添加Capacity调度器的源码注释

hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/NodeType.java

@@ -22,7 +22,12 @@
  * Resource classification.
  */
 public enum NodeType {
-  NODE_LOCAL(0), RACK_LOCAL(1), OFF_SWITCH(2);
+  //请求规定了必须运行在某个的服务器节点
+  NODE_LOCAL(0),
+  // 请求规定了必须运行在某个机架上
+  RACK_LOCAL(1),
+  // 这个请求对本地化没有要求
+  OFF_SWITCH(2);
 
   private final int index;
 

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

@@ -1363,6 +1363,7 @@ protected void nodeUpdate(RMNode rmNode) {
     readLock.lock();
     try {
       setLastNodeUpdateTime(Time.now());
+      // 优先处理心跳逻辑。
       super.nodeUpdate(rmNode);
     } finally {
       readLock.unlock();
@@ -1370,6 +1371,7 @@ protected void nodeUpdate(RMNode rmNode) {
 
     // Try to do scheduling
     if (!scheduleAsynchronously) {
+      // 在不开启异步调度的时候，需要借助节点的心跳进行调度。默认建议开启异步调度。
       writeLock.lock();
       try {
         // reset allocation and reservation stats before we start doing any
@@ -1484,6 +1486,13 @@ private void updateSchedulerHealth(long now, NodeId nodeId,
       .getAssignmentInformation().getReserved());
   }
 
+  /**
+   * 是否需要继续分配资源。
+   * @param assignment 已经分配的资源信息
+   * @param offswitchCount 不限制节点和机架的个数。
+   * @param assignedContainers 已经分配的Container个数
+   * @return
+   */
   private boolean canAllocateMore(CSAssignment assignment, int offswitchCount,
       int assignedContainers) {
     // Current assignment shouldn't be empty
@@ -1529,6 +1538,7 @@ private CandidateNodeSet<FiCaSchedulerNode> getCandidateNodeSet(
   /**
    * We need to make sure when doing allocation, Node should be existed
    * And we will construct a {@link CandidateNodeSet} before proceeding
+   * @param withNodeHeartbeat 如果是未开启异步调度，则需要通过心跳来实现，当前值为true,否则为false
    */
   private void allocateContainersToNode(NodeId nodeId,
       boolean withNodeHeartbeat) {
@@ -1544,6 +1554,7 @@ private void allocateContainersToNode(NodeId nodeId,
       // Only check if we can allocate more container on the same node when
       // scheduling is triggered by node heartbeat
       if (null != assignment && withNodeHeartbeat) {
+        // 如果是由心跳触发的分配，会尝试再次分配。
         if (assignment.getType() == NodeType.OFF_SWITCH) {
           offswitchCount++;
         }
@@ -1693,6 +1704,12 @@ private void allocateFromReservedContainer(FiCaSchedulerNode node,
     submitResourceCommitRequest(getClusterResource(), assignment);
   }
 
+  /**
+   * 为Container申请或者预留资源。
+   * @param candidates
+   * @param withNodeHeartbeat
+   * @return
+   */
   private CSAssignment allocateOrReserveNewContainers(
       CandidateNodeSet<FiCaSchedulerNode> candidates,
       boolean withNodeHeartbeat) {
@@ -1787,6 +1804,12 @@ private CSAssignment allocateContainersOnMultiNodes(
     return allocateOrReserveNewContainers(candidates, false);
   }
 
+  /**
+   * 将资源分派给指定的节点
+   * @param candidates
+   * @param withNodeHeartbeat
+   * @return
+   */
   @VisibleForTesting
   CSAssignment allocateContainersToNode(
       CandidateNodeSet<FiCaSchedulerNode> candidates,