Python 3x faster at restoring .sav with large arrays

Prev Go to previous topic

Go to next topic

Last Post 29 Jul 2012 08:03 PM by anon

2 Replies

Sort:


You are not authorized to post a reply.

Author

Messages

anon

New Member

Posts:

29 Jul 2012 08:03 PM
I have recently come across a bottleneck with my application where the RESTORE procedure takes 6-7 seconds to restore 6 large arrays. The arrays are each [4049, 2189] four of them are integers the other two are float. My question to the IDL devs is why does this restore take so long? I have tested the same restore with the readsav function from Pythons scipy library and the Python library was 3x faster. How is it that Python is faster at reading IDL's own file type? Is there any plans to speed this up in future versions of IDL?

Deleted User

New Member

Posts:

01 Aug 2012 10:33 AM
An interesting observation. You might want to contact Exelis VIS technical support with your question. Let them know what version if IDL you are using as well as the commands you are using to restore those file both in Python and IDL..

Deleted User

New Member

Posts:

01 Aug 2012 06:48 PM
Ok I have submitted this to technical support and I look forward to their response.

You are not authorized to post a reply.