Developing a Multi-Threaded Benchmarking App for Linux

Message boards : Number crunching : Developing a Multi-Threaded Benchmarking App for Linux
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 · 6 · Next

AuthorMessage
Profile Mike Special Project $75 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 01
Posts: 34253
Credit: 79,922,639
RAC: 80
Germany
Message 1967848 - Posted: 29 Nov 2018, 22:58:51 UTC - in response to Message 1967843.  

No, no, no.
Last code change by Joe Segur was in 2015.
That`s just 3 years ago.
Each windows application has been hand optimized.

Ok, so where do we go to lookup the app commit changes for all the apps for those of us that don't have perfect mimetic memory.


That`s in the development section of Lunatics.
Only a few of us Lunatics have access to it.

OK, but without such knowledge of a hidden developer area that most of us don't have access to all anyone can do about the history of the development of the apps is to make guesses from the release date and any docs accompanying the app.

My guesstimate was off by a decade.


No problem Keith.
I`m always here, but i only reply usually on important things.


With each crime and every kindness we birth our future.
ID: 1967848 · Report as offensive
Profile Mike Special Project $75 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 01
Posts: 34253
Credit: 79,922,639
RAC: 80
Germany
Message 1967850 - Posted: 29 Nov 2018, 23:00:37 UTC - in response to Message 1967846.  

Way off topic but does anyone know if Joe was from NY?


Northern New York State


With each crime and every kindness we birth our future.
ID: 1967850 · Report as offensive
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 1967852 - Posted: 29 Nov 2018, 23:03:53 UTC - in response to Message 1967850.  
Last modified: 29 Nov 2018, 23:03:58 UTC

Was afraid of that.

Josef Winsor Segur
DECEASED | Keeseville, NY
ID: 1967852 · Report as offensive
Profile Mike Special Project $75 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 01
Posts: 34253
Credit: 79,922,639
RAC: 80
Germany
Message 1967853 - Posted: 29 Nov 2018, 23:06:58 UTC - in response to Message 1967852.  

Was afraid of that.

Josef Winsor Segur
DECEASED | Keeseville, NY


Thanks Zalster.
That`s what we`ve feared.


With each crime and every kindness we birth our future.
ID: 1967853 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1967863 - Posted: 29 Nov 2018, 23:41:26 UTC - in response to Message 1967848.  

No problem Keith.
I`m always here, but i only reply usually on important things.

And we do in fact appreciate that you are the goto expert in the matter of the SoG app tunings and we always defer to your expertise in this matter.

I too was afraid that was the reason why there hasn't been any sign of Joe in the forums for years. RIP.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1967863 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14649
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1967983 - Posted: 30 Nov 2018, 9:19:11 UTC - in response to Message 1967839.  

No, no, no.
Last code change by Joe Segur was in 2015.
That`s just 3 years ago.
Each windows application has been hand optimized.

Ok, so where do we go to lookup the app commit changes for all the apps for those of us that don't have perfect mimetic memory.

That`s in the development section of Lunatics.
Only a few of us Lunatics have access to it.
The discussion and explanations takes place at Lunatics, but the code itself is held in an open-source repository, as required under the GPL. It looks like Joe's last contribution was http://setisvn.ssl.berkeley.edu/trac/changeset/2896.

R.I.P.
He was one of the best.
ID: 1967983 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1968030 - Posted: 30 Nov 2018, 15:59:21 UTC - in response to Message 1967983.  

No, no, no.
Last code change by Joe Segur was in 2015.
That`s just 3 years ago.
Each windows application has been hand optimized.

Ok, so where do we go to lookup the app commit changes for all the apps for those of us that don't have perfect mimetic memory.

That`s in the development section of Lunatics.
Only a few of us Lunatics have access to it.
The discussion and explanations takes place at Lunatics, but the code itself is held in an open-source repository, as required under the GPL. It looks like Joe's last contribution was http://setisvn.ssl.berkeley.edu/trac/changeset/2896.

R.I.P.
He was one of the best.

OK, thanks for the link Richard, I was afraid the code was obfuscated to any future developer from following in his footsteps and being able to further develop the cpu code to account for newer instruction sets and cpu architectures. My guesstimate about how old the code is was way off I see.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1968030 · Report as offensive
OzzFan Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Apr 02
Posts: 15691
Credit: 84,761,841
RAC: 28
United States
Message 1968075 - Posted: 30 Nov 2018, 23:31:55 UTC - in response to Message 1967852.  

