-
Bug
-
Resolution: Unresolved
-
Major
-
None
-
None
-
None
-
Oracle JDK 1.7
Windows XP 32bit
i5
3.24 GB RAM
When loading spreadsheets, the size of a spreadsheet in memory is between 40 to 90 times the size on disk:
170KB xlsx file on disk = 28MB in heap
511KB xlsx file on disk = 43MB in heap
6MB xlsx file on disk = 346MB in heap
7.4MB xlsx file on disk = 425MB in heap
31MB xlsx file on disk > 1024MB in heap
When making a web application with several score users all working on different spreadsheets, this requires a large amount of memory allocated to the VM via -Xms
How do consumers of the ZK spreadsheet API reduce the large footprint in heap?