Welcome to MilkyWay@home

milkyway v0.2 code release

Message boards : Application Code Discussion : milkyway v0.2 code release
Message board moderation

To post messages, you must log in.

1 · 2 · Next

AuthorMessage
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 6441 - Posted: 22 Nov 2008, 14:06:02 UTC
Last modified: 24 Nov 2008, 18:30:12 UTC

Here's the code for the new application (v0.2) (update - now in tar or zip):
http://milkyway.cs.rpi.edu/milkyway/download/code_release/milkyway_release_0.2.tar
http://milkyway.cs.rpi.edu/milkyway/download/code_release/milkyway_release_0.2.zip

I've included makefiles for linux and osx:
milkyway/bin/make.osx
milkyway/bin/make.linux

You'll have to change the paths to your boinc code and libraries in the makefile for these to work.

Also, i've included a few test files:
milkyway/bin/test_files/stars.txt
milkyway/bin/test_files/search_parameters.txt
milkyway/bin/test_files/astronomy_parameters.txt
milkyway/bin/test_files/astronomy_parameters-cut-large.txt
milkyway/bin/test_files/astronomy_parameters-cut-medium.txt
milkyway/bin/test_files/astronomy_parameters-nocut-small.txt
milkyway/bin/test_files/astronomy_parameters-unconvolved-small.txt

astronomy_parameters.txt tests having a slice of the volume cut out, with a small integral calculation (using the convolution).

astronomy_parameters-nocut-small.txt tests the same as astronomy_parameters.txt except without cutting out a slice of the volume
astronomy_parameters-unconvolved-small.txt doesn't use the convolution or any cutting of volume.

Generally, at the end of the parameters file:
convolve: 30
sgr_coordinates: 0
wedge: 86
r[min,max,steps]: 16.0, 22.5, 175
mu[min,max,steps]: 310, 420, 400
nu[min,max,steps]: -1.25, 1.25, 20
number_cuts: 1
r_cut[min,max,steps][3]: 16.0, 22.5, 175
mu_cut[min,max,steps][3]: 21.6, 22.3, 100
nu_cut[min,max,steps][3]: -1.25, 1.25, 20

you can modifiy the number of steps and add/remove additional cuts for testing (less steps = less integral calculation). setting sgr_coordinates to 1 uses a different method to calculate the integral (Nate will have to fill you in on exactly what that is). Setting convolve to 0 means the code won't run the convolution, any value > 0 is the number of points calculated around an integral (or star point).

Let me know if you have any problems using these, like missing files or anything like that. Also, i'm not a master of makefiles so if they don't work i apologize beforehand :)
ID: 6441 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile DeleteNull

Send message
Joined: 5 Apr 08
Posts: 11
Credit: 105,359,710
RAC: 0
Message 6443 - Posted: 22 Nov 2008, 15:48:16 UTC

Nice, but the files probability.h/c are missing in the astronony folder?!

ID: 6443 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 6447 - Posted: 22 Nov 2008, 16:10:31 UTC - in response to Message 6443.  

Nice, but the files probability.h/c are missing in the astronony folder?!


thanks for catching that, i've replaced the tar file with an updated version.
ID: 6447 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
jedirock
Avatar

Send message
Joined: 8 Nov 08
Posts: 178
Credit: 6,140,854
RAC: 0
Message 6450 - Posted: 22 Nov 2008, 16:29:04 UTC

I seem to be missing evaluation/mpi_evaluator.h. Referenced from astronomy/evaluation_state.h.
ID: 6450 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile DeleteNull

Send message
Joined: 5 Apr 08
Posts: 11
Credit: 105,359,710
RAC: 0
Message 6451 - Posted: 22 Nov 2008, 16:36:26 UTC
Last modified: 22 Nov 2008, 16:38:00 UTC

o.k., thanks. But now:
../evaluation/mpi_evaluator.h
is missing.

EDIT: jedirock is the same file missing.
ID: 6451 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 6453 - Posted: 22 Nov 2008, 17:10:43 UTC - in response to Message 6450.  

