couchdb-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Paul J. Davis" <>
Subject Re: CouchDB and Data Compression with Snappy
Date Mon, 19 Sep 2011 19:12:44 GMT

There really isn't much in the way of documentation to Snappy because its nearly automatic.
For the most part you just have to make sure that your version of Erlang is new enough to
use it and that it's configured to use Snappy (which it is by defualt IIRC).

As to compression of network traffic, attachments are the only thing that is currently compressed
by CouchDB. Raw JSON data isn't gziped or anything before sending it across the network. Its
been considered but no one to my knowledge has implemented anything for it.

Also, trunk means its in the repository only, there's no official tarball for it. The info
for the SVN repo is at [1] and you can get (completely unofficial) dynamically built tarballs
from GitHub's mirror at [2].


On Monday, September 19, 2011 at 1:58 PM, Christopher Moloney wrote:

> Hello,
> I am considering CouchDB for a project at work. One of the requirements is data compression.
> I have found this ticket on Snappy Compression for CouchDB:
> From the ticket it appears that Snappy was added to the trunk in May. And if I understand
the ticket, it will be released with version 1.2 of CouchDB. Is that correct?
> Can someone direct me to more documentation on this? I would like to see if the data
is compressed when written to disk, and whether it is compressed when replicating over the
> Where can I download the trunk version? I did not see a link on the download page:
> Thanks,
> Chris

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message