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-1130) binary optimization in couch_file
Date Mon, 18 Apr 2011 22:31:06 GMT

    [ https://issues.apache.org/jira/browse/COUCHDB-1130?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13021318#comment-13021318
] 

Randall Leeds commented on COUCHDB-1130:
----------------------------------------

@Paul: I have no idea.
And looking at the docs I think I understand a bit better now.

I think what we're doing is akin to the second code example in http://www.erlang.org/doc/efficiency_guide/binaryhandling.html#id64095
Except, according to that excerpt, even if we switched the argument order (as in my patch),
we avoid creating sub-binaries but we still create a extra match contexts.

This code path's only ever hit when we're pulling objects > 4096KB (?SIZE_BLOCK) out of
the couch file, and even then I expect the gains are quite small. But who knows. I'll try
to profile it in a bit. I'm super curious and I'm enjoying the education.

> binary optimization in couch_file
> ---------------------------------
>
>                 Key: COUCHDB-1130
>                 URL: https://issues.apache.org/jira/browse/COUCHDB-1130
>             Project: CouchDB
>          Issue Type: Improvement
>          Components: Database Core
>    Affects Versions: 1.0.2
>            Reporter: Randall Leeds
>            Priority: Minor
>         Attachments: 0001-refactor-remove_block_prefixes-2-for-optimization.patch
>
>
> I've had this patch sitting around since January and kept forgetting to file the ticket.
Hurray spring cleaning.
> Just for fun I ran erlc with +bin_opt_info, which gives information about how the Erlang
VM can optimize creating binary objects.
> What follows is the commit message from my patch.
> Even if I'm wrong about the last point, it can't hurt.
> What think you all?
> -------------------
> Running erlc with +bin_opt_info gives an INFO message stating that "matching
> anything else but a plain variable to the left of a binary pattern will prevent
> delayed sub binary optimization; SUGGEST changing argument order"
> I guess matching 0 is triggering this. If I understand correctly, this change
> will allow the compiler to skip creating a sub-binary that starts at the block
> boundary in the third clause and delay creation until we strip the leading byte
> in the 0 clause. This means one less 1-byte binary every time we read across a
> block boundary.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message