I seem to be missing evaluation/mpi_evaluator.h. Referenced from astronomy/evaluation_state.h.


weird, it shouldn't need that. i've fixed the download. you should just be able to remove that line from evaluation_state.h if you don't want to redownload the whole thing.
ID: 6453 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile DeleteNull

Send message
Joined: 5 Apr 08
Posts: 11
Credit: 105,359,710
RAC: 0
Message 6455 - Posted: 22 Nov 2008, 17:18:06 UTC

I have downloaded the new tar, but anyway now missing:
/searches/asynchronous_newton_method.<h/c>

ID: 6455 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
jedirock
Avatar

Send message
Joined: 8 Nov 08
Posts: 178
Credit: 6,140,854
RAC: 0
Message 6471 - Posted: 23 Nov 2008, 11:34:42 UTC

Got the code compiled, but just trying to play with compiler flags before touching source. However, I can't seem to get the app_info right for comparing to the existing application. What's the application name? astronomy, milkyway, or something else?
ID: 6471 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 6475 - Posted: 23 Nov 2008, 12:23:00 UTC - in response to Message 6455.  

I have downloaded the new tar, but anyway now missing:
/searches/asynchronous_newton_method.



do you know what file is including that? it shouldn't be there.
ID: 6475 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 6476 - Posted: 23 Nov 2008, 12:23:18 UTC - in response to Message 6471.  

Got the code compiled, but just trying to play with compiler flags before touching source. However, I can't seem to get the app_info right for comparing to the existing application. What's the application name? astronomy, milkyway, or something else?


the name of this one is milkyway.
ID: 6476 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
John Clark

Send message
Joined: 4 Oct 08
Posts: 1734
Credit: 64,228,409
RAC: 0
Message 6490 - Posted: 23 Nov 2008, 16:42:14 UTC - in response to Message 6441.  
Last modified: 23 Nov 2008, 16:42:39 UTC

Here's the code for the new application (v0.2): http://milkyway.cs.rpi.edu/milkyway/download/milkyway_release_0.2.tar
:)



Travis

The link you give us for the 0.2 test client release downloads OK. But, on a Windows XP Pro system I have not been able to open and unzip the file. This even occurs when I uninstall WinZip from the system.

Can you or Dave advise me how I can prepare the second faster test client for use on a windows system, please?

Clearly many with Windows OSes have cracked the problem, but I need help.
ID: 6490 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 6493 - Posted: 23 Nov 2008, 17:39:29 UTC - in response to Message 6490.  

Here's the code for the new application (v0.2): http://milkyway.cs.rpi.edu/milkyway/download/milkyway_release_0.2.tar
:)



Travis

The link you give us for the 0.2 test client release downloads OK. But, on a Windows XP Pro system I have not been able to open and unzip the file. This even occurs when I uninstall WinZip from the system.

Can you or Dave advise me how I can prepare the second faster test client for use on a windows system, please?

Clearly many with Windows OSes have cracked the problem, but I need help.


This application should automatically download when you get a workunit for the new application (anything named nm_testX at the moment), you don't need to compile it yourself. Just make sure you have 'run test applications' enabled in your project preferences: http://milkyway.cs.rpi.edu/milkyway/prefs.php?subset=project

I'm not generating too many workunits for the new application at the moment, because we're still testing it to make sure everything is working correctly, so if it doesn't download immediately don't be surprised.
ID: 6493 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile DeleteNull

Send message
Joined: 5 Apr 08
Posts: 11
Credit: 105,359,710
RAC: 0
Message 6494 - Posted: 23 Nov 2008, 17:39:46 UTC - in response to Message 6471.  

Got the code compiled, but just trying to play with compiler flags before touching source. However, I can't seem to get the app_info right for comparing to the existing application. What's the application name? astronomy, milkyway, or something else?


<app_info>
<app>
<name>milkyway</name>
</app>
<file_info>
<name>your_application_name</name>
<executable/>
</file_info>
<app_version>
<app_name>milkyway</app_name>
<version_num>2</version_num>
<file_ref>
<file_name>your_application_name</file_name>
<main_program/>
</file_ref>
</app_version>
</app_info>
ID: 6494 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr Mystery

Send message
Joined: 21 Nov 08
Posts: 90
Credit: 2,601
RAC: 0
Message 6495 - Posted: 23 Nov 2008, 17:39:51 UTC - in response to Message 6490.  

Here's the code for the new application (v0.2): http://milkyway.cs.rpi.edu/milkyway/download/milkyway_release_0.2.tar
:)




The link you give us for the 0.2 test client release downloads OK. But, on a Windows XP Pro system I have not been able to open and unzip the file.

Can you or Dave advise me how I can prepare the second faster test client for use on a windows system, please?


