narendly commented on a change in pull request #490: DRAFT: Add the metric 
components for WAGED rebalancer
URL: https://github.com/apache/helix/pull/490#discussion_r328921464
 
 

 ##########
 File path: 
helix-core/src/main/java/org/apache/helix/controller/rebalancer/waged/WagedRebalancer.java
 ##########
 @@ -255,11 +261,14 @@ private void 
refreshBaseline(ResourceControllerDataProvider clusterData,
       Map<HelixConstants.ChangeType, Set<String>> clusterChanges, Map<String, 
Resource> resourceMap,
       final CurrentStateOutput currentStateOutput) throws 
HelixRebalanceException {
     LOG.info("Start calculating the new best possible assignment.");
+    LatencyMetric partialRebalanceMetric = new 
PartialRebalanceMetric("PartialRebalanceLatency");
 
 Review comment:
   @i3wangyi 
   I'm not sure if I understand your concern entirely.
   
   It actually is meant to be like that - this is just one metric we are 
adding. Here I've only added one metric just as an example, but every "step" of 
the pipeline like the way you referred to, will have an implementation. 
   
   So you could say that we will at least have a Metric object for each step. 
We will also have a metric for the entire step. We'll emit appropriate 
gauges/counters/Histogram for more detailed analysis.
   
   In short, we want to do exactly what you're describing. But at the code 
level, we won't have any dependency between metrics or hierarchy since JMX will 
treat every metric equally anyways (whether it represents step 0 or the entire 
duration of the pipeline, they are just another metric to JMX). Hope that 
clears things up.
   
   

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


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@helix.apache.org
For additional commands, e-mail: reviews-h...@helix.apache.org

Reply via email to