Mailing-List: contact user-help@storm.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@storm.apache.org
Subject: Re: Need Help regarding topology with numWorker>1
To: user@storm.apache.org
References: 
 <CAHhryHZW1-064UfZ2Zp9NTSzM4hyEHvVb4jU=AB2YJwOr+Sxqg@mail.gmail.com>
 <CAGC92pqgyJnN2BcJKshtUfYNTUzY5u3_bYrVTJapsEkWit9DLg@mail.gmail.com>
 <CAHhryHZ=E+=aZpVfaxMtNeMQ3V=EsrVPi3i6qR9oGoO556T5Yw@mail.gmail.com>
 <CAGC92pqdApoqFx=nTSPMjTh0VdAU+FERdfERmj31Z23A5YqrJQ@mail.gmail.com>
 <56BCE552.3080601@apache.org>
 <CAHhryHZDRWHC4QR5=MsocmEnrjv+iumz-X_3fVR--9kuqXgrLg@mail.gmail.com>
 <CAHhryHbi4C9Xu90gCm4GO4ormPCoxsWBLRvQUR2L_-g64eej2A@mail.gmail.com>
 <CAMHaYAfTFTg9pOw45g8a9f-6FOqqXuGGPO4ivtyetHwPMa7CaA@mail.gmail.com>
 <CAHhryHYTFPMj3o047_qFj95ycg9jczBWH11ni1UEiCXS5CEkxQ@mail.gmail.com>
From: "Matthias J. Sax" <mjsax@apache.org>
Message-ID: <56BF5DA9.9090902@apache.org>
Date: Sat, 13 Feb 2016 17:45:29 +0100
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101
 Icedove/38.5.0
MIME-Version: 1.0
In-Reply-To: 
 <CAHhryHYTFPMj3o047_qFj95ycg9jczBWH11ni1UEiCXS5CEkxQ@mail.gmail.com>
Content-Type: multipart/signed; micalg=pgp-sha256;
 protocol="application/pgp-signature";
 boundary="1Sfg9ErA9P7HnO8JBQumvSKLVBmTJIvop"

This is an OpenPGP/MIME signed message (RFC 4880 and 3156)
--1Sfg9ErA9P7HnO8JBQumvSKLVBmTJIvop
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

If your workload does not saturate a single machine, it will of course
be more efficient to run within a single worker as you avoid intra-JVM
communication.

As long as CPU, memory, or network of a single machine is not utilized
completely, you will not benefit from multiple workers from a
performance point of view.

Btw: I am a not sure what you mean by "I have to create a scenario..."?

On 02/12/2016 11:42 PM, Rudraneel chakraborty wrote:
> Thanks. Actually i have to create a scenario where 2 worker performs
> better than one worker. But in reality , topology with a single worker
> performs considerably better.=20
>=20
> I sending csv lines to kafka (5 partitions ) and reading them from a
> topology with kafka spout (parallelism hint 5)
>=20
> Any thoughts?=20
>=20
> On Thursday, 11 February 2016, Nathan Leung <ncleung@gmail.com
> <mailto:ncleung@gmail.com>> wrote:
>=20
>     Any situation where you require more CPU than 1 server can provide
>     for you - there are tuning parameters (e.g. localOrShuffleGrouping)=

>     that you can use to reduce the amount of data sent over the network=
 too.
>=20
>     Any situation where you need to have tolerance in case of machine
>     failure.
>=20
>=20
>=20
>     On Thu, Feb 11, 2016 at 2:58 PM, Rudraneel chakraborty
>     <rudraneel.chakraborty@gmail.com
>     <javascript:_e(%7B%7D,'cvml','rudraneel.chakraborty@gmail.com');>>
>     wrote:
>=20
>         Topology param numWorkers i meant=20
>=20
>=20
>         On Thursday, 11 February 2016, Rudraneel chakraborty
>         <rudraneel.chakraborty@gmail.com
>         <javascript:_e(%7B%7D,'cvml','rudraneel.chakraborty@gmail.com')=
;>>
>         wrote:
>=20
>             More specifically , i have seen a topology performs better
>             if it is assigned a single worker compared to more than one=

