Github user revans2 commented on a diff in the pull request:
https://github.com/apache/storm/pull/2419#discussion_r151700529
--- Diff:
storm-server/src/test/java/org/apache/storm/scheduler/resource/strategies/scheduling/TestGenericResourceAwareStrategy.java
---
@@ -187,16 +199,21 @@ public void testGenericResourceAwareStrategy() {
rs.prepare(conf);
rs.schedule(topologies, cluster);
+ //We need to have 3 slots on 3 separate hosts to make the GPU
situation work
+
HashSet<HashSet<ExecutorDetails>> expectedScheduling = new
HashSet<>();
- expectedScheduling.add(new HashSet<>(Arrays.asList(new
ExecutorDetails(0, 0)))); //Spout
+ expectedScheduling.add(new HashSet<>(Arrays.asList(new
ExecutorDetails(3, 3)))); //bolt-3 - 500 MB, 50% CPU, 2 GPU
--- End diff --
This test was also wrong because it was not taking the GPUs into account.
That is why I called out all of the resources in the comments for the
scheduling. bolt-3 requires 2 GPUs. Each node has 2 GPUs on it. Having 2
instances of a bolt-3 on the same node, which is what the scheduling was
before, would go over the GPU count for the node.
---
