[jira] [Commented] (YARN-2768) Avoid cloning Resource in FSAppAttempt#updateDemand
[
https://issues.apache.org/jira/browse/YARN-2768?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15144100#comment-15144100
]
Sangjin Lee commented on YARN-2768:
---
This looks like a great candidate to backport to 2.7.x and 2.6.x.
> Avoid cloning Resource in FSAppAttempt#updateDemand
> ---
>
> Key: YARN-2768
> URL: https://issues.apache.org/jira/browse/YARN-2768
> Project: Hadoop YARN
> Issue Type: Improvement
> Components: fairscheduler
>Reporter: Hong Zhiguo
>Assignee: Hong Zhiguo
>Priority: Minor
> Fix For: 2.8.0
>
> Attachments: YARN-2768.patch, profiling_FairScheduler_update.png
>
>
> See the attached picture of profiling result. The clone of Resource object
> within Resources.multiply() takes up **85%** (19.2 / 22.6) CPU time of the
> function FairScheduler.update().
> The code of FSAppAttempt.updateDemand:
> {code}
> public void updateDemand() {
> demand = Resources.createResource(0);
> // Demand is current consumption plus outstanding requests
> Resources.addTo(demand, app.getCurrentConsumption());
> // Add up outstanding resource requests
> synchronized (app) {
> for (Priority p : app.getPriorities()) {
> for (ResourceRequest r : app.getResourceRequests(p).values()) {
> Resource total = Resources.multiply(r.getCapability(),
> r.getNumContainers());
> Resources.addTo(demand, total);
> }
> }
> }
> }
> {code}
> The code of Resources.multiply:
> {code}
> public static Resource multiply(Resource lhs, double by) {
> return multiplyTo(clone(lhs), by);
> }
> {code}
> The clone could be skipped by directly update the value of this.demand.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (YARN-2768) Avoid cloning Resource in FSAppAttempt#updateDemand
[
https://issues.apache.org/jira/browse/YARN-2768?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14647578#comment-14647578
]
Hudson commented on YARN-2768:
--
FAILURE: Integrated in Hadoop-Hdfs-trunk-Java8 #261 (See
[https://builds.apache.org/job/Hadoop-Hdfs-trunk-Java8/261/])
YARN-2768. Avoid cloning Resource in FSAppAttempt#updateDemand. (Hong Zhiguo
via kasha) (kasha: rev 5205a330b387d2e133ee790b9fe7d5af3cd8bccc)
*
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/main/java/org/apache/hadoop/yarn/util/resource/Resources.java
*
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FSAppAttempt.java
* hadoop-yarn-project/CHANGES.txt
Avoid cloning Resource in FSAppAttempt#updateDemand
---
Key: YARN-2768
URL: https://issues.apache.org/jira/browse/YARN-2768
Project: Hadoop YARN
Issue Type: Improvement
Components: fairscheduler
Reporter: Hong Zhiguo
Assignee: Hong Zhiguo
Priority: Minor
Fix For: 2.8.0
Attachments: YARN-2768.patch, profiling_FairScheduler_update.png
See the attached picture of profiling result. The clone of Resource object
within Resources.multiply() takes up **85%** (19.2 / 22.6) CPU time of the
function FairScheduler.update().
