hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jean-Marc Spaggiari <jean-m...@spaggiari.org>
Subject Re: hbase timerange scan
Date Thu, 05 Nov 2015 16:25:58 GMT
I will be very happy to be wrong ;) That means it more efficient than what
I think ;)

2015-11-05 11:21 GMT-05:00 Dave Latham <latham@davelink.net>:

> Don't think that's correct.  If you look at
> StoreFileScanner.shouldUseScanner you can see that it will skip entire
> store files if the time range for a scan does not intersect with the time
> range of data in the store file.  However, without tiered compaction there
> is nothing built in to optimize grouping of data into store files by time
> range for efficiency in time range scans.
>
> On Thu, Nov 5, 2015 at 7:57 AM, Jean-Marc Spaggiari <
> jean-marc@spaggiari.org
> > wrote:
>
> > Hi Shushant,
> >
> > If my memory deserves me correctly, someone asked the same question
> about a
> > year ago, and one of the commiters looked into the code to figure that
> > there is no mechanism to skip some of the files even when timestamp is
> > provided. It ended up being a full table scan.
> >
> > You might be able to find that using google and the mailing list.
> >
> > JMS
> >
> > 2015-11-04 23:07 GMT-05:00 Shushant Arora <shushantarora09@gmail.com>:
> >
> > > Does hbase timerange scan is full table scan without the start and stop
> > > key?
> > > Or is it take care of HFile meta data about min and max timerange n
> > HFile .
> > > And how it optimises this metadata after compaction of multiple files?
> > >
> >
>

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