_bundle5_3s_south_pt2_2_1651669798_11233861 application Milkyway@home Separation created 15 May 2022, 9:34:34 UTC minimum quorum 1 initial replication 1 max # of error/total/success tasks 2, 9, 6 validation Pending Task click for details Computer Sent Time reported or deadline explain Status Run time (sec) CPU time (sec) Credit Application 271031951 928280 15 May 2022, 9:43:24 UTC 15 May 2022, 23:26:13 UTC Completed, waiting for validation 2,773.68 2,758.37 pending Milkyway@home Separation v1.46 x86_64-pc-linux-gnu What validation do rhey want when minimum quorum 1 initial replication 1 max # of error/total/success tasks 2, 9, 6 validation Pending
5) Message boards : News : Nbody WU Flush (Message 73535) Posted 15 May 2022 by Jean-David Beyer Post: On my linux machine, I get similar results: all are 64-bit. Red Hat Enterprise Linux release 8.6 (Ootpa) This is a client machine, not a server. [/usr/bin]$ file boinc* boinc: ELF 64-bit LSB shared object, x86-64, version 1 (GNU/Linux), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 3.2.0, BuildID[sha1]=4a444f877bad66d3e3c69815b610c2f44dec51c0, stripped boinc_client: ELF 64-bit LSB shared object, x86-64, version 1 (GNU/Linux), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 3.2.0, BuildID[sha1]=4a444f877bad66d3e3c69815b610c2f44dec51c0, stripped boinccmd: ELF 64-bit LSB shared object, x86-64, version 1 (GNU/Linux), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 3.2.0, BuildID[sha1]=004ad0d0739c3b4a05a6afd9e5f716f730d63226, stripped boincmgr: ELF 64-bit LSB shared object, x86-64, version 1 (GNU/Linux), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 3.2.0, BuildID[sha1]=30032e4e8cb1a106b666ce63ac8b11e722fccb82, stripped boincscr: ELF 64-bit LSB shared object, x86-64, version 1 (GNU/Linux), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 3.2.0, BuildID[sha1]=1b24af4798f6e15af7d5fd66681355a7005d82ad, stripped [/usr/lib64]$ file ld-linux* ld-2.28* ld-linux-x86-64.so.2: symbolic link to ld-2.28.so ld-2.28.so: ELF 64-bit LSB shared object, x86-64, version 1 (GNU/Linux), dynamically linked, BuildID[sha1]=67aa0f1504abcfc4befb2b61a329a30a9984e0db, with debug_info, not stripped [/usr/lib64]$ file boinc libboinc_api.so.7: symbolic link to libboinc_api.so.7.16.11 libboinc_api.so.7.16.11: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked, BuildID[sha1]=b534348dcb43689c3eaacc94f317305a5be2a982, stripped libboinc_graphics2.so.7: symbolic link to libboinc_graphics2.so.7.16.11 libboinc_graphics2.so.7.16.11: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked, BuildID[sha1]=900728d6af6e7c77e639e569d657326824973d4d, stripped libboinc_opencl.so.7: symbolic link to libboinc_opencl.so.7.16.11 libboinc_opencl.so.7.16.11: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked, BuildID[sha1]=011ba4e89c630388ff1694e4cc56db3d55b53fe7, stripped
6) Questions and Answers : Unix/Linux : How do tasks like de_nbody_08_31_2021_v176_40k__data__13_1647295263_3642347_2 work? (Message 73507) Posted 14 May 2022 by Jean-David Beyer Post: You're guessing correctly, the task is probably using 8 cores. N-Body Simulation project is multicore and can be set up to use 1 to 16 cores per task via app_config.xml. By default each task will use all available cores (up to 16). BOINC will even sometimes pause other running tasks, like you've noticed. I'm currently running this project on Ubuntu 20.04 (via WSL2 on Windows 10) and both top and htop commands also give the same reading as you describe. It seems to be the (strange) way they interpret multicore usage. I found that 3-5 core setup produces highest throughputs. Here's the app_config.xml that I have to run it using 4 cores. Thank-you. I set it to run with only 4 cores and it is now working that way. I am not sure what the optimum is. I am now running # perf stat -aB -e cache-references,cache-misses to see how the processor cash is doing. So I am by no means running out of processor cache. This is unusually good number. When ClimatePrediction is running big (N216) models, the processor cache missis go up to about 50%, but now only milkyway and universe@home are the only Boinc tasks (and not much else) that are running. Performance counter stats for 'system wide': 10,736,725,787 cache-references 122,353,406 cache-misses # 1.140 % of all cache refs 60.247759271 seconds time elapsed
7) Questions and Answers : Unix/Linux : How do tasks like de_nbody_08_31_2021_v176_40k__data__13_1647295263_3642347_2 work? (Message 73501) Posted 14 May 2022 by Jean-David Beyer Post: I am running a machine using Red Hat Enterprise Linux release 8.5 (Ootpa). When my machine runs a task like de_nbody_08_31_2021_v176_40k__data__13_1647295263_3642347_2, it seems to run just fine, and sends back up to the server where it is marked valid. But if I run the top command to look at only Boinc tasks, that kind of task seems to be assigned to one CPU, but that CPU seems to run at 650% (or so) of capacity, and the other Boinc tasks go to sleep. I get the impression that it is using six to eight processors, but top does not say so. So this is not a complaint, but I would like to know what is really going on.