Spark Streaming AutoScaling not always taking effect even with available resources
I am trying to setup Autoscaling for our Spark Streaming jobs running on YARN along the lines of Elastic Spark Streaming. fwiw, I decided against using Spark's internal Auto-scaling because we needed to aggressively scale up/down than mentioned there along with a few other parameters to tune the scaling but I did use similar code to perform the scaling.
So, I use the Yarn Scheduler Backend instance I retrieve from the SparkContext at the beginning of the job to perform the scaling operations periodically. What I see is that YARN does not respond to the scale up requests immediately even when it has plenty of resources left (both in terms of cores and memory). It does respond after about an hour of the jobs functioning (after about 5-6 invocations of requestTotalExecutors) and the behavior is not even deterministic here in terms of when it starts. The scale down requests by calling killExecutors(Seq) still works as expected - so its only the scale up thats not working for some reason.
I am not sure if there is some configuration on YARN or Spark that I missed which is causing this weird behavior and am here for help.
apache-spark spark-streaming yarn autoscaling
asked Nov 19 '18 at 21:23
noobNeverythingnoobNeverything
55110
add a comment |
I am trying to setup Autoscaling for our Spark Streaming jobs running on YARN along the lines of Elastic Spark Streaming. fwiw, I decided against using Spark's internal Auto-scaling because we needed to aggressively scale up/down than mentioned there along with a few other parameters to tune the scaling but I did use similar code to perform the scaling.
So, I use the Yarn Scheduler Backend instance I retrieve from the SparkContext at the beginning of the job to perform the scaling operations periodically. What I see is that YARN does not respond to the scale up requests immediately even when it has plenty of resources left (both in terms of cores and memory). It does respond after about an hour of the jobs functioning (after about 5-6 invocations of requestTotalExecutors) and the behavior is not even deterministic here in terms of when it starts. The scale down requests by calling killExecutors(Seq) still works as expected - so its only the scale up thats not working for some reason.
I am not sure if there is some configuration on YARN or Spark that I missed which is causing this weird behavior and am here for help.
apache-spark spark-streaming yarn autoscaling
asked Nov 19 '18 at 21:23
noobNeverythingnoobNeverything
55110
add a comment |
I am trying to setup Autoscaling for our Spark Streaming jobs running on YARN along the lines of Elastic Spark Streaming. fwiw, I decided against using Spark's internal Auto-scaling because we needed to aggressively scale up/down than mentioned there along with a few other parameters to tune the scaling but I did use similar code to perform the scaling.
So, I use the Yarn Scheduler Backend instance I retrieve from the SparkContext at the beginning of the job to perform the scaling operations periodically. What I see is that YARN does not respond to the scale up requests immediately even when it has plenty of resources left (both in terms of cores and memory). It does respond after about an hour of the jobs functioning (after about 5-6 invocations of requestTotalExecutors) and the behavior is not even deterministic here in terms of when it starts. The scale down requests by calling killExecutors(Seq) still works as expected - so its only the scale up thats not working for some reason.
I am not sure if there is some configuration on YARN or Spark that I missed which is causing this weird behavior and am here for help.
apache-spark spark-streaming yarn autoscaling
asked Nov 19 '18 at 21:23
noobNeverythingnoobNeverything
55110
I am trying to setup Autoscaling for our Spark Streaming jobs running on YARN along the lines of Elastic Spark Streaming. fwiw, I decided against using Spark's internal Auto-scaling because we needed to aggressively scale up/down than mentioned there along with a few other parameters to tune the scaling but I did use similar code to perform the scaling.
So, I use the Yarn Scheduler Backend instance I retrieve from the SparkContext at the beginning of the job to perform the scaling operations periodically. What I see is that YARN does not respond to the scale up requests immediately even when it has plenty of resources left (both in terms of cores and memory). It does respond after about an hour of the jobs functioning (after about 5-6 invocations of requestTotalExecutors) and the behavior is not even deterministic here in terms of when it starts. The scale down requests by calling killExecutors(Seq) still works as expected - so its only the scale up thats not working for some reason.
I am not sure if there is some configuration on YARN or Spark that I missed which is causing this weird behavior and am here for help.
apache-spark spark-streaming yarn autoscaling
apache-spark spark-streaming yarn autoscaling
asked Nov 19 '18 at 21:23
noobNeverythingnoobNeverything
55110
asked Nov 19 '18 at 21:23
noobNeverythingnoobNeverything
55110
edited Nov 19 '18 at 22:44
noobNeverything
asked Nov 19 '18 at 21:23
noobNeverythingnoobNeverything
55110
asked Nov 19 '18 at 21:23
noobNeverythingnoobNeverything
55110
asked Nov 19 '18 at 21:23
noobNeverythingnoobNeverything
55110
55110
add a comment |
add a comment |
0
active
oldest
votes
Your Answer
StackExchange.ifUsing("editor", function () {
StackExchange.using("externalEditor", function () {
StackExchange.using("snippets", function () {
StackExchange.snippets.init();
});
});
}, "code-snippets");
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "1"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});
function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});
}
});
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53382849%2fspark-streaming-autoscaling-not-always-taking-effect-even-with-available-resourc%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
0
active
oldest
votes
0
active
oldest
votes
active
oldest
votes
active
oldest
votes
Thanks for contributing an answer to Stack Overflow!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53382849%2fspark-streaming-autoscaling-not-always-taking-effect-even-with-available-resourc%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
