[
https://issues.apache.org/jira/browse/HAMA-511?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13207741#comment-13207741
]
ChiaHung Lin commented on HAMA-511:
-----------------------------------
As I understand this strategy is called local checkpoint or independent
checkpoint. The issues of this design is that it would have domino effect[1],
resulting in the recovery process starting from the initial state and the
frequency may be higher than expected. In addition, due to lack of consistent
state for the whole system at specific time point, every individual
checkpointed data can not be garbage collected[2], indicating that the system
needs to preserve large amount of the checkpointed data so that rollback/
recovery would be possible.
Coordinated checkpoint synchronizes at specific time point in order to form a
consistent state. Although this is not a perfect solution, it is somehow
reliable compared with local checkpoint and relative simpler than
communication-induced checkpoint.
[1]. Brian Randell. System Structure for Software Fault Tolerance.
[2]. Titos Saridakis. Design Patterns for Checkpoint-Based Rollback Recovery
> Submitting heterogenous supersteps with precedence constraints on Hama
> ----------------------------------------------------------------------
>
> Key: HAMA-511
> URL: https://issues.apache.org/jira/browse/HAMA-511
> Project: Hama
> Issue Type: New Feature
> Reporter: Suraj Menon
> Priority: Minor
> Attachments: Defining supersteps for BSP.pdf
>
>
> Hama should support submission of jobs with support for:
> 1) Skipping unwanted superstep synchronization.
> 2) Run supersteps with heterogenous nature of computation
> 3) Scheduling supersteps with precedence constraints.
> An explanation of these is provided in the attachment.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
