Project

General

Profile

Task #1687

Story #975: Refactor to support synchronization across multiple MNs

Distributed job execution

Added by Robert Waltz over 12 years ago. Updated over 12 years ago.

Status:
Closed
Priority:
Normal
Assignee:
Robert Waltz
Category:
d1_synchronization
Target version:
Start date:
2011-07-20
Due date:
% Done:

100%

Milestone:
CCI-0.6.2
Product Version:
*
Story Points:
Sprint:

Description

A job will execute on a given schedule. Though a trigger will execute on a home machine, the execution of the job may be distributed to a remote machine. Hazelcast allows for distributed executor service.

SyncObjectExecutor creates a single thread on each CN that runs a SyncObjectTask. SyncObjectTask polls on the syncObjectQueue (a hazelcast queue) for synchronization tasks to run. Each CN will run a maximum configurable set of tasks for each membernode being synchronized. A sync task is associated with an object. For each object that needs synchronization a TransferObjectTask thread is executed.

an object that fails to synchronize is submitted to the MN via the sync failure interface.

History

#1 Updated by Robert Waltz over 12 years ago

  • Status changed from New to Closed

Also available in: Atom PDF

Add picture from clipboard (Maximum size: 14.8 MB)