The code of FSAppAttempt.updateDemand:
{code}
public void updateDemand() {
demand = Resources.createResource(0);
// Demand is current consumption plus outstanding requests
Resources.addTo(demand, app.getCurrentConsumption());
// Add up outstanding resource requests
synchronized (app) {
for (Priority p : app.getPriorities()) {
for (ResourceRequest r : app.getResourceRequests(p).values()) {
Resource total = Resources.multiply(r.getCapability(),
r.getNumContainers());
Resources.addTo(demand, total);
}
}
}
}
{code}
The code of Resources.multiply:
{code}
public static Resource multiply(Resource lhs, double by) {
return multiplyTo(clone(lhs), by);
}
{code}
The clone could be skipped by directly update the value of this.demand.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (YARN-2768) Avoid cloning Resource in FSAppAttempt#updateDemand
[
https://issues.apache.org/jira/browse/YARN-2768?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14647497#comment-14647497
]
Hudson commented on YARN-2768:
--
FAILURE: Integrated in Hadoop-Yarn-trunk-Java8 #272 (See
[https://builds.apache.org/job/Hadoop-Yarn-trunk-Java8/272/])
YARN-2768. Avoid cloning Resource in FSAppAttempt#updateDemand. (Hong Zhiguo
via kasha) (kasha: rev 5205a330b387d2e133ee790b9fe7d5af3cd8bccc)
* hadoop-yarn-project/CHANGES.txt
*
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FSAppAttempt.java
*
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/main/java/org/apache/hadoop/yarn/util/resource/Resources.java
Avoid cloning Resource in FSAppAttempt#updateDemand
---
Key: YARN-2768
URL: https://issues.apache.org/jira/browse/YARN-2768
Project: Hadoop YARN
Issue Type: Improvement
Components: fairscheduler
Reporter: Hong Zhiguo
Assignee: Hong Zhiguo
Priority: Minor
Fix For: 2.8.0
Attachments: YARN-2768.patch, profiling_FairScheduler_update.png
See the attached picture of profiling result. The clone of Resource object
within Resources.multiply() takes up **85%** (19.2 / 22.6) CPU time of the
function FairScheduler.update().
The code of FSAppAttempt.updateDemand:
{code}
public void updateDemand() {
demand = Resources.createResource(0);
// Demand is current consumption plus outstanding requests
Resources.addTo(demand, app.getCurrentConsumption());
// Add up outstanding resource requests
synchronized (app) {
for (Priority p : app.getPriorities()) {
for (ResourceRequest r : app.getResourceRequests(p).values()) {
Resource total = Resources.multiply(r.getCapability(),
r.getNumContainers());
Resources.addTo(demand, total);
}
}
}
}
{code}
The code of Resources.multiply:
{code}
public static Resource multiply(Resource lhs, double by) {
return multiplyTo(clone(lhs), by);
}
{code}
The clone could be skipped by directly update the value of this.demand.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (YARN-2768) Avoid cloning Resource in FSAppAttempt#updateDemand
[
https://issues.apache.org/jira/browse/YARN-2768?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14647824#comment-14647824
]
Hudson commented on YARN-2768:
--
SUCCESS: Integrated in Hadoop-Mapreduce-trunk #2218 (See
[https://builds.apache.org/job/Hadoop-Mapreduce-trunk/2218/])
YARN-2768. Avoid cloning Resource in FSAppAttempt#updateDemand. (Hong Zhiguo
via kasha) (kasha: rev 5205a330b387d2e133ee790b9fe7d5af3cd8bccc)
*
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FSAppAttempt.java
*
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/main/java/org/apache/hadoop/yarn/util/resource/Resources.java
* hadoop-yarn-project/CHANGES.txt
Avoid cloning Resource in FSAppAttempt#updateDemand
---
Key: YARN-2768
URL: https://issues.apache.org/jira/browse/YARN-2768
Project: Hadoop YARN
Issue Type: Improvement
Components: fairscheduler
Reporter: Hong Zhiguo
Assignee: Hong Zhiguo
Priority: Minor
Fix For: 2.8.0
Attachments: YARN-2768.patch, profiling_FairScheduler_update.png
See the attached picture of profiling result. The clone of Resource object
within Resources.multiply() takes up **85%** (19.2 / 22.6) CPU time of the
function FairScheduler.update().
