You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@zeppelin.apache.org by "liuxun (JIRA)" <ji...@apache.org> on 2018/07/11 11:09:00 UTC

[jira] [Created] (ZEPPELIN-3611) High utilization module design

liuxun created ZEPPELIN-3611:
--------------------------------

             Summary: High utilization module design
                 Key: ZEPPELIN-3611
                 URL: https://issues.apache.org/jira/browse/ZEPPELIN-3611
             Project: Zeppelin
          Issue Type: Sub-task
          Components: zeppelin-server
    Affects Versions: 0.9.0
            Reporter: liuxun
             Fix For: 0.9.0


h3. When you create an Interpreter process in Zeppelin-Server, you can create an Interpreter process in another Zeppelin-Server from a remote server with cluster resources, and establish a session.
 # *Server resource reporting*

Zeppelin-Server collects server hardware CPU and memory usage, periodically reports to the cluster, and stores metadata in Cluster MetaData.

 # *Create an interpreter process remotely*

When Zeppelin-Server creates an Interpreter process, first find out the server with idle resources from Cluster MetaData, and create an Interpreter process remotely through Thrift.

 # *Reuse the Interpreter process*

When Zeppelin-Server creates an Interpreter process, it first looks up the Cluster MetaData to see if the required Interpreter process already exists. If it already exists, it obtains the IP and port of the Interpreter process's Thrift through the Cluster MetaData to create the user's note and Interpreter. Session association;



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)