You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hbase.apache.org by Ted Yu <yu...@gmail.com> on 2011/08/17 21:40:46 UTC
prefix compression and HBASE-4027
Jacek / Pi:
I hope you two can discuss the implications of Jacek's proposal in
HBASE-4218 w.r.t. Pi's current work in HBASE-4027.
Agreeing on the access pattern is an important aspect.
Cheers
Re: prefix compression and HBASE-4027
Posted by Jacek Migdal <ja...@fb.com>.
It will still have ByteBuffer, but the data might be no longer uncompressed there.
>From the description of your JIRA I don't see any problem there.
Thanks for letting me know.
Jacek
From: Li Pi <li...@cloudera.com>>
Date: Wed, 17 Aug 2011 12:50:38 -0700
To: Ted Yu <yu...@gmail.com>>
Cc: Microsoft Office User <ja...@fb.com>>, <de...@hbase.apache.org>>
Subject: Re: prefix compression and HBASE-4027
As long as HFileBlocks have a ByteBuffer where the data is stored, it should be fine.
On Wed, Aug 17, 2011 at 12:40 PM, Ted Yu <yu...@gmail.com>> wrote:
Jacek / Pi:
I hope you two can discuss the implications of Jacek's proposal in HBASE-4218 w.r.t. Pi's current work in HBASE-4027.
Agreeing on the access pattern is an important aspect.
Cheers
Re: prefix compression and HBASE-4027
Posted by Li Pi <li...@cloudera.com>.
As long as HFileBlocks have a ByteBuffer where the data is stored, it should
be fine.
On Wed, Aug 17, 2011 at 12:40 PM, Ted Yu <yu...@gmail.com> wrote:
> Jacek / Pi:
> I hope you two can discuss the implications of Jacek's proposal in
> HBASE-4218 w.r.t. Pi's current work in HBASE-4027.
> Agreeing on the access pattern is an important aspect.
>
> Cheers
>
Re: prefix compression and HBASE-4027
Posted by Ted Yu <yu...@gmail.com>.
Including Jason in the discussion as well.
On Wed, Aug 17, 2011 at 12:40 PM, Ted Yu <yu...@gmail.com> wrote:
> Jacek / Pi:
> I hope you two can discuss the implications of Jacek's proposal in
> HBASE-4218 w.r.t. Pi's current work in HBASE-4027.
> Agreeing on the access pattern is an important aspect.
>
> Cheers
>