The code of FSAppAttempt.updateDemand:
{code}
public void updateDemand() {
demand = Resources.createResource(0);
// Demand is current consumption plus outstanding requests
Resources.addTo(demand, app.getCurrentConsumption());
// Add up outstanding resource requests
synchronized (app) {
for (Priority p : app.getPriorities()) {
for (ResourceRequest r : app.getResourceRequests(p).values()) {
Resource total = Resources.multiply(r.getCapability(),
r.getNumContainers());
Resources.addTo(demand, total);
}
}
}
}
{code}
The code of Resources.multiply:
{code}
public static Resource multiply(Resource lhs, double by) {
return multiplyTo(clone(lhs), by);
}
{code}
The clone could be skipped by directly update the value of this.demand.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (YARN-2768) Avoid cloning Resource in FSAppAttempt#updateDemand
[
https://issues.apache.org/jira/browse/YARN-2768?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14647692#comment-14647692
]
Hudson commented on YARN-2768:
--
SUCCESS: Integrated in Hadoop-Hdfs-trunk #2199 (See
[https://builds.apache.org/job/Hadoop-Hdfs-trunk/2199/])
YARN-2768. Avoid cloning Resource in FSAppAttempt#updateDemand. (Hong Zhiguo
via kasha) (kasha: rev 5205a330b387d2e133ee790b9fe7d5af3cd8bccc)
* hadoop-yarn-project/CHANGES.txt
*
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/main/java/org/apache/hadoop/yarn/util/resource/Resources.java
*
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FSAppAttempt.java
Avoid cloning Resource in FSAppAttempt#updateDemand
---
Key: YARN-2768
URL: https://issues.apache.org/jira/browse/YARN-2768
Project: Hadoop YARN
Issue Type: Improvement
Components: fairscheduler
Reporter: Hong Zhiguo
Assignee: Hong Zhiguo
Priority: Minor
Fix For: 2.8.0
Attachments: YARN-2768.patch, profiling_FairScheduler_update.png
See the attached picture of profiling result. The clone of Resource object
within Resources.multiply() takes up **85%** (19.2 / 22.6) CPU time of the
function FairScheduler.update().
The code of FSAppAttempt.updateDemand:
{code}
public void updateDemand() {
demand = Resources.createResource(0);
// Demand is current consumption plus outstanding requests
Resources.addTo(demand, app.getCurrentConsumption());
// Add up outstanding resource requests
synchronized (app) {
for (Priority p : app.getPriorities()) {
for (ResourceRequest r : app.getResourceRequests(p).values()) {
Resource total = Resources.multiply(r.getCapability(),
r.getNumContainers());
Resources.addTo(demand, total);
}
}
}
}
{code}
The code of Resources.multiply:
{code}
public static Resource multiply(Resource lhs, double by) {
return multiplyTo(clone(lhs), by);
}
{code}
The clone could be skipped by directly update the value of this.demand.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (YARN-2768) Avoid cloning Resource in FSAppAttempt#updateDemand
[
https://issues.apache.org/jira/browse/YARN-2768?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14647783#comment-14647783
]
Hudson commented on YARN-2768:
--
SUCCESS: Integrated in Hadoop-Mapreduce-trunk-Java8 #269 (See
[https://builds.apache.org/job/Hadoop-Mapreduce-trunk-Java8/269/])
YARN-2768. Avoid cloning Resource in FSAppAttempt#updateDemand. (Hong Zhiguo
via kasha) (kasha: rev 5205a330b387d2e133ee790b9fe7d5af3cd8bccc)
*
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FSAppAttempt.java
* hadoop-yarn-project/CHANGES.txt
*
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/main/java/org/apache/hadoop/yarn/util/resource/Resources.java
Avoid cloning Resource in FSAppAttempt#updateDemand
---
Key: YARN-2768
URL: https://issues.apache.org/jira/browse/YARN-2768
Project: Hadoop YARN
Issue Type: Improvement
Components: fairscheduler
Reporter: Hong Zhiguo
Assignee: Hong Zhiguo
Priority: Minor
Fix For: 2.8.0
Attachments: YARN-2768.patch, profiling_FairScheduler_update.png
See the attached picture of profiling result. The clone of Resource object
within Resources.multiply() takes up **85%** (19.2 / 22.6) CPU time of the
function FairScheduler.update().
