Details
-
Improvement
-
Status: Open
-
Major
-
Resolution: Unresolved
-
None
-
None
-
None
-
None
-
Operability
Description
Need an interface to gauge the performance metrics of the jobs/processes that are managed by Falcon. Some of them happen to be:
- how is a process performing with respect to its expected duration for SLA compliance
- how much time is being spent in from launching the process itself to the successful completion of the process
- how many temporary failures and permanent failures has the process experienced during a particular time window
- how is a data feed doing with respect to the SLA compliance
- how much resources are being consumed by a process (say in hadoop's case mapreduce slots, this can help chip in capacity planning)
- how much resources are being consumed on an aggregate level by a group of processes
and some more, as and when they appear to be requirement.
thanks,
Nikhil