Please also note that m7 IS NOT HBase and have no connection with Apache
HBase at all. Please do not let your vendor tell otherwise.

Enis

On Thu, Sep 22, 2016 at 12:00 AM, Deepak Khandelwal <
dkhandelwal....@gmail.com> wrote:

> Ok thx everyone.
>
> Will check with Mapr
>
> On Wednesday, September 21, 2016, Heng Chen <heng.chen.1...@gmail.com>
> wrote:
>
> > Not sure hbase m7 is which version of hbase in community.
> >
> > Is your batch load job some kind of bulk load or just call HTable API
> > to dump data to HBase?
> >
> >
> > 2016-09-22 14:30 GMT+08:00 Dima Spivak <dimaspi...@apache.org
> > <javascript:;>>:
> > > Hey Deepak,
> > >
> > > Assuming I understand your question, I think you'd be better served
> > > reaching out to MapR directly. Our community isn't involved in M7 so
> the
> > > average user (or dev) wouldn't know about the ins and outs of that
> > > offering.
> > >
> > > On Wednesday, September 21, 2016, Deepak Khandelwal <
> > > dkhandelwal....@gmail.com <javascript:;>> wrote:
> > >
> > >> Hi all
> > >>
> > >> I am facing an issue while accessing data from an hbase m7 table which
> > has
> > >> about 50 million records.
> > >>
> > >> In a single Api request, we make 3 calls to hbase m7.
> > >> 1. Single Multi get to fetch about 30 records
> > >> 2. Single multi-put to update about 500 records
> > >> 3. Single multi-get to fetch about 15 records
> > >>
> > >> We consistently get the response in less than 200 seconds for approx
> > >> 99%calls. We have a tps of about 200 with 8vm's.
> > >> But we get issue everyday between 4pm and 6pm when Api response time
> > gets
> > >> significant increase to from 200ms to 7-8sec. This happens because we
> > have
> > >> a daily batch load That runs between 4and 6pm that puts multiple
> entries
> > >> into same hbase table.
> > >>
> > >> We are trying to find a solution to this problem that why response
> time
> > >> increases when batch load runs. We cannot change the time of batch
> job.
> > Is
> > >> there anything we could do to resolve this issue?any help or pointers
> > would
> > >> be much appreciated. Thanks
> > >>
> > >
> > >
> > > --
> > > -Dima
> >
>

Reply via email to