hadoop-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Tim Wintle <tim.win...@teamrubber.com>
Subject Re: google snappy
Date Wed, 23 Mar 2011 17:11:39 GMT
On Wed, 2011-03-23 at 10:03 -0700, Jean-Daniel Cryans wrote:
> Somebody obviously needs to publish some benchmarks, but knowing
> Snappy's origin I can believe that claim. 

There were some benchmarks in the original Bigtable presentation

Results from compressing bigtable blocks:

Algorithm  % remaining	Encoding	Decoding
Gzip       13.4%        21MB/s          118MB/s
LZO        20.5%        135MB/s         410MB/s
Zippy      22.2%        172MB/s         409MB/s

(Zippy is apparently what's now been renamed snappy)

Tim Wintle


Mime
View raw message