Skip to content

Conversation

@DifferentSC
Copy link
Contributor

This PR resolves #1131 via

  • Implement recovery-based scaling out.
  • Add necessary avro formats and function calls.

@DifferentSC DifferentSC requested a review from taegeonum May 21, 2018 13:40
Copy link
Contributor

@taegeonum taegeonum left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I've left some comments. Thanks!

"type": "double"
},
{
"name": "NumEventProcessors",
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why do we need this information?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can't we send the group load info from Task by dividing the load by the number of event processors?

innerList.add(taskHostname);
return innerMap.putIfAbsent(taskHostname, TaskStats.newBuilder()
.setTaskLoad(0.0)
.setNumEventProcessors(1)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

why is set to 1?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

What happen if a task fails before it is updated?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Implement basic automatic scaling-out

2 participants