2014 ISIT Plenary Lecture
Codes for Distributed Storage -- Asking More of an Old Friend
P. Vijay Kumar
Indian Institute of Science
Given the explosive growth in the amount of data that keeps getting generated and the inescapable mortality of a storage device, there is renewed interest within the storage industry in identifying techniques that will enable data to be stored in a manner that is efficient, reliable and which provides easy access to the data. Codes for distributed storage place data across nodes in a network in a redundant fashion and are designed precisely with these objectives in mind. This talk will focus on two recent approaches to code design that attempt to overcome the shortcomings of prevalent coding schemes that are based either on replication or the use of Reed-Solomon codes.
The first approach makes use of a class of codes known as regenerating codes, where the emphasis is on low storage overhead and minimization of the amount of data download needed for the repair of a failed node. Codes employed under the second approach are termed as locally repairable codes and here, the focus is on sharply reducing the number of nodes accessed during node repair. Both approaches call for additional structure within a block code and introduce new performance measures. We will overview these recent exciting developments in coding theory highlighting some of the innovative ideas that have been put forward, discuss applications, and make connections with other research topics of interest to the information-theory community.