Mailing-List: contact user-help@hbase.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@hbase.apache.org
Received-SPF: pass (nike.apache.org: domain of prakash.kadel@gmail.com
 designates 209.85.210.47 as permitted sender)
Subject: Re: coprocessor enabled put very slow, help please~~~
References: <78023CDA-6661-4BE1-B5FF-3762119E51A2@gmail.com>
 <7434AC93-B867-4DAC-8A7B-0A5A4CC386CA@gmail.com>
 <1361149643.73801.YahooMailNeo@web140601.mail.bf1.yahoo.com>
 <B398DD4D-8FB7-4A61-B491-CCCEE1F25BC4@gmail.com>
 <1361150275.19655.YahooMailNeo@web140602.mail.bf1.yahoo.com>
From: Prakash Kadel <prakash.kadel@gmail.com>
Content-Type: text/plain;
	charset=us-ascii
In-Reply-To: <1361150275.19655.YahooMailNeo@web140602.mail.bf1.yahoo.com>
Message-Id: <9C786227-90FE-46BB-B2A3-01FD182181E2@gmail.com>
Date: Mon, 18 Feb 2013 10:32:16 +0900
To: "user@hbase.apache.org" <user@hbase.apache.org>
Content-Transfer-Encoding: quoted-printable
Mime-Version: 1.0 (1.0)

one more question.

even if the coprocessors are making insertions to different region, since i u=
se  "postCheckAndPut" shouldnt there be not much prefomance slow down?

thanks

Sincerely,
Prakash Kadel

On Feb 18, 2013, at 10:17 AM, lars hofhansl <larsh@apache.org> wrote:

> Index maintenance will always be slower. An interesting comparison would b=
e to also update your indexes from the M/R and see whether that performs bet=
ter.
>=20
>=20
>=20
> ________________________________
> From: Prakash Kadel <prakash.kadel@gmail.com>
> To: "user@hbase.apache.org" <user@hbase.apache.org>=20
> Sent: Sunday, February 17, 2013 5:13 PM
> Subject: Re: coprocessor enabled put very slow, help please~~~
>=20
> thank you lars,
> That is my guess too. I am confused, isnt that something that cannot be co=
ntrolled. Is this approach of creating some kind of index wrong?
>=20
> Sincerely,
> Prakash Kadel
>=20
> On Feb 18, 2013, at 10:07 AM, lars hofhansl <larsh@apache.org> wrote:
>=20
>> Presumably the coprocessor issues Puts to another region server in most c=
ases, that could explain it being (much) slower.
>>=20
>>=20
>>=20
>> ________________________________
>> From: Prakash Kadel <prakash.kadel@gmail.com>
>> To: "user@hbase.apache.org" <user@hbase.apache.org>=20
>> Sent: Sunday, February 17, 2013 4:52 PM
>> Subject: Re: coprocessor enabled put very slow, help please~~~
>>=20
>> Forgot to mention. I am using 0.92.
>>=20
>> Sincerely,
>> Prakash
>>=20
>> On Feb 18, 2013, at 9:48 AM, Prakash Kadel <prakash.kadel@gmail.com> wrot=
e:
>>=20
>>> hi,
>>>     i am trying to insert few million documents to hbase with mapreduce.=
 To enable quick search of docs i want to have some indexes, so i tried to u=
se the coprocessors, but they are slowing down my inserts. Arent the coproce=
ssors not supposed to increase the latency?=20
>>> my settings:
>>>      3 region servers
>>>     60 maps
>>> each map inserts to doc table.(checkAndPut)
>>> regionobserver coprocessor does a postCheckAndPut and inserts some rows t=
o a index table.
>>>=20
>>>=20
>>> Sincerely,
>>> Prakash