RuiLi8080 opened a new pull request #3283: URL: https://github.com/apache/storm/pull/3283
## What is the purpose of the change Add topology config for RAS scheduler to attempt to schedule system components/executors across all workers evenly instead of scheduling them all together after having finished all topology components. ## How was the change tested 1. test with even distribution option: `storm jar /home/y/lib64/jars/storm-starter.jar org.apache.storm.starter.WordCountTopology wc1 -c topology..system.components.even.distribution=true -c topology.acker.executors=8 -c topology.component.resources.onheap.memory.mb=100 -c topology.acker.resources.onheap.memory.mb=150 -c topology.workers=4 -c experimental.topology.ras.order.executors.by.proximity.needs=true -c topology.worker.max.heap.size.mb=1000` ![image](https://user-images.githubusercontent.com/15622046/84088095-5941a000-a9b1-11ea-87e0-26d3b75fc455.png) 2. test without even distribution option `storm jar /home/y/lib64/jars/storm-starter.jar org.apache.storm.starter.WordCountTopology wc-no-even-distribution -c topology.acker.executors=8 -c topology.component.resources.onheap.memory.mb=100 -c topology.acker.resources.onheap.memory.mb=150 -c topology.workers=4 -c experimental.topology.ras.order.executors.by.proximity.needs=true -c topology.worker.max.heap.size.mb=1000` ![image](https://user-images.githubusercontent.com/15622046/84088124-6bbbd980-a9b1-11ea-9839-b65b46329353.png) ---------------------------------------------------------------- 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