You are viewing a plain text version of this content. The canonical link for it is here.
Posted to hdfs-dev@hadoop.apache.org by "Robert Chansler (Created) (JIRA)" <ji...@apache.org> on 2011/11/03 18:25:32 UTC
[jira] [Created] (HDFS-2535) A Model for Data Durability
A Model for Data Durability
---------------------------
Key: HDFS-2535
URL: https://issues.apache.org/jira/browse/HDFS-2535
Project: Hadoop HDFS
Issue Type: Task
Components: documentation
Reporter: Robert Chansler
Assignee: Robert Chansler
Attachments: LosingBlocks.xlsx
This is a statistical model considering data durability in HDFS in the presence of Data Node failures. The attached spreadsheet considers the probability of losing a block with three replicas in the case of uncorrelated failures of DNs. Also included is a section that looks at the consequences of simultaneous failures.
The model parameters reflect experience at Yahoo with a large cluster. But it is easy to change the parameters in the spreadsheet. Number of replicas is not a easily adjusted parameter. And while published reports (the Google papers) suggest that node failures are not really uncorrelated, this does give some practical insight into HDFS durability.
I and others have quoted from this work in the past. I thought it good to make the details conveniently available.
--
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
[jira] [Resolved] (HDFS-2535) A Model for Data Durability
Posted by "Robert Chansler (Resolved) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-2535?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Robert Chansler resolved HDFS-2535.
-----------------------------------
Resolution: Fixed
> A Model for Data Durability
> ---------------------------
>
> Key: HDFS-2535
> URL: https://issues.apache.org/jira/browse/HDFS-2535
> Project: Hadoop HDFS
> Issue Type: Task
> Components: documentation
> Reporter: Robert Chansler
> Assignee: Robert Chansler
> Attachments: LosingBlocks.xlsx
>
>
> This is a statistical model considering data durability in HDFS in the presence of Data Node failures. The attached spreadsheet considers the probability of losing a block with three replicas in the case of uncorrelated failures of DNs. Also included is a section that looks at the consequences of simultaneous failures.
> The model parameters reflect experience at Yahoo with a large cluster. But it is easy to change the parameters in the spreadsheet. Number of replicas is not a easily adjusted parameter. And while published reports (the Google papers) suggest that node failures are not really uncorrelated, this does give some practical insight into HDFS durability.
> I and others have quoted from this work in the past. I thought it good to make the details conveniently available.
--
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