perf-tool: data aggregation

[gireeshpunathil]

Is it possible / meaningful to produce a summary view of the monitor data from JSON format to an aggregate view? If so, what aggregations make sense? when it should be performed? how it should be represented?

This can have some discussions

[gireeshpunathil]

@fmhwong - in terms of aggregations, what views would be useful to you?

• a per object contention count (total)?
• a per object contention participant count (average)?
• a per object contention latency (average)?

also, when / how would these aggregations be recorded in the log?

• part of each contention record, inline? (handy, but consumes CPU)
• only at the end (vmstop event) ?