Interconnection Networks and Packet Buffer Architectures for Massive Data Centers

PhD Thesis Proposal Defence


Title: "Interconnection Networks and Packet Buffer Architectures for Massive
Data Centers"

by

Mr. Dong LIN


ABSTRACT:

Massive data centers are being built around the world to provide various 
cloud computing services. As a result, data center networking has recently 
been a hot research topic in both academia and industry. A fundamental 
challenge in this research is the design of the data center network that 
interconnects the massive number of servers, and provides efficient and 
fault-tolerant routing service to upper-layer applications. In response to 
this challenge, the research community have begun exploring novel 
interconnect topologies. One approach is to use commodity switches or 
servers to scale out the network, such as Portland, VL2, DCell, BCube and 
FiConn. The other approach is to exploit optical devices to build 
high-capacity switches, such as Helios, HyPaC, PETASW, Data Vortex and 
OSMOSIS. Understandably, this research is still in its infancy. For the 
first approach, the proposed solutions either scale too fast (i.e., double 
exponentially) or too slow, suffer from performance bottlenecks, or can be 
quite costly. For the second approach, where the entire interconnection 
network can be regarded as a “giant” switch, its performance heavily 
relies on well-designed packet buffers that support multiple queues, 
provide large capacity and short response time.

In this proposal, five different kinds of algorithms/architectures are 
presented, addressing on both of these issues respectively. Using 
small-port-count commodity switches, I propose two cost-effective and 
gracefully scalable Data Center Interconnects (DCIs) called Hyper-BCube 
and LacoNet that combine the advantages of both DCell and BCube 
architectures while avoiding their limitations. On the other hand, aiming 
at scalable packet buffers, I propose three memory management algorithms 
along with their architectures termed Radom-Round-Robin, distributed and 
hierarchical schemes. Preliminary experimental results show that these 
schemes can outperform the traditional algorithms significantly in terms 
of low time complexity, short access delay and guaranteed performance.

My future work will be carried out along following directions. First, I 
will conduct more experiments and simulations to comprehensively evaluate 
the performance of the proposed algorithms/architectures in different 
settings and environments. Second, the fault-tolerant and load-balancing 
routing algorithms for the proposed DCIs are still required to be further 
investigated.


Date:                   Tuesday, 21 June 2011

Time:                   10:00am - 12:00noon

Venue:                  Room 4475
                         lifts 25/26

Committee Members:      Prof. Mounir Hamdi (Supervisor)
                         Dr. Brahim Bensaou (Chairperson)
 			Dr. Lin Gu
 			Dr. Jogesh Muppala


**** ALL are Welcome ****