[
https://issues.apache.org/jira/browse/STORM-898?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15042145#comment-15042145
]
ASF GitHub Bot commented on STORM-898:
--------------------------------------
Github user jerrypeng commented on a diff in the pull request:
https://github.com/apache/storm/pull/921#discussion_r46726363
--- Diff: storm-core/src/jvm/backtype/storm/scheduler/resource/User.java ---
@@ -0,0 +1,344 @@
+/**
+ * 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.scheduler.Cluster;
+import backtype.storm.scheduler.TopologyDetails;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.util.Collection;
+import java.util.Comparator;
+import java.util.HashMap;
+import java.util.LinkedList;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+import java.util.TreeSet;
+
+public class User {
+ private String userId;
+ //Topologies yet to be scheduled sorted by priority for each user
+ private TreeSet<TopologyDetails> pendingQueue = new
TreeSet<TopologyDetails>(new PQsortByPriorityAndSubmittionTime());
+
+ //Topologies yet to be scheduled sorted by priority for each user
+ private TreeSet<TopologyDetails> runningQueue = new
TreeSet<TopologyDetails>(new PQsortByPriorityAndSubmittionTime());
+
+ //Topologies that was attempted to be scheduled but wasn't successull
+ private TreeSet<TopologyDetails> attemptedQueue = new
TreeSet<TopologyDetails>(new PQsortByPriorityAndSubmittionTime());
+
+ private TreeSet<TopologyDetails> invalidQueue = new
TreeSet<TopologyDetails>(new PQsortByPriorityAndSubmittionTime());
+
+ private Map<String, Double> resourcePool = new HashMap<String,
Double>();
+
+ private static final Logger LOG = LoggerFactory.getLogger(User.class);
+
+ public User(String userId) {
+ this.userId = userId;
+ }
+
+ public User(String userId, Map<String, Double> resourcePool) {
+ this(userId);
+ if (resourcePool != null) {
+ this.resourcePool.putAll(resourcePool);
+ }
+ if (this.resourcePool.get("cpu") == null) {
+ this.resourcePool.put("cpu", 0.0);
+ }
+ if (this.resourcePool.get("memory") == null) {
+ this.resourcePool.put("memory", 0.0);
+ }
+ }
+
+ public User getCopy() {
+ User newUser = new User(this.userId, this.resourcePool);
+ for (TopologyDetails topo : this.pendingQueue) {
+ newUser.addTopologyToPendingQueue(topo);
+ }
+ for (TopologyDetails topo : this.runningQueue) {
+ newUser.addTopologyToRunningQueue(topo);
+ }
+ for (TopologyDetails topo : this.attemptedQueue) {
+ newUser.addTopologyToAttemptedQueue(topo);
+ }
+ for (TopologyDetails topo : this.invalidQueue) {
+ newUser.addTopologyToInvalidQueue(topo);
+ }
+ return newUser;
+ }
+
+ public String getId() {
+ return this.userId;
+ }
+
+ public void addTopologyToPendingQueue(TopologyDetails topo, Cluster
cluster) {
+ this.pendingQueue.add(topo);
+ if (cluster != null) {
+ cluster.setStatus(topo.getId(), "Scheduling Pending");
+ }
+ }
+
+ public void addTopologyToPendingQueue(TopologyDetails topo) {
+ this.addTopologyToPendingQueue(topo, null);
+ }
+
+ public void addTopologyToRunningQueue(TopologyDetails topo, Cluster
cluster) {
+ this.runningQueue.add(topo);
+ if (cluster != null) {
+ cluster.setStatus(topo.getId(), "Fully Scheduled");
+ }
+ }
+
+ public void addTopologyToRunningQueue(TopologyDetails topo) {
+ this.addTopologyToRunningQueue(topo, null);
+ }
+
+ public Set<TopologyDetails> getTopologiesPending() {
+ TreeSet<TopologyDetails> ret = new TreeSet<TopologyDetails>(new
PQsortByPriorityAndSubmittionTime());
+ ret.addAll(this.pendingQueue);
+ return ret;
+ }
+
+ public void addTopologyToAttemptedQueue(TopologyDetails topo) {
+ this.attemptedQueue.add(topo);
+ }
+
+ public void addTopologyToInvalidQueue(TopologyDetails topo) {
+ this.invalidQueue.add(topo);
+ }
+
+ public Set<TopologyDetails> getTopologiesRunning() {
+ TreeSet<TopologyDetails> ret = new TreeSet<TopologyDetails>(new
PQsortByPriorityAndSubmittionTime());
+ ret.addAll(this.runningQueue);
+ return ret;
+ }
+
+ public Set<TopologyDetails> getTopologiesAttempted() {
+ TreeSet<TopologyDetails> ret = new TreeSet<TopologyDetails>(new
PQsortByPriorityAndSubmittionTime());
+ ret.addAll(this.attemptedQueue);
+ return ret;
+ }
+
+ public Set<TopologyDetails> getTopologiesInvalid() {
+ TreeSet<TopologyDetails> ret = new TreeSet<TopologyDetails>(new
PQsortByPriorityAndSubmittionTime());
+ ret.addAll(this.invalidQueue);
+ return ret;
+ }
+
+ public Map<String, Number> getResourcePool() {
+ if (this.resourcePool != null) {
+ return new HashMap<String, Number>(this.resourcePool);
+ }
+ return null;
+ }
+
+ public void moveTopoFromPendingToRunning(TopologyDetails topo, Cluster
cluster) {
+ moveTopology(topo, this.pendingQueue, "pending",
this.runningQueue, "running");
+ if (cluster != null) {
+ cluster.setStatus(topo.getId(), "Fully Scheduled");
+ }
+ }
+
+ public void moveTopoFromPendingToRunning(TopologyDetails topo) {
+ this.moveTopoFromPendingToRunning(topo, null);
+ }
+
+
+ public void moveTopoFromPendingToAttempted(TopologyDetails topo,
Cluster cluster) {
+ moveTopology(topo, this.pendingQueue, "pending",
this.attemptedQueue, "attempted");
+ if (cluster != null) {
+ cluster.setStatus(topo.getId(), "Scheduling Attempted but
Failed");
+ }
+ }
+
+ public void moveTopoFromPendingToAttempted(TopologyDetails topo) {
+ this.moveTopoFromPendingToAttempted(topo, null);
+ }
+
+
+ public void moveTopoFromPendingToInvalid(TopologyDetails topo, Cluster
cluster) {
+ moveTopology(topo, this.pendingQueue, "pending",
this.invalidQueue, "invalid");
+ if (cluster != null) {
+ cluster.setStatus(topo.getId(), "Scheduling Attempted but
topology is invalid");
+ }
+ }
+
+ public void moveTopoFromPendingToInvalid(TopologyDetails topo) {
+ this.moveTopoFromPendingToInvalid(topo, null);
+ }
+
+
+ public void moveTopoFromRunningToPending(TopologyDetails topo, Cluster
cluster) {
+ moveTopology(topo, this.runningQueue, "running",
this.pendingQueue, "pending");
+ if (cluster != null) {
+ cluster.setStatus(topo.getId(), "Scheduling Pending");
+ }
+ }
+
+ public void moveTopoFromRunningToPending(TopologyDetails topo) {
+ this.moveTopoFromRunningToPending(topo, null);
+ }
+
+
+ private void moveTopology(TopologyDetails topo, Set<TopologyDetails>
src, String srcName, Set<TopologyDetails> dest, String destName) {
--- End diff --
I am not quite following what you are suggesting. Can you explain your
suggestion in more detail. Thanks!
> 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)
