Calvin Lab Auditorium
Codes for distributed storage are designed to both reliably maintain the stored data and efficiently maintain the guaranteed reliability of the storage network under node failures. These properties have already made such codes important for data storage practice. But code design also determines the ways in which data can be retrieved from coded storage, and thus how available it is under given data request patterns and the request scheduling policies. This talk will look into data retrieval quality of service in coded distributed storage and its implications on code design.