Was afraid of that.

Josef Winsor Segur
DECEASED | Keeseville, NY


That's terrible news to me. I always admired his skill and his ability to articulate technical details in understandable terms. And he never really got bogged down into the heated discussions these forums have had. He seemed to be a great man.
ID: 1968075 · Report as offensive
Profile RueiKe Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 14 Feb 16
Posts: 492
Credit: 378,512,430
RAC: 785
Taiwan
Message 1968118 - Posted: 1 Dec 2018, 7:00:23 UTC

Initial release of the benchMT tool is now available on GitHub:
https://github.com/Ricks-Lab/benchMT

I will monitor this thread for any feedback or reported issues.
GitHub: Ricks-Lab
Instagram: ricks_labs
ID: 1968118 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1968122 - Posted: 1 Dec 2018, 7:23:09 UTC

Thanks for all the work you have done on the benchmark tool, Rick. I know I will be using it a fair bit with the new Turing card going into the next project. About the only thing I wish could have been tested better was the AP apps and tasks, but with the dearth of them lately, kinda hard to do. I expect a Arecibo tape will show up eventually so I can put the apps through their paces.

I verified what I thought I had tried before and could not get the Windows work unit generator to work under Wine.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1968122 · Report as offensive
Profile Brent Norman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester

Send message
Joined: 1 Dec 99
Posts: 2786
Credit: 685,657,289
RAC: 835
Canada
Message 1968131 - Posted: 1 Dec 2018, 9:20:32 UTC - in response to Message 1968122.  

Hmm, got me thinking. I have a whole wack of AP files 7zipped from Oct 2016.
I can't remember exactly why we were looking at them at time, I think some 10% were running super fast.
I could upload a few, but unsure if recording format changed.
By a few I mean I have a 600kps upload speed and 2G of files, LOL.
ID: 1968131 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13720
Credit: 208,696,464
RAC: 304
Australia
Message 1968133 - Posted: 1 Dec 2018, 9:37:21 UTC - in response to Message 1968131.  

I could upload a few, but unsure if recording format changed.

Should be OK, APv7 applications came out around Oct 2014.
Grant
Darwin NT
ID: 1968133 · Report as offensive
Profile Brent Norman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester

Send message
Joined: 1 Dec 99
Posts: 2786
Credit: 685,657,289
RAC: 835
Canada
Message 1968222 - Posted: 1 Dec 2018, 23:03:55 UTC - in response to Message 1968131.  

I uploaded some AP files from Oct 2016 to my web server, if anyone wants them I will PM you a link to them.
ID: 1968222 · Report as offensive
Profile RueiKe Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 14 Feb 16
Posts: 492
Credit: 378,512,430
RAC: 785
Taiwan
Message 1968287 - Posted: 2 Dec 2018, 12:47:18 UTC

I have just released benchRP which will parse the output of the Windows based MB_bench_213 benchmark utility. It will parse command line arguments and convert to individual data columns with the header containing the argument name with the corresponding values in row by job. benchRP can also be used to expand the argument field of the benchMT psv file. This file format is useful for the import into analytics tools for the analysis of the sensitivity of processing time and similarity to command line tuning parameters. benchRP can be downloaded here:
https://github.com/Ricks-Lab/benchRP
GitHub: Ricks-Lab
Instagram: ricks_labs
ID: 1968287 · Report as offensive
Profile RueiKe Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 14 Feb 16
Posts: 492
Credit: 378,512,430
RAC: 785
Taiwan
Message 1969283 - Posted: 8 Dec 2018, 8:27:03 UTC

I am working on an updated version of the tool that will also output angle range and signal counts. I have used this feature to assess the current sample WUs included with the package. Here is a summary of the current work units included. Seems like it could be optimized. Let me know of any recommendations on what would be an ideal set of sample WUs for benchmarking.


