You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-issues@hadoop.apache.org by GitBox <gi...@apache.org> on 2020/12/01 15:34:49 UTC
[GitHub] [hadoop] jiwq commented on a change in pull request #2494: YARN-10380: Import logic of multi-node allocation in CapacityScheduler
jiwq commented on a change in pull request #2494:
URL: https://github.com/apache/hadoop/pull/2494#discussion_r533502841
##########
File path: hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/capacity/CapacityScheduler.java
##########
@@ -544,44 +544,73 @@ static void schedule(CapacityScheduler cs) throws InterruptedException{
if(nodeSize == 0) {
return;
}
- int start = random.nextInt(nodeSize);
+ if (!cs.multiNodePlacementEnabled) {
+ // First randomize the start point
+ int current = 0;
+ int start = random.nextInt(nodeSize);
- // To avoid too verbose DEBUG logging, only print debug log once for
- // every 10 secs.
- boolean printSkipedNodeLogging = false;
- if (Time.monotonicNow() / 1000 % 10 == 0) {
- printSkipedNodeLogging = (!printedVerboseLoggingForAsyncScheduling);
- } else {
- printedVerboseLoggingForAsyncScheduling = false;
- }
+ // To avoid too verbose DEBUG logging, only print debug log once for
+ // every 10 secs.
+ boolean printSkipedNodeLogging = false;
+ if (Time.monotonicNow() / 1000 % 10 == 0) {
+ printSkipedNodeLogging = (!printedVerboseLoggingForAsyncScheduling);
+ } else {
+ printedVerboseLoggingForAsyncScheduling = false;
+ }
+
+ // Allocate containers of node [start, end)
+ for (FiCaSchedulerNode node : nodes) {
+ if (current++ >= start) {
+ if (shouldSkipNodeSchedule(node, cs, printSkipedNodeLogging)) {
+ continue;
+ }
+ cs.allocateContainersToNode(node.getNodeID(), false);
+ }
+ }
- // Allocate containers of node [start, end)
- for (FiCaSchedulerNode node : nodes) {
- if (current++ >= start) {
+ current = 0;
+
+ // Allocate containers of node [0, start)
+ for (FiCaSchedulerNode node : nodes) {
+ if (current++ > start) {
+ break;
+ }
if (shouldSkipNodeSchedule(node, cs, printSkipedNodeLogging)) {
continue;
}
cs.allocateContainersToNode(node.getNodeID(), false);
}
- }
-
- current = 0;
- // Allocate containers of node [0, start)
- for (FiCaSchedulerNode node : nodes) {
- if (current++ > start) {
- break;
+ if (printSkipedNodeLogging) {
+ printedVerboseLoggingForAsyncScheduling = true;
}
- if (shouldSkipNodeSchedule(node, cs, printSkipedNodeLogging)) {
- continue;
+ } else {
+ //Get all partitions
Review comment:
```suggestion
// Get all partitions
```
##########
File path: hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/capacity/CapacityScheduler.java
##########
@@ -544,44 +544,73 @@ static void schedule(CapacityScheduler cs) throws InterruptedException{
if(nodeSize == 0) {
return;
}
- int start = random.nextInt(nodeSize);
+ if (!cs.multiNodePlacementEnabled) {
+ // First randomize the start point
+ int current = 0;
+ int start = random.nextInt(nodeSize);
- // To avoid too verbose DEBUG logging, only print debug log once for
- // every 10 secs.
- boolean printSkipedNodeLogging = false;
- if (Time.monotonicNow() / 1000 % 10 == 0) {
- printSkipedNodeLogging = (!printedVerboseLoggingForAsyncScheduling);
- } else {
- printedVerboseLoggingForAsyncScheduling = false;
- }
+ // To avoid too verbose DEBUG logging, only print debug log once for
+ // every 10 secs.
+ boolean printSkipedNodeLogging = false;
+ if (Time.monotonicNow() / 1000 % 10 == 0) {
+ printSkipedNodeLogging = (!printedVerboseLoggingForAsyncScheduling);
+ } else {
+ printedVerboseLoggingForAsyncScheduling = false;
+ }
+
+ // Allocate containers of node [start, end)
+ for (FiCaSchedulerNode node : nodes) {
+ if (current++ >= start) {
+ if (shouldSkipNodeSchedule(node, cs, printSkipedNodeLogging)) {
+ continue;
+ }
+ cs.allocateContainersToNode(node.getNodeID(), false);
+ }
+ }
- // Allocate containers of node [start, end)
- for (FiCaSchedulerNode node : nodes) {
- if (current++ >= start) {
+ current = 0;
+
+ // Allocate containers of node [0, start)
+ for (FiCaSchedulerNode node : nodes) {
+ if (current++ > start) {
+ break;
+ }
if (shouldSkipNodeSchedule(node, cs, printSkipedNodeLogging)) {
continue;
}
cs.allocateContainersToNode(node.getNodeID(), false);
}
- }
-
- current = 0;
- // Allocate containers of node [0, start)
- for (FiCaSchedulerNode node : nodes) {
- if (current++ > start) {
- break;
+ if (printSkipedNodeLogging) {
+ printedVerboseLoggingForAsyncScheduling = true;
}
- if (shouldSkipNodeSchedule(node, cs, printSkipedNodeLogging)) {
- continue;
+ } else {
+ //Get all partitions
+ List<String> partitions = cs.nodeTracker.getPartitions();
+ int partitionSize = partitions.size();
+ // First randomize the start point
+ int start = random.nextInt(partitionSize);
+ int current = 0;
+ // Allocate containers of partition [start, end)
+ for (String partititon : partitions) {
+ if (current++ >= start) {
+ cs.allocateContainersToNode(cs.getCandidateNodeSet(partititon),
Review comment:
Due to `getCandiateNodeSet(String)` method maybe return null, so call `allocateContainersToNode` can cause NPE.
##########
File path: hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/capacity/CapacityScheduler.java
##########
@@ -531,11 +531,11 @@ private static boolean shouldSkipNodeSchedule(FiCaSchedulerNode node,
/**
* Schedule on all nodes by starting at a random point.
+ * Schedule on all partitions by starting at a random partition
+ * when multiNodePlacementEnabled is true.
* @param cs
*/
static void schedule(CapacityScheduler cs) throws InterruptedException{
- // First randomize the start point
- int current = 0;
Review comment:
Why remove this variable and add new within if-else block?
----------------------------------------------------------------
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:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: common-issues-help@hadoop.apache.org