[
https://issues.apache.org/jira/browse/STORM-898?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15058695#comment-15058695
]
ASF GitHub Bot commented on STORM-898:
--------------------------------------
Github user rfarivar commented on a diff in the pull request:
https://github.com/apache/storm/pull/921#discussion_r47691194
--- Diff:
storm-core/src/jvm/backtype/storm/scheduler/resource/RAS_Nodes.java ---
@@ -0,0 +1,159 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ * <p>
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * <p>
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package backtype.storm.scheduler.resource;
+
+import backtype.storm.Config;
+import backtype.storm.scheduler.Cluster;
+import backtype.storm.scheduler.ExecutorDetails;
+import backtype.storm.scheduler.SchedulerAssignment;
+import backtype.storm.scheduler.SupervisorDetails;
+import backtype.storm.scheduler.Topologies;
+import backtype.storm.scheduler.WorkerSlot;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.util.Collection;
+import java.util.HashMap;
+import java.util.Map;
+
+public class RAS_Nodes {
+
+ private Map<String, RAS_Node> nodeMap;
+
+ private static final Logger LOG =
LoggerFactory.getLogger(RAS_Nodes.class);
+
+ public RAS_Nodes(Cluster cluster, Topologies topologies) {
+ this.nodeMap = getAllNodesFrom(cluster, topologies);
+ }
+
+ public static Map<String, RAS_Node> getAllNodesFrom(Cluster cluster,
Topologies topologies) {
+ Map<String, RAS_Node> nodeIdToNode = new HashMap<String,
RAS_Node>();
+ for (SupervisorDetails sup : cluster.getSupervisors().values()) {
+ //Node ID and supervisor ID are the same.
+ String id = sup.getId();
+ boolean isAlive = !cluster.isBlackListed(id);
+ LOG.debug("Found a {} Node {} {}",
+ isAlive ? "living" : "dead", id, sup.getAllPorts());
+ LOG.debug("resources_mem: {}, resources_CPU: {}",
sup.getTotalMemory(), sup.getTotalCPU());
+ nodeIdToNode.put(sup.getId(), new RAS_Node(id,
sup.getAllPorts(), isAlive, sup, cluster, topologies));
+ }
+ for (Map.Entry<String, SchedulerAssignment> entry :
cluster.getAssignments().entrySet()) {
+ String topId = entry.getValue().getTopologyId();
+ for (WorkerSlot workerSlot : entry.getValue().getSlots()) {
+ String id = workerSlot.getNodeId();
+ RAS_Node node = nodeIdToNode.get(id);
+ if (node == null) {
+ LOG.info("Found an assigned slot on a dead supervisor
{} with executors {}",
+ workerSlot, RAS_Node.getExecutors(workerSlot,
cluster));
+ node = new RAS_Node(id, null, false, null, cluster,
topologies);
+ nodeIdToNode.put(id, node);
+ }
+ if (!node.isAlive()) {
+ //The supervisor on the node down so add an orphaned
slot to hold the unsupervised worker
+ node.addOrphanedSlot(workerSlot);
+ }
+ if (node.assignInternal(workerSlot, topId, true)) {
+ LOG.warn("Bad scheduling state, " + workerSlot + "
assigned multiple workers, unassigning everything...");
+ node.free(workerSlot);
+ }
+ }
+ }
+ updateAvailableResources(cluster, topologies, nodeIdToNode);
+ return nodeIdToNode;
+ }
+
+ /**
+ * updates the available resources for every node in a cluster
+ * by recalculating memory requirements.
+ *
+ * @param cluster the cluster used in this calculation
+ * @param topologies container of all topologies
+ * @param nodeIdToNode a map between node id and node
+ */
+ private static void updateAvailableResources(Cluster cluster,
+ Topologies topologies,
+ Map<String, RAS_Node>
nodeIdToNode) {
+ //recompute memory
+ if (cluster.getAssignments().size() > 0) {
+ for (Map.Entry<String, SchedulerAssignment> entry :
cluster.getAssignments()
+ .entrySet()) {
+ Map<ExecutorDetails, WorkerSlot> executorToSlot =
entry.getValue()
+ .getExecutorToSlot();
+ Map<ExecutorDetails, Double> topoMemoryResourceList =
topologies.getById(entry.getKey()).getTotalMemoryResourceList();
+
+ if (topoMemoryResourceList == null ||
topoMemoryResourceList.size() == 0) {
+ continue;
+ }
+ for (Map.Entry<ExecutorDetails, WorkerSlot> execToSlot :
executorToSlot
+ .entrySet()) {
+ WorkerSlot slot = execToSlot.getValue();
+ ExecutorDetails exec = execToSlot.getKey();
+ RAS_Node node = nodeIdToNode.get(slot.getNodeId());
+ if (!node.isAlive()) {
+ continue;
+ // We do not free the assigned slots (the orphaned
slots) on the inactive supervisors
+ // The inactive node will be treated as a
0-resource node and not available for other unassigned workers
+ }
+ if (topoMemoryResourceList.containsKey(exec)) {
+ node.consumeResourcesforTask(exec,
topologies.getById(entry.getKey()));
+ } else {
+ LOG.warn("Resource Req not found...Scheduling
Task{} with memory requirement as on heap - {} and off heap - {} and CPU
requirement as {}",
--- End diff --
Too long, break (new line)
> Add priorities and per user resource guarantees to Resource Aware Scheduler
> ---------------------------------------------------------------------------
>
> Key: STORM-898
> URL: https://issues.apache.org/jira/browse/STORM-898
> Project: Apache Storm
> Issue Type: New Feature
> Components: storm-core
> Reporter: Robert Joseph Evans
> Assignee: Boyang Jerry Peng
> Attachments: Resource Aware Scheduler for Storm.pdf
>
>
> In a multi-tenant environment we would like to be able to give individual
> users a guarantee of how much CPU/Memory/Network they will be able to use in
> a cluster. We would also like to know which topologies a user feels are the
> most important to keep running if there are not enough resources to run all
> of their topologies.
> Each user should be able to specify if their topology is production, staging,
> or development. Within each of those categories a user should be able to give
> a topology a priority, 0 to 10 with 10 being the highest priority (or
> something like this).
> If there are not enough resources on a cluster to run a topology assume this
> topology is running using resources and find the user that is most over their
> guaranteed resources. Shoot the lowest priority topology for that user, and
> repeat until, this topology is able to run, or this topology would be the one
> shot. Ideally we don't actually shoot anything until we know that we would
> have made enough room.
> If the cluster is over-subscribed and everyone is under their guarantee, and
> this topology would not put the user over their guarantee. Shoot the lowest
> priority topology in this workers resource pool until there is enough room to
> run the topology or this topology is the one that would be shot. We might
> also want to think about what to do if we are going to shoot a production
> topology in an oversubscribed case, and perhaps we can shoot a non-production
> topology instead even if the other user is not over their guarantee.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
