hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ted Yu <yuzhih...@gmail.com>
Subject Re: Filters with TimeRange (should get executed only in regions matching TimeRange)
Date Wed, 12 Jul 2017 15:22:23 GMT
You may know this method:

Scan#setTimeRange(long minStamp, long maxStamp)

Please take a look at its javadoc

You can see how the time range is used, e.g., by
StoreFileScanner#shouldUseScanner()
to narrow the search.

FYI

On Wed, Jul 12, 2017 at 8:04 AM, Veerraju Tadimeti <tvvraju@gmail.com>
wrote:

> Rowkey is (hiphen seprated hashed primary keys 1-0000-xxxx-aaa-bbbb-cccc)
> .  I believe Hbase takes care of Rowkey (startRow and stopRow) during
> scan.  I want to get rid of regions with unwanted timerange. In filter, I
> am basically returning eitire cells (complete ROW) instead of updated cells
> in that region.
>
> My Story is to get the all rows  ( with all cells : old cells and updated
> cells in that timerange ) falling in that timerange and key range and also
> with scan Raw true.
>
> Thanks,
> Raju,
> (972)273-0155.
>
> On Tue, Jul 11, 2017 at 10:13 PM, Ted Yu <yuzhihong@gmail.com> wrote:
>
> > Can you tell us more about your row key design ?
> >
> > Thanks
> >
> > On Tue, Jul 11, 2017 at 3:03 PM, Veerraju Tadimeti <tvvraju@gmail.com>
> > wrote:
> >
> > > hi,
> > >
> > > If I implement a filter, it does full range scan.  Is there a way to
> > > implement a Filter ( with TimeRange, startRow  and stopRow ) without
> > doing
> > > Full range scan.
> > >
> > > Basically, if we pass TimeRange to scan, it wont do Full range scan.  I
> > > want write a filter to get executed only in the regions which fall
> under
> > > TimeRange.
> > >
> > > Thank you in advance.
> > >
> > >
> > > Thanks,
> > > Raju,
> > > (972)273-0155.
> > >
> >
>

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