[
https://issues.apache.org/jira/browse/HADOOP-13327?focusedWorklogId=513677&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-513677
]
ASF GitHub Bot logged work on HADOOP-13327:
-------------------------------------------
Author: ASF GitHub Bot
Created on: 18/Nov/20 18:10
Start Date: 18/Nov/20 18:10
Worklog Time Spent: 10m
Work Description: joshelser commented on pull request #2102:
URL: https://github.com/apache/hadoop/pull/2102#issuecomment-729861275
Yikes. These got mis-filed in my mail. I'm sorry, Steve.
> Declare that HSYNC is sufficient and the only one you should look for. If
you implement HFLUSH and not HSYNC, you are of no use to applications. retain
support for HFLUSH probe, but in the spec only mention it as of historical
interest.
My only worry with this is that, today, HBase is only running with HFlush
and not HSync because of performance reasons. The last wag I remember was in
the ballpark of a 30% performance hit for writes if we force hsync instead of
hflush. Let me re-read your latest and see if my mind changes.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
Issue Time Tracking
-------------------
Worklog Id: (was: 513677)
Time Spent: 3h 50m (was: 3h 40m)
> Add OutputStream + Syncable to the Filesystem Specification
> -----------------------------------------------------------
>
> Key: HADOOP-13327
> URL: https://issues.apache.org/jira/browse/HADOOP-13327
> Project: Hadoop Common
> Issue Type: Sub-task
> Components: fs
> Affects Versions: 2.8.0
> Reporter: Steve Loughran
> Assignee: Steve Loughran
> Priority: Major
> Labels: pull-request-available
> Attachments: HADOOP-13327-002.patch, HADOOP-13327-003.patch,
> HADOOP-13327-branch-2-001.patch
>
> Time Spent: 3h 50m
> Remaining Estimate: 0h
>
> Write down what a Filesystem output stream should do. While core the API is
> defined in Java, that doesn't say what's expected about visibility,
> durability, etc —and Hadoop Syncable interface is entirely ours to define.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
