Spotfire Ideas Portal
At our company, Web Player RAM is a critical bottle neck of Spotfire. In order to manage the Spotfire environment we need to be able to identify reports that are too big based on the actual memory they consume. We've learned that this information is available in Memory Statistics Log - Total Data Table Size and Total Data View Size. However upon closer look it turns out this log only contains information associated with a Session ID and scheduled reports, which haven't been opened are not included there.
Therefore the log doesn't provide complete insight into the memory usage and we need to come up with workarounds for identifying the big reports.
It would also be good to record how much disk space is also consumed. Sometimes reports are required to be cached to disk, sometimes cache gets overflowed to disk. Understanding both RAM and DISK usage for sessions and Scheduled Updates would help identify both performance issues and also possible RAM expansion. I would also suggest this also gets extended to Automation services and even TERR / Python if possible.
Agreed, it is not easy to determin report usage and memory management when scheduled reports are cached to memory but then get paged to disk reducing performance.
Memory is not unlimited but needs to be managed.