|
Questions and Answers :
Windows :
Why is Climate prediction eating my disk space like candy
Message board moderation
Author | Message |
---|---|
Send message Joined: 5 Apr 05 Posts: 2 Credit: 52,803 RAC: 0 |
I have 4 projects running one one of my machines, the disk space was set to 2 gig, climate prediction somehow grew to 19.5 gig, this is well beying the 16% of the 2 gig resources I placed in the settings. The files grew to the extent that no other project would connect, I would get a red error message stating that the resources for the other 3 projects were not sufficent for the projects to continue. Disk/resource usage was set as follows: Seti and this project are set at 16.67%, while einstein, and lhc are set at 33.3%. I have since gone in and deleted 18gig that was in a temp folder, but that still did not start the other projects up again. Why did the program expand so radicaly beyond the 2 gig limit I set. the basic specs for the system are amd3200xp+, 1 gigmemory, 40gig hard drive, and windows xppro sp2 As probibly a temporary fix i jumped my disk space to 5gig allowed usage by boink, but I need answers, or this project will on my machines go the way of the do-do bird. In favor of the other projects, as they consume far less resources |
Send message Joined: 5 Sep 04 Posts: 7629 Credit: 24,240,330 RAC: 0 |
As you have your computers hidden, we can only guess. So, guess No. 1) the large file that you deleted was an stdoutdae.txt file in one of the folders of a model. Some people have reported getting this file, although I have no idea why. The only file of this name should be in your BOINC folder, and be less than 1Meg. It is a list of the messages created by BOINC, and which appear in the Messages tab. Guess No. 2) You are creating a sulphur model. These grow to about 2.5Gig. To see why climate[rediction files are big, read the Science pages to see what it is this project is doing. Briefly: cpdn starts with a basic set of data, adds some parameter changes, and then spends weeks creating a 3D model of the Earth\'s atmosphere at half hour intervals. The other projects, in contrast, take \'large\' chunks of data and search them for certain predefined values. This is a lot simpler, and doesn\'t create large files. |
![]() ![]() Send message Joined: 5 Aug 04 Posts: 426 Credit: 2,426,069 RAC: 0 |
As you have your computers hidden, we can only guess. To add another possiblity to this answer. If you have any completed (or aborted) models they tend to leave a fairly large directory behind. You can go backup and delete any that are not referenced in your work tab. (you can even skip the backup as well). BOINC WIKI ![]() ![]() BOINCing since 2002/12/8 |
![]() Send message Joined: 5 Aug 04 Posts: 907 Credit: 299,864 RAC: 0 |
I\'m going to trim CPDN disk usage in the future (i.e. the upcoming coupled model & beyond) -- but a big problem is if it crashes right off the bat, i.e. a segfault that will crash the \"monitor\" program that is supposed to do the cleanup! |
Send message Joined: 5 Apr 05 Posts: 2 Credit: 52,803 RAC: 0 |
I will unhide my systems for 2 weeks, so you can see them. As to the file deleted, I deleted the contents of a tmp folder having well over 1000 files, which had grown to 18gig. If I understand the directory names right, the information I deleted was not in a sulfur model. the directory name is 24hl_000120137. What is strange to me is the fact that I have climate prediction running on 2 of my systems, and only on one do I have this build up of material. on my system named tivo As you have your computers hidden, we can only guess. |
![]() Send message Joined: 5 Aug 04 Posts: 1496 Credit: 95,522,203 RAC: 0 |
Another possibility: There is an option in the graphic display to record the pretty pictures for playback. I don\'t know where the output goes or how it\'s identified (someone please help) but it is said to grow quite large. I think Tolu posted, don\'t recall which Board, that he\'ll make it less easy to turn that feature \"on\" accidently. "We have met the enemy and he is us." -- Pogo Greetings from coastal Washington state, the scenic US Pacific Northwest. |
Send message Joined: 5 Sep 04 Posts: 7629 Credit: 24,240,330 RAC: 0 |
Ah, you\'re right, AstroWX. I\'d forgotten that! Found his post: <a href=\"http://climateapps2.oucs.ox.ac.uk/cpdnboinc/forum_thread.php?id=3588\"> here.</a> |
Send message Joined: 11 Jan 05 Posts: 4 Credit: 390,068 RAC: 0 |
If I understand the directory names right, the information I deleted was not in a sulfur model. the directory name is 24hl_000120137. That WU errored-out. Could that have been caused by the deletion of the contents of that directory, or did the error occur before that? <a href="http://www.boinc.be"><img src="http://www.boinc.be/boinc_be.gif"></a> |
©2024 climateprediction.net