You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@samza.apache.org by "Shanthoosh Venkataraman (JIRA)" <ji...@apache.org> on 2018/07/26 23:54:00 UTC

[jira] [Created] (SAMZA-1788) Introduce LocationIdProvider abstraction:

Shanthoosh Venkataraman created SAMZA-1788:
----------------------------------------------

             Summary: Introduce  LocationIdProvider abstraction:
                 Key: SAMZA-1788
                 URL: https://issues.apache.org/jira/browse/SAMZA-1788
             Project: Samza
          Issue Type: Task
            Reporter: Shanthoosh Venkataraman
            Assignee: Shanthoosh Venkataraman


Currently in standalone, hostName of the standalone processor is used as LocationId. However, for containerized environments like azure cloud, kubernetes this defaulting does  not work. Standalone processors can be launched from different kubernetes container on a physical machine(where each docker container has different locatliyID than other docker container within same machine). 

To solve this problem, we introduce locationID abstraction to support plugging in uniqueId identifying the execution environment of the processor. In case of containerized environments, LocationId is a combination of multiple fields (sliceId, containerId, hostname) instead of simple physical hostname. By default hostname will be used as LocationId(if not configured by the user).

A new abstraction LocationIdProvider will be introduced to generate locationId for a physical execution environment. All the processors of an application registered from an locationID should be able to share(read/write) their local state stores. Any custom LocationIdProvider is expected to honor this contract when generating the locationID.



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