Design and Evaluation of Data Center Network Topologies

PhD Thesis Proposal Defence


Title: "Design and Evaluation of Data Center Network Topologies"

by

Mr. Yang LIU


ABSTRACT:

Large-scale data centers form the core infrastructure support for the ever 
expanding cloud based services. Thus the performance and dependability 
characteristics of data centers will have significant impact on the scalability 
of these services. In particular, the data center network needs to be agile and 
reconfigurable in order to respond quickly to ever changing application demands 
and service requirements. Significant research work has been done on designing 
the data center network topologies in order to improve the performance of data 
centers.

In this proposal, we first present a abstract survey of data center network 
designs and topologies that have published recently. We start with a discussion 
on various representative data center network topologies, and compare them with 
respect to several properties in order to highlight their advantages and 
disadvantages. A good understanding of the state-of-the-art in data center 
networks would enable the design of future architectures in order to improve 
performance and dependability of data centers.

With the plethora of data center network (DCN) topologies that have been 
proposed in the literature, there is a sore need for a standardized method for 
evaluating and comparing various alternate DCN architectures. Considering these 
needs, we designed DCNSim, a general purpose DCN simulator that supports most 
well-known DCN topologies proposed in the literature. Our simulator can 
generate various metrics for the topologies, including static metrics like 
average path length, aggregated bottleneck throughput, routing failure rate; 
and dynamic metrics like packet loss rate, average buffer size and link 
utilization. The modular and flexible architecture of the simulator permits 
easy extension to support any future proposed topologies and compute new 
metrics.

With the support of DCNSim, we present an evaluation of the fault-tolerance 
characteristics of several important data center network topologies. These 
enable us to present an objective comparison of the network topologies under 
faulty conditions. The concept of fault region is useful to describe associated 
failures.

Optical data center networks (DCNs) are becoming increasingly popular for their 
technological advantages including flexibility and high bandwidth. However, the 
existing algorithm applied in literature cannot fully utilize such model and 
its flexibility. We focuses on addressing these two main questions which are of 
key importance to fully exploit the advantages of the optical DCNs. To this 
end, we present an algorithm for constructing the network. By integrating 
largest flow first and traffic aware shortest path routing together, our new 
topology construction algorithm significantly improves the throughput by $90\%$ 
over the existing one under the evaluated traffic patterns.

On base of above work, we propose future research plan that helps to build a 
more solid thesis to fulfill the requirement of the degree.


Date:                   Monday, 18 March 2013

Time:                   4:00pm - 6:00pm

Venue:                  Room 4480
                         lifts 25/26

Committee Members:      Dr. Jogesh Muppala (Supervisor)
                         Dr. Brahim Bensaou (Chairperson)
 			Dr. Gary Chan
 			Dr. Pan Hui


**** ALL are Welcome ****