- Posts: 4
SAM run stability
- uricohn
- Topic Author
Less
More
20 Aug 2012 02:56 #777
by uricohn
SAM run stability was created by uricohn
Hi Paul
When I ran the latest rev of SAM through a SAMUL code I recieved the follwing error message and then SAM crashed (got stuck)
Did I do something wrong or is it Hardware issue
Warm Regards
Uri
When I ran the latest rev of SAM through a SAMUL code I recieved the follwing error message and then SAM crashed (got stuck)
Did I do something wrong or is it Hardware issue
Warm Regards
Uri
Please Log in or Create an account to join the conversation.
- pgilman
Less
More
- Posts: 5423
20 Aug 2012 14:21 #778
by pgilman
Replied by pgilman on topic SAM run stability
Hi Uri,
If your script involves running many simulations (a few hundred or more), SAM may be running out of memory. You may want to try using the "false" parameter with the simulate() function to help reduce the amount of memory SAM requires. Without the "false" parameter, SAM generates a large amount of data for the time series data viewer in the user interface. If you don't need to view the data on the Results page, you can use "simulate(false)" to prevent SAM from generating the data. (That will also speed up simulations.)
If that does not solve the problem, then you may want to break up your analysis into runs of less than 200 or 300 at a time. Your computer's memory resources determine how many runs SAM can manage without running out of memory.
This is an issue we are aware of, and hope to fix in a future version.
Is that what is causing the error message to appear in your case?
Best regards,
Paul.
If your script involves running many simulations (a few hundred or more), SAM may be running out of memory. You may want to try using the "false" parameter with the simulate() function to help reduce the amount of memory SAM requires. Without the "false" parameter, SAM generates a large amount of data for the time series data viewer in the user interface. If you don't need to view the data on the Results page, you can use "simulate(false)" to prevent SAM from generating the data. (That will also speed up simulations.)
If that does not solve the problem, then you may want to break up your analysis into runs of less than 200 or 300 at a time. Your computer's memory resources determine how many runs SAM can manage without running out of memory.
This is an issue we are aware of, and hope to fix in a future version.
Is that what is causing the error message to appear in your case?
Best regards,
Paul.
Please Log in or Create an account to join the conversation.
- ajgong
Less
More
- Posts: 6
23 Oct 2012 12:32 #779
by ajgong
Replied by ajgong on topic SAM run stability
I'm also running into problems with large batches (the 9.27 beta crashes once it hits around 1,860,000K memory or about 180 simulations). This happens on two different computers, one which as 3GB memory and one which has 6GB memory. Are there any functions that allow me to clear the temp files or any thing else that SAM has stored up in memory?
I've tried using a few of the built in functions after every simulation, 4 simulations, 30 simulations, or 120 simulations, but it crashes at about the 180th one each time. I'm not sure if I have the syntax correct or not:
ClearSimResults
DeleteTempFiles
ClearCache
I've also been using simulate(false) and writing the results to a .csv file if that matters
Any thoughts?
edit:
I seem to be able to run the program for a while (1200 simulations in 28 mins) at about 123,000K memory with the following syntax:
simulate("batch", false)
...
ClearSimResults("batch")
DeleteTempFiles()
ClearCache()
However I do get errors that say:
"Can't create a window of class wxWindowsClassNR (error 1406: cannot create a top-level child window.)"
"Can't create a window of class wxWindowsClassNR (error 1158: the current process has used all of its system allowance of handles for Windows Manager objects)
Based on timestamps, the above errors occured a few times each second, although not always in pairs, for 20 minutes while the script was running. Not a major issue, but maybe SAM isn't relinquishing handles or something.
I've tried using a few of the built in functions after every simulation, 4 simulations, 30 simulations, or 120 simulations, but it crashes at about the 180th one each time. I'm not sure if I have the syntax correct or not:
ClearSimResults
DeleteTempFiles
ClearCache
I've also been using simulate(false) and writing the results to a .csv file if that matters
Any thoughts?
edit:
I seem to be able to run the program for a while (1200 simulations in 28 mins) at about 123,000K memory with the following syntax:
simulate("batch", false)
...
ClearSimResults("batch")
DeleteTempFiles()
ClearCache()
However I do get errors that say:
"Can't create a window of class wxWindowsClassNR (error 1406: cannot create a top-level child window.)"
"Can't create a window of class wxWindowsClassNR (error 1158: the current process has used all of its system allowance of handles for Windows Manager objects)
Based on timestamps, the above errors occured a few times each second, although not always in pairs, for 20 minutes while the script was running. Not a major issue, but maybe SAM isn't relinquishing handles or something.
Please Log in or Create an account to join the conversation.
- pgilman
Less
More
- Posts: 5423
24 Oct 2012 11:30 #780
by pgilman
Replied by pgilman on topic SAM run stability
Thank you for the detailed description.
We've made some improvements in the Beta version that should help with the memory issues for very large numbers of simulations. Unfortunately, the improvements do not completely eliminate the problem, but should allow for more simulations per batch. The memory issues are related to how SAM handles user interface objects.
Best regards,
Paul.
We've made some improvements in the Beta version that should help with the memory issues for very large numbers of simulations. Unfortunately, the improvements do not completely eliminate the problem, but should allow for more simulations per batch. The memory issues are related to how SAM handles user interface objects.
Best regards,
Paul.
Please Log in or Create an account to join the conversation.
Moderators: pgilman