Bulk Data Transfer among Cloud Data Centers: Online Algorithms and SDN Implementation

This project studies efficient online optimization algorithms for large-scale data transfer among data centers in a geographically-distributed cloud system, as well as their SDN (Software Defined Networking)-facilitated implementation. Big data analytics, content distribution, and various web applications (social networking, search engine) have become dominating applications on today’s cloud platforms. Moving bulk volumes of data from one data center to a remote one is common in these applications, for aggregation of distributed datasets for processing by a MapReduce-like framework, replication of video contents to be closer to viewers, migration of virtual machine images for failure resilience, etc. The data transfers are commonly associated with different deadlines, depending on their delay sensitivity. In this project, we seek to design efficient online algorithms for
dynamic, globally-optimal bulk data transfer in a geo-distributed cloud system, and engineer the solution practically in an emerging SDN network.

Faculty Supervisor:

Zongpeng Li

Student:

Ruiting Zhou

Partner:

Hidaca Inc.

Discipline:

Computer science

Sector:

Information and communications technologies

University:

University of Calgary