couchdb-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Randall Leeds (JIRA)" <j...@apache.org>
Subject [jira] Commented: (COUCHDB-1027) More efficient implementation of the DB updater BTree functions
Date Mon, 17 Jan 2011 20:37:43 GMT

    [ https://issues.apache.org/jira/browse/COUCHDB-1027?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12982854#action_12982854
] 

Randall Leeds commented on COUCHDB-1027:
----------------------------------------

Nice! I bet there's more like this lurking around.

> More efficient implementation of the DB updater BTree functions
> ---------------------------------------------------------------
>
>                 Key: COUCHDB-1027
>                 URL: https://issues.apache.org/jira/browse/COUCHDB-1027
>             Project: CouchDB
>          Issue Type: Improvement
>          Components: Database Core
>            Reporter: Filipe Manana
>            Assignee: Filipe Manana
>             Fix For: 1.1, 1.2, 2.0
>
>         Attachments: COUCHDB-1027.patch
>
>
> Currently, some of the BTree functions (by_id BTree reduce function and and by_seq BTree
split function) traverse the same input lists several times to produce the output. This is
not optimal. For example, the btree_by_id_reduce function, for a database with 300 000+ documents,
can easily get a list of 50 or more #full_doc_info records. This functions traverses that
list 4 times - 2 filtered list comprehensions plus two length/1 calls.
> The following patch changes the implementation to produce the same output with a single
list traversal operation.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message