Mailing-List: contact solr-user-help@lucene.apache.org; run by ezmlm
Precedence: bulk
Reply-To: solr-user@lucene.apache.org
Received-SPF: pass (athena.apache.org: message received from 54.164.171.186
 which is an MX secondary for solr-user@lucene.apache.org)
Subject: Re: Replication for SolrCloud
References: 
 <CAAfTDLVnvtJqccohwZLzxjhoH5ZjcJwywbjgYWMNwxCSDLZXhA@mail.gmail.com>
 <CAN4YXvccGVGLWd7N4H10kM+ghkVEb0=ROc0-egJbBhQk9YmxUw@mail.gmail.com>
 <CAGESXAROy-3+LaW8tehi87D0ZX29Bs+rDzHOXSmUx0bG3u26rA@mail.gmail.com>
 <55320581.60307@devoteam.com>
 <CAGESXAR5dr=Pzf6Q1svbunkUzSw8W7dRJVgFu-F0TP_8OaYsog@mail.gmail.com>
 <CAOOKt51fpLgD1UBRMXX9uUV7PH9TLPOg9ejdeA6F8Bnh_GkgGA@mail.gmail.com>
 <CAN4YXvfNtJy9rOVSuCHxzK-cF75eJGV_WJ6bVUHx+aEoL-NPqQ@mail.gmail.com>
 <CAGESXAQ4OF=8PxtLskN8hdht=JyuUk2my3j89R9paJjF79OjbA@mail.gmail.com>
From: "juergen.wagner@devoteam.com" <juergen.wagner@devoteam.com>
Content-Type: text/plain;
	charset=utf-8
In-Reply-To: 
 <CAGESXAQ4OF=8PxtLskN8hdht=JyuUk2my3j89R9paJjF79OjbA@mail.gmail.com>
Message-Id: <708D87FF-89CC-42C9-8F20-A1FD7A1312AC@devoteam.com>
Date: Sun, 19 Apr 2015 13:46:10 +0200
To: "solr-user@lucene.apache.org" <solr-user@lucene.apache.org>
Content-Transfer-Encoding: quoted-printable
Mime-Version: 1.0 (1.0)

In simple words:

HDFS is good for file-oriented replication. Solr is good for index replicati=
on.

Consequently, if atomic file update operations of an application (like Solr)=
 are not atomic on a file level, HDFS is not adequate - like for Solr with l=
ive index updates. Running Solr on HDFS (as a file system) will pose limitat=
ions due to HDFS properties. Indexing, however, still won't use Hadoop.

If you produce indexes and distribute them as finalized, read-only structure=
s (e.g., through Hadoop jobs), HDFS is fine. Solr does not need to be much a=
ware of HDFS.

The third one in the picture is records-based replication to be handled by H=
base, Cassandra or Zookeeper, depending on requirements.

Cheers,
J=C3=BCrgen=