A Survey on Congestion Control in Data Center Networks

PhD Qualifying Examination


Title: "A Survey on Congestion Control in Data Center Networks"

by

Mr. Ahmed MOHAMED ABDELMONIEM SAYED


Abstract:

Cloud computing defines a new computation approach that has led to the 
restructuring of the whole IT industry, by making computing available to 
customers at reasonable costs in a manner similar to governments or private 
companies providing utilities (water, electricity, gas) to the citizens.

Cloud computing is becoming widely popular with the public and private IT 
sectors, leading to a dramatic increase in the rate of deployment of new data 
centers. It is widely believed that today any piece of data or information that 
circulates in the Internet originates in fact from some data center. Data 
centers consist of tens of thousands of servers mutually interconnected via 
high speed network interconnects, running a large number of applications 
serving a huge number of users simultaneously. As such most often these 
applications adopt a multi-tiered design model where several services residing 
on distributed servers work together to satisfy a single client request. Hence, 
the overall performance of such applications depends greatly on the ability of 
the underlying communication network to provide efficient and timely data 
transfers.

In this survey, we first review the architectural design of data center 
networks and inspect how such networks would affect application performance. In 
particular we will focus on network congestion, as it is one of the major 
problems in communication networks, and review the three major categories of 
congestion control mechanisms proposed for the Internet. We will also discuss 
the major limitations of TCP congestion control in tackling congestion in data 
centers. In the third chapter, we will study the different solutions that has 
been proposed over the past few years to address TCP's shortcomings in data 
centers and classify then into three categories: sender-based approaches, 
receiver-based approaches and finally switch-based approaches. We will further 
investigate their ability to provide delay guarantees and classify them into 
deadline-agnostic solutions and deadline-aware solutions. Finally, we compare 
the proposed methods qualitatively, draw our conclusions on the reviewed 
proposals' and discuss some future ideas for further improvement.


Date:			Thursday, 26 February 2015

Time:                  	1:30am - 3:30pm

Venue:                  Room 5503
                         Lifts 25/26

Committee Members:	Dr. Brahim Bensaou (Supervisor)
 			Dr. Jogesh Muppala (Chairperson)
 			Dr. Kai Chen
 			Dr. Lin Gu


**** ALL are Welcome ****