Mailing-List: contact hdfs-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hdfs-issues@hadoop.apache.org
Date: Mon, 25 Jun 2012 23:49:43 +0000 (UTC)
From: "Sumadhur Reddy Bolli (JIRA)" <jira@apache.org>
To: hdfs-issues@hadoop.apache.org
Message-ID: <537078260.54178.1340668183942.JavaMail.jiratomcat@issues-vm>
Subject: [jira] [Created] (HDFS-3564) Make the replication policy pluggable
 to allow custom replication policies
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

Sumadhur Reddy Bolli created HDFS-3564:
------------------------------------------

             Summary: Make the replication policy pluggable to allow custom=
 replication policies
                 Key: HDFS-3564
                 URL: https://issues.apache.org/jira/browse/HDFS-3564
             Project: Hadoop HDFS
          Issue Type: Improvement
          Components: name-node
            Reporter: Sumadhur Reddy Bolli


ReplicationTargetChooser currently determines the placement of replicas in =
hadoop. Making the replication policy pluggable would help in having custom=
 replication policies that suit the environment.=20

Eg1: Enabling placing replicas across different datacenters(not just racks)
Eg2: Enabling placing replicas across multiple(more than 2) racks
Eg3: Cloud environments like azure have logical concepts like fault and upg=
rade domains. Each fault domain spans multiple upgrade domains and each upg=
rade domain spans multiple fault domains. Machines are spread typically eve=
nly across both fault and upgrade domains. Fault domain failures are typica=
lly catastrophic/unplanned failures and data loss possibility is high. An u=
pgrade domain can be taken down by azure for maintenance periodically. Each=
 time an upgrade domain is taken down a small percentage of machines in the=
 upgrade domain(typically 1-2%) are replaced due to disk failures, thus los=
ing data. Assuming the default replication factor 3, any 3 data nodes going=
 down at the same time would mean potential data loss. So, it is important =
to have a policy that spreads replicas across both fault and upgrade domain=
s to ensure practically no data loss. The problem here is two dimensional a=
nd the default policy in hadoop is one-dimensional. Custom policies to addr=
ess issues like these can be written if we make the policy pluggable.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrato=
rs: https://issues.apache.org/jira/secure/ContactAdministrators!default.jsp=
a
For more information on JIRA, see: http://www.atlassian.com/software/jira