-
Notifications
You must be signed in to change notification settings - Fork 5.5k
feat(server): Add http support for internal resource manager communication #26635
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: master
Are you sure you want to change the base?
Changes from all commits
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -161,6 +161,8 @@ At least one resource manager is needed for a cluster, and more can be added to | |
| discovery-server.enabled=true | ||
| discovery.uri=http://example.net:8080 (Point to resource manager host/vip) | ||
| thrift.server.ssl.enabled=true | ||
| resource-manager.http-server-enabled=false | ||
| internal-communication.resource-manager-communication-protocol=THRIFT | ||
|
|
||
| * ``Coordinator`` | ||
|
|
||
|
|
@@ -175,6 +177,7 @@ A cluster can have a pool of coordinators. Each coordinator will run a subset of | |
| query.max-memory-per-node=1GB | ||
| discovery.uri=http://example.net:8080 (Point to resource manager host/vip) | ||
| resource-manager-enabled=true | ||
| internal-communication.resource-manager-communication-protocol=THRIFT | ||
|
|
||
| * ``Worker`` | ||
|
|
||
|
|
@@ -188,6 +191,7 @@ A cluster can have a pool of workers. They send their heartbeats to the resource | |
| query.max-memory-per-node=1GB | ||
| discovery.uri=http://example.net:8080 (Point to resource manager host/vip) | ||
| resource-manager-enabled=true | ||
| internal-communication.resource-manager-communication-protocol=THRIFT | ||
|
|
||
| These properties require some explanation: | ||
|
|
||
|
|
@@ -231,6 +235,15 @@ These properties require some explanation: | |
| the host and port of the Presto coordinator. This URI must not end | ||
| in a slash. | ||
|
|
||
| * ``internal-communication.resource-manager-communication-protocol``: | ||
| The protocol used for communication with the resource manager. This | ||
| can be set to ``THRIFT`` or ``HTTP``. | ||
|
|
||
| * ``resource-manager.http-server-enabled``: | ||
|
Contributor
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Could we use the value of
Contributor
Author
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. The issue is that the C++ workers use the HTTP protocol regardless if the coordinator communicates in thrift. So you could want to keep the thrift implementation & keep the http server on. We could have |
||
| Whether to enable the resource manager HTTP server or not. If | ||
| ``internal-communication.resource-manager-communication-protocol=HTTP``, this | ||
| must be set to ``true``. | ||
|
|
||
| The following flags can help one tune the disaggregated coordinator cluster’s resource groups to the desired consistency: | ||
|
|
||
| * ``concurrency-threshold-to-enable-resource-group-refresh (default: 1.0)`` | ||
|
|
||
Uh oh!
There was an error while loading. Please reload this page.