The code of FSAppAttempt.updateDemand:
{code}
public void updateDemand() {
demand = Resources.createResource(0);
// Demand is current consumption plus outstanding requests
Resources.addTo(demand, app.getCurrentConsumption());
// Add up outstanding resource requests
synchronized (app) {
for (Priority p : app.getPriorities()) {
for (ResourceRequest r : app.getResourceRequests(p).values()) {
Resource total = Resources.multiply(r.getCapability(),
r.getNumContainers());
Resources.addTo(demand, total);
}
}
}
}
{code}
The code of Resources.multiply:
{code}
public static Resource multiply(Resource lhs, double by) {
return multiplyTo(clone(lhs), by);
}
{code}
The clone could be skipped by directly update the value of this.demand.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (YARN-2768) Avoid cloning Resource in FSAppAttempt#updateDemand
[
https://issues.apache.org/jira/browse/YARN-2768?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14647447#comment-14647447
]
Hudson commented on YARN-2768:
--
FAILURE: Integrated in Hadoop-Yarn-trunk #1002 (See
[https://builds.apache.org/job/Hadoop-Yarn-trunk/1002/])
YARN-2768. Avoid cloning Resource in FSAppAttempt#updateDemand. (Hong Zhiguo
via kasha) (kasha: rev 5205a330b387d2e133ee790b9fe7d5af3cd8bccc)
* hadoop-yarn-project/CHANGES.txt
*
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/main/java/org/apache/hadoop/yarn/util/resource/Resources.java
*
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FSAppAttempt.java
Avoid cloning Resource in FSAppAttempt#updateDemand
---
Key: YARN-2768
URL: https://issues.apache.org/jira/browse/YARN-2768
Project: Hadoop YARN
Issue Type: Improvement
Components: fairscheduler
Reporter: Hong Zhiguo
Assignee: Hong Zhiguo
Priority: Minor
Fix For: 2.8.0
Attachments: YARN-2768.patch, profiling_FairScheduler_update.png
See the attached picture of profiling result. The clone of Resource object
within Resources.multiply() takes up **85%** (19.2 / 22.6) CPU time of the
function FairScheduler.update().
The code of FSAppAttempt.updateDemand:
{code}
public void updateDemand() {
demand = Resources.createResource(0);
// Demand is current consumption plus outstanding requests
Resources.addTo(demand, app.getCurrentConsumption());
// Add up outstanding resource requests
synchronized (app) {
for (Priority p : app.getPriorities()) {
for (ResourceRequest r : app.getResourceRequests(p).values()) {
Resource total = Resources.multiply(r.getCapability(),
r.getNumContainers());
Resources.addTo(demand, total);
}
}
}
}
{code}
The code of Resources.multiply:
{code}
public static Resource multiply(Resource lhs, double by) {
return multiplyTo(clone(lhs), by);
}
{code}
The clone could be skipped by directly update the value of this.demand.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (YARN-2768) Avoid cloning Resource in FSAppAttempt#updateDemand
[
https://issues.apache.org/jira/browse/YARN-2768?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14646401#comment-14646401
]
Hudson commented on YARN-2768:
--
FAILURE: Integrated in Hadoop-trunk-Commit #8241 (See
[https://builds.apache.org/job/Hadoop-trunk-Commit/8241/])
YARN-2768. Avoid cloning Resource in FSAppAttempt#updateDemand. (Hong Zhiguo
via kasha) (kasha: rev 5205a330b387d2e133ee790b9fe7d5af3cd8bccc)
*
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FSAppAttempt.java
*
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/main/java/org/apache/hadoop/yarn/util/resource/Resources.java
* hadoop-yarn-project/CHANGES.txt
Avoid cloning Resource in FSAppAttempt#updateDemand
---
Key: YARN-2768
URL: https://issues.apache.org/jira/browse/YARN-2768
Project: Hadoop YARN
Issue Type: Improvement
Components: fairscheduler
Reporter: Hong Zhiguo
Assignee: Hong Zhiguo
Priority: Minor
Fix For: 2.8.0
Attachments: YARN-2768.patch, profiling_FairScheduler_update.png
See the attached picture of profiling result. The clone of Resource object
within Resources.multiply() takes up **85%** (19.2 / 22.6) CPU time of the
function FairScheduler.update().
The code of FSAppAttempt.updateDemand:
{code}
public void updateDemand() {
demand = Resources.createResource(0);
// Demand is current consumption plus outstanding requests
Resources.addTo(demand, app.getCurrentConsumption());
// Add up outstanding resource requests
synchronized (app) {
for (Priority p : app.getPriorities()) {
for (ResourceRequest r : app.getResourceRequests(p).values()) {
Resource total = Resources.multiply(r.getCapability(),
r.getNumContainers());
Resources.addTo(demand, total);
}
}
}
}
{code}
The code of Resources.multiply:
{code}
public static Resource multiply(Resource lhs, double by) {
return multiplyTo(clone(lhs), by);
}
{code}
The clone could be skipped by directly update the value of this.demand.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
