Both tutorials are from a Simons institute Information Theory Boot camp.
Piggybacking RS codes
Measurements from Facebook's data-center showing the effects of node failures on the network and disk-IO. A practical solution (called “Piggybacking”) that ameliorates this problem is also presented.
A tutorial on how Hadoop HDFS stores files and how Reed-Solomon and modern distributed storage codes are designed for Facebook clusters.
Talk given by Alex Dimakis on a seminar organized by Milibo, May 2013.
An animated video describing a simple graph-based exact-MBR code construction, along with a gentle introduction to regenerating codes. Please click on the thumbnail above to view the video (and keep your speakers on! ).
A. G. Dimakis, K. Ramchandran, Y. Wu, C. Suh,
Proceedings of the IEEE, March 2011, Vol 99, No 3.