[GitHub] huijunw opened a new issue #2964: stmgr-stmgr backpressure metric to tell peer stmgr

[GitBox <gi...@apache.org>]

huijunw opened a new issue #2964: stmgr-stmgr backpressure metric to tell peer stmgr
URL: https://github.com/apache/incubator-heron/issues/2964
 
 
   The existing backpressure metrics measure Time spent in back pressure:
   ```
   __server/__time_spent_back_pressure_aggr: back pressure in all scenarios
   __server/__time_spent_back_pressure_initiated: back pressure when sending data to both local heron-instance and remote stmgrs
   __time_spent_back_pressure_by_compid/container_#_task_#: back pressure when sending data to local heron-instance
   ```
   ```
   __server/__time_spent_back_pressure_aggr >= __server/__time_spent_back_pressure_initiated >= sum(__time_spent_back_pressure_by_compid/container_#_task_#)
   ```
   ```
   __server/__time_spent_back_pressure_aggr = __server/__time_spent_back_pressure_initiated + __time_spent_back_pressure_signaled_from_remote
   __server/__time_spent_back_pressure_initiated = sum(__time_spent_back_pressure_by_stmgrid/stmgr_#) + sum(__time_spent_back_pressure_by_compid/container_#_task_#)
   ```
   `__time_spent_back_pressure_by_compid` tells the peer heron-instance. Similarly, we may add `__time_spent_back_pressure_by_stmgrid` to tell the peer stmgr, which may help debug which container is slow in stmgr-stmgr backpressure scenario.
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services