hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ted Yu <yuzhih...@gmail.com>
Subject Re: Hbase data version performance impact
Date Wed, 01 Feb 2017 00:14:55 GMT
bq. load all value at client side

You can do the above or, if there're a lot of versions for each cell, you
can implement Coprocessor which retrieves the versions on server side and
returns desired one(s) back to client.

On Tue, Jan 31, 2017 at 4:09 PM, Manjeet Singh <manjeet.chandhok@gmail.com>
wrote:

> Hi Ted
>
> Its mean I have no other way to load specific version value the only way to
> load all value at client side and get version of my choice.
>
> Manjeet
> On 1 Feb 2017 00:25, "Ted Yu" <yuzhihong@gmail.com> wrote:
>
> > For #3, you need to retrieve multiple versions (to get to V2).
> >
> > Take a look at TestVisibilityLabelsWithDeletes#
> > testDeleteColumnWithLatestTimeStampUsingMultipleVersions
> > around line 1368.
> >
> > FYI
> >
> > On Tue, Jan 31, 2017 at 9:58 AM, Manjeet Singh <
> manjeet.chandhok@gmail.com
> > >
> > wrote:
> >
> > > Hi All
> > >
> > > can anyone tell me what is the Impact of using Hbase data version
> > >
> > > Q1 what is the impact if I have 1000 version set or what if its 10000
> > >
> > > Q2 Is it good idea to put non default version in different column
> family?
> > >
> > >
> > > Q3 Does version support to get specific version value?
> > >  as showing below I want to use only A2
> > > rk header header2
> > > 001 A1 V1
> > > 001 A2 V2
> > > 001 A3 V3
> > >
> > > Q4 Any idea to use concept of Nested table in Hbase? I am considering
> > > version
> > >
> > >
> > > Thanks
> > > Manjeet
> > >
> > >
> > >
> > > --
> > > luv all
> > >
> >
>

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