>             worker.=20
>=20
>             I want a situation where a topology performs better with
>             more than one worker.
>=20
>             And it doesnt matter if both workers are on same supervisor=

>             or different supervisor
>=20
>             On Thursday, 11 February 2016, Matthias J. Sax
>             <mjsax@apache.org> wrote:
>=20
>                 I am not sure what you mean:
>=20
>                   - number of worker slots per supervisor
>                 or
>                   - topology parameter "number of workers"
>=20
>                 Can you clarify?
>=20
>                 -Matthias
>=20
>                 On 02/11/2016 05:14 AM, anshu shukla wrote:
>                 > Not  like that..  But i have used workers equal to
>                 number of cores. Each
>                 > vm with 8 corea.
>                 >
>                 > On 11 Feb 2016 9:07 am, "Rudraneel chakraborty"
>                 > <rudraneel.chakraborty@gmail.com
>                 > <mailto:rudraneel.chakraborty@gmail.com>> wrote:
>                 >
>                 >     more than one worker on same node ? Did u use
>                 custom schedule r?
>                 >     because by default, the workers would be spread
>                 throughout the cluster
>                 >
>=20
>=20
>=20
>             --=20
>             Rudraneel Chakraborty
>             Carleton University Real Time and Distributed Systems Reser=
ach
>=20
>=20
>=20
>         --=20
>         Rudraneel Chakraborty
>         Carleton University Real Time and Distributed Systems Reserach
>=20
>=20
>=20
>=20
> --=20
> Rudraneel Chakraborty
> Carleton University Real Time and Distributed Systems Reserach
>=20


--1Sfg9ErA9P7HnO8JBQumvSKLVBmTJIvop
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQIcBAEBCAAGBQJWv12tAAoJEFCVK48prEZ4G2MP/3Kt8qYhaKL19ybbzpiguE6Z
Ao3B7HSM54y/Kskyzi/7E/GxiK+IjXLl5VZNpZ9yIQ/5SW04OPmUYVDwLFe2sYpt
7U0RTE+aVAF454Fh4yBTrFSfI3D0gBHHAQGxJoF2F9aQ5aWHaLmQr5sNzKjFSFMb
/TdgO/kPtx5EfJSAhR1dkjesJMOC6i3GZRRYymK6AJ4906x9Ky54pgycWAZbFhiJ
3INh3+CLriAqOklbA5+wJmXW0T++J1EnRgZInNor3fwq/wuG/o2XCcNzyQW8Oab+
+8Hhx0FBX+UU1isH1bxjHi+EdEWL31DeXb4h6Dy5CXi+orf6hipp0Q81XsWHidre
jJ2Byfik/q7B3eAIvUPRQEZ2BbwTwPkAF5n0K9XUbC+WpY694ko8Zc0MLqZxxv/h
LN8zqFeTHkhxg3G1CLFquf4zCisyMT4F2qBewwkO5JLV/ZlQO8H3VjXyKZYEBNxe
ZTczXdiriNw8fu1uHnJBDcDPLS8rUhwUgQ9Fi8q+rlrF4ULuMKJ7mBTFPRQd1n9/
jfkNGcPyEVU7IeH6mXYgMUFHUXQ444j4Mv2/aj3UCPu1I8dIVd2S06/qeNj9/etO
h5l1TktQE+0InQGH4OPTyo0Ix4Mps/4+um9us2MZ6o/1hInT6uc2bMILuyc21wEf
+ZWIhFQwdxVUXLPs9GJg
=a+L/
-----END PGP SIGNATURE-----

--1Sfg9ErA9P7HnO8JBQumvSKLVBmTJIvop--