<div dir="ltr">I've never read that paper before and can not find a free copy online. Based on the abstract it seems to be a parity based algorithm (erasure codes) and would not be directly applicable to replication based dispersion/data placement.<div><br></div><div>There is current work to enable an erasure code scheme for data encoding [1] - but the current effort is focused on using the existing swift ring implementation for partition desperation and placement.</div><div><br></div><div>If you research overlaps with the general idea of taking a object and splitting into n chunks such that any k of the m chunks can construct the file object you may want to align with the current development on-going in this area [2].</div><div><br></div><div>Best regards,</div><div><br></div><div>1. <a href="https://github.com/openstack/swift-specs/blob/master/specs/swift/erasure_coding.rst">https://github.com/openstack/swift-specs/blob/master/specs/swift/erasure_coding.rst</a></div><div>2. <a href="https://review.openstack.org/#/q/status:open+project:openstack/swift+branch:feature/ec,n,z">https://review.openstack.org/#/q/status:open+project:openstack/swift+branch:feature/ec,n,z</a></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Mon, Nov 10, 2014 at 2:23 PM, Aniket K <span dir="ltr"><<a href="mailto:aniket.k@tcs.com" target="_blank">aniket.k@tcs.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div><font face="sans-serif">Dear Team,</font>
<br>
<br><font face="sans-serif">In OpenStack Swift, built-in replication
technique is used to provide 3x+ data redundancy that is enabling OpenStack
to provide high availability. We are studying use of <b>Information Dispersal
Algorithm(IDA)</b> as a technique instead of replication to provide high
availability along with security.</font>
<br>
<br><font face="sans-serif">We would like to know,</font>
<br><font face="sans-serif">1. Was
it considered in Swift before? If so, can you please share your opinion?</font>
<br><font face="sans-serif">2. Will
this fit into existing Swift architecture that supports eventual consistency?
</font>
<br><font face="sans-serif">3. Do
you see advantages like storage efficiency, security etc. of using this
technique over replication?</font>
<br>
<br><font face="sans-serif">Any pointer/suggestions/guidance in
this regard will be appreciated.</font>
<br>
<br><font face="sans-serif">More information about IDA can be found
at:</font>
<br><a href="http://en.wikipedia.org/wiki/Secret_sharing" target="_blank"><font color="blue" face="sans-serif"><u>http://en.wikipedia.org/wiki/Secret_sharing</u></font></a>
<br><font face="sans-serif"><br>
Thanks in advance.</font>
<br>
<br><font face="sans-serif">Regards,<br>
Aniket Kulkarni</font></div><p>=====-----=====-----=====<br>
Notice: The information contained in this e-mail<br>
message and/or attachments to it may contain <br>
confidential or privileged information. If you are <br>
not the intended recipient, any dissemination, use, <br>
review, distribution, printing or copying of the <br>
information contained in this e-mail message <br>
and/or attachments to it are strictly prohibited. If <br>
you have received this communication in error, <br>
please notify us by reply e-mail or telephone and <br>
immediately and permanently delete the message <br>
and any attachments. Thank you</p>
<p></p><br>_______________________________________________<br>
OpenStack-dev mailing list<br>
<a href="mailto:OpenStack-dev@lists.openstack.org">OpenStack-dev@lists.openstack.org</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
<br></blockquote></div><br></div>