[DOC] Features overview suggests Cinders to increase reliability

[Andrew Lazarev]

"Using Cinder volumes increases reliability of data which is important for HDFS service."

This is not true. We strongly not recommend to use any Cinder devices with replication enabled. HDFS makes replication by itself and replication on device level will multiply number of chunks one more time. So, if HDFS has replication factor equal to 3 and Cinder configured to use Swift with replication factor 3, resulting replication factor will be 9. That's why we recommend using Cinder for devices that don't replicate data. This is not something called 'reliability'.

Need to describe this in docs.

[Duncan Thomas (duncan-thomas)]

Cinder does *not* use swift, except for backup.

I'm not sure that this bug makes any sense. The use of replication in cinder is designed for disaster recovery type scenarios, and currently has little effect on per-volume reliability.

[John Griffith (john-griffith)]

Hey Andrew, maybe a link to the doc would help here? I suspect there's an overloaded term here somewhere and we just need to be more explicit in said doc.

[Andrew Lazarev (alazarev)]

Link: http://docs.openstack.org/developer/sahara/userdoc/features.html#cinder-support

[Daniele Venzano (venza)]

Using Cinder for HDFS has performance issues that are not well understood.
Once you lose data locality, computation will no longer happen near the data and jobs will make heavy use of the network. Many of the assumptions that Hadoop/Spark make are no longer valid when HDFS is backed by network volumes.

While Sahara supporting Cinder is a good thing, I would be wary of even suggesting such a solution without a well-thought document of the pros and cons that people can read and understand.

[Sergey Reshetnyak (sreshetniak)]

This bug is > 60 days without activity. We are unsetting assignee and milestone and setting status to Incomplete in order to allow its expiry in 60 days.

If the bug is still valid, then update the bug status.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/292859

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/292859
Committed: https://git.openstack.org/cgit/openstack/sahara/commit/?id=1af4701b854d14134e5b9fdba6ba73f5326c361b
Submitter: Jenkins
Branch: master

commit 1af4701b854d14134e5b9fdba6ba73f5326c361b
Author: Nikita Konovalov <email address hidden>
Date: Tue Mar 15 15:30:48 2016 +0300

    Updated volumes section in docs

    Improved description for cinder volumes usage mechanism.
    Described the volumes backed by distributed storage case.

    Change-Id: Ie5a436daaca9e5ed009aa131cf6487be3057182a
    Closes-bug: #1371360

[Doug Hellmann (doug-hellmann)]

This issue was fixed in the openstack/sahara 5.0.0.0b1 development milestone.