GitHub: Ricks-Lab
Instagram: ricks_labs
ID: 1969283 · Report as offensive
Profile RueiKe Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 14 Feb 16
Posts: 492
Credit: 378,512,430
RAC: 785
Taiwan
Message 1969465 - Posted: 9 Dec 2018, 9:03:50 UTC

I just released a new version of benchMT:
https://github.com/Ricks-Lab/benchMT/releases/tag/v1.1.0

Changes include the following:
    Command line options can now be specified in mode lines of the BenchCFG file. Options given on the command line will override modes specified in the CFG file.
    An alternative CFG file can now be specified as a command line option.
    Signal Counts and Angle Range are now included in the psv and txt summary files.
    Remove app -device N arg if specified, since -device is automatically added based on slot assignment.
    Added --gpu_devices x,y command line option to specify which GPU devices the user would like to include in the benchmark run.
    Added a lock_file in the working directory to prevent a second occurrence of benchMT from using the same directory.
    Updated reference WUs in the WU_test/safe directory.
    Changed --ref_signals option to --std_signals for clarity.


GitHub: Ricks-Lab
Instagram: ricks_labs
ID: 1969465 · Report as offensive
Profile RueiKe Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 14 Feb 16
Posts: 492
Credit: 378,512,430
RAC: 785
Taiwan
Message 1969975 - Posted: 12 Dec 2018, 6:32:31 UTC

I just released a new version of benchMT:
https://github.com/Ricks-Lab/benchMT/releases/tag/v1.2.0

Changes include the following:
    Fixed a problem with the when lock_file was created and checked. Now placed before slot initialization.
    Fixed issue where program would exit if Reference file didn't exist. Now an error message is printed and no comparison results are printed to summary files.
    Added commmand line option --no_ref which will not create reference results when selected. This is useful for characterizing potential reference WUs.
    Added color to status display.
    Modified so that status display will not show skipped jobs (Reference data already exists).
    Updated reference WUs in the WU_test/safe directory. Still need a WU with a Gaussian signal.


GitHub: Ricks-Lab
Instagram: ricks_labs
ID: 1969975 · Report as offensive
Profile RueiKe Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 14 Feb 16
Posts: 492
Credit: 378,512,430
RAC: 785
Taiwan
Message 1973491 - Posted: 4 Jan 2019, 23:30:53 UTC

I have just released a new version of benchMT:
https://github.com/Ricks-Lab/benchMT/releases/tag/v1.3.0

Changes include the following:
    SETI MB apps are now run without a shell, using shlex to parse args for the subprocess command.
    Implemented --force_ref option to force generation of reference results, even if they already exist.
    Implemented job execution with time command. Time relevant data is written to summary and psv files.
    Added job execution error checking. Bad exit status will result in updated error fields in summary/psv files and status display.


I have implemented the use of time to execute jobs and reporting back the results from time for use in comparing application performance. There are many parameters available using time, and I have only selected a few in this release. Let me know if there are others that would be useful.


GitHub: Ricks-Lab
Instagram: ricks_labs
ID: 1973491 · Report as offensive
Profile RueiKe Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 14 Feb 16
Posts: 492
Credit: 378,512,430
RAC: 785
Taiwan
Message 1975895 - Posted: 19 Jan 2019, 5:34:36 UTC

I have just released a new version of benchMT:
https://github.com/Ricks-Lab/benchMT/releases/tag/v1.4.0

Changes include the following:
    Write run_name to the psv output file. Useful when wanting to analyze data from multiple runs.
    Include nVidia stock MB app in the distribution.
    Include error message when job fails to spawn.


GitHub: Ricks-Lab
Instagram: ricks_labs
ID: 1975895 · Report as offensive
Profile Tom M
Volunteer tester

Send message
Joined: 28 Nov 02
Posts: 5124
Credit: 276,046,078
RAC: 462
Message 1987244 - Posted: 26 Mar 2019, 14:01:48 UTC

Downloaded it. Unzipped it in my download directory. Commented out the gpu test and uncommented a 4.1 something cpu test. Figured out how to point it to where my Boinc is installed (on the Desktop) via the command line.

Ran it.
Came back with an error result. Apparently the actual app/task didn't run.

What am I missing? Do I need to create and copy it to a specific folder?

I will post what I was seeing after maintenance.

Tom
A proud member of the OFA (Old Farts Association).
ID: 1987244 · Report as offensive
Previous · 1 · 2 · 3 · 4 · 5 · 6 · Next

Message boards : Number crunching : Developing a Multi-Threaded Benchmarking App for Linux


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.