Makes you wonder, dunnit. Get a beginner's guide to tarballs and compiling?


ID: 6495 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 6498 - Posted: 23 Nov 2008, 17:49:48 UTC - in response to Message 6495.  

Here's the code for the new application (v0.2): http://milkyway.cs.rpi.edu/milkyway/download/milkyway_release_0.2.tar
:)




The link you give us for the 0.2 test client release downloads OK. But, on a Windows XP Pro system I have not been able to open and unzip the file.

Can you or Dave advise me how I can prepare the second faster test client for use on a windows system, please?


Makes you wonder, dunnit. Get a beginner's guide to tarballs and compiling?


Be nice :P Anyways, it's now available in tar or zip format (see the first post in the thread). Also, for future reference, if you want to unzip a .tar file on windows you could use one of the following:

gzip - http://www.gzip.org/
winrar - http://www.rarlab.org

I'm sure there's more (and probably better ways to do it), but I haven't used windows in quite awhile now, so maybe some other people can update that :P
ID: 6498 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
jedirock
Avatar

Send message
Joined: 8 Nov 08
Posts: 178
Credit: 6,140,854
RAC: 0
Message 6502 - Posted: 23 Nov 2008, 17:55:46 UTC - in response to Message 6494.  

Thanks Stoffel, but I do have an app_info. I just didn't know what the application name was, and there were no WUs last night to test different names. Both problems were solved.
ID: 6502 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
John Clark

Send message
Joined: 4 Oct 08
Posts: 1734
Credit: 64,228,409
RAC: 0
Message 6506 - Posted: 23 Nov 2008, 18:15:00 UTC - in response to Message 6498.  
Last modified: 23 Nov 2008, 18:21:11 UTC

Also, for future reference, if you want to unzip a .tar file on windows you could use one of the following:

gzip - http://www.gzip.org/
winrar - http://www.rarlab.org

I'm sure there's more (and probably better ways to do it), but I haven't used windows in quite awhile now, so maybe some other people can update that :P


Thanks

I was looking at descriptions of the .TAR file extension (first time I've seen it) and this quickly moved to discussing .RAR files as well.

I have a copy of 7Zip on another rig, which is also mentioned in this file extension article, so I may have an unzip tool already.

I will play with it.

NOTE: I will ignore The Milky Bar Kid's snide comment.

I am sure in my specialist field I could possibly snipe at his ignorance of - (a) chemicals capable of guiding research on cell necrosis and apoptosis; or (b) the design and build of solid propellant rocket engines. But, then again, this is easy when I do not know him, his interests nor specialisms.
ID: 6506 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 6509 - Posted: 23 Nov 2008, 18:48:08 UTC - in response to Message 6506.  

I am sure in my specialist field I could possibly snipe at his ignorance of - (a) chemicals capable of guiding research on cell necrosis and apoptosis; or (b) the design and build of solid propellant rocket engines. But, then again, this is easy when I do not know him, his interests nor specialisms.


lol, it still is funny you haven't come across a tar file yet (or .tgz or .tar.gz) :D
ID: 6509 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
John Clark

Send message
Joined: 4 Oct 08
Posts: 1734
Credit: 64,228,409
RAC: 0
Message 6511 - Posted: 23 Nov 2008, 18:55:05 UTC - in response to Message 6509.  
Last modified: 23 Nov 2008, 19:01:53 UTC

lol, it still is funny you haven't come across a tar file yet (or .tgz or .tar.gz) :D


True though LOL

I see that WinZip is supposed to recognise that file extension, but my copy certainly did not (and it's not too old yet).
ID: 6511 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Michael Roberts
Avatar

Send message
Joined: 25 Dec 07
Posts: 10
Credit: 217,283
RAC: 0
Message 6534 - Posted: 23 Nov 2008, 23:07:00 UTC

1. I'm using linux. I don't see how to use the makefiles -- the dependencies for all -- the $(PROGS) definition -- do not appear anywhere as targets. I get:

xxx> make -f make.linux
make: *** No rule to make target `boinc_astronomy', needed by `all'.  Stop.
xxx>


2. If you add a comment line to the beginning of the makefiles, life will be easier for those using emacs/xemacs (and perhaps other editors?):

# -*- Makefile -*-


This turns on Makefile colourisation for that file in a generic editor configuration.
ID: 6534 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
1 · 2 · Next

Message boards : Application Code Discussion : milkyway v0.2 code release

©2024 Astroinformatics Group