couchdb-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Paul Joseph Davis (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (COUCHDB-2097) Avoid performance regression with a single fd
Date Mon, 24 Feb 2014 17:26:29 GMT

    [ https://issues.apache.org/jira/browse/COUCHDB-2097?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13910535#comment-13910535
] 

Paul Joseph Davis commented on COUCHDB-2097:
--------------------------------------------

Just checked and it looks like the majority of the performance analysis has been reported
here:

https://gist.github.com/mikewallace1979/8ed91d91443f9d1ee828

Mike's general conclusions:

The `couch_file` and `couch_server` optimizations provide faster reads under concurrent read/write
load than the original multiple fd optimization in `couch_db_updater` for large and small
documents. Writes are slower than with the multiple fd optimization however they are significantly
faster than when using a single updater fd with no optimizations.

I'll also attach the entire long form analysis to this ticket so we have a copy for posterity.

> Avoid performance regression with a single fd
> ---------------------------------------------
>
>                 Key: COUCHDB-2097
>                 URL: https://issues.apache.org/jira/browse/COUCHDB-2097
>             Project: CouchDB
>          Issue Type: Improvement
>      Security Level: public(Regular issues) 
>          Components: BigCouch
>            Reporter: Paul Joseph Davis
>              Labels: release
>
> Part of one of our large enhancements required that we remove a CouchDB performance optimization
on having two file handles to each .couch file. We need to make sure that this doesn't negatively
impact performance.



--
This message was sent by Atlassian JIRA
(v6.1.5#6160)

Mime
View raw message