Topics: Software De fined Networking
Authors: Sai Qian Zhang (University of Toronto, Canada); Qi Zhang (University of Toronto & University of Waterloo, Canada); Ali Tizghadam (TELUS & University of Toronto, Canada); Byungchul Park, Hadi Bannazadeh and Alberto Leon-Garcia (University of Toronto, Canada); Raouf Boutaba (University of Waterloo, Canada)
Presenter Bio: ALI TIZGHADAM is currently a Senior Research Associate and Adjunct
Lecturer in the Department of Electrical & Computer Engineering at
the University of Toronto. He received his M.A.Sc in 1994 for University
of Tehran in Electrical Engineering. Then he went to the industry for
about 10 years where he gained an abundance of experience in
telecommunications especially in local exchange switches and access
networks. He started his PhD in 2003 and finished 2008. His major
interest are in Autonomic Network Control and Management, Traffic
Engineering, Network Planning, Intelligent Transportation Systems (ITS),
Connected Vehicles and Smart Transportation, and Green Communications .
Abstract: In Software Defined Networking (SDN), a central controller can achieve
fine-grained control over individual flows by installing appropriate
forwarding rules in switches and routers. This allows the network to
realize a wide variety of functionalities and objectives. But at the
same time, these flexibility and versatility come at the expense of (1) a
huge burden on the limited Ternary Content Addressable Memory (TCAM)
space, and (2) limited scalability due to the large number of forwarding
rules that are required in the network. To address these limitations,
we present Sector, a switch memory-aware routing scheme that reduces the
number of entries in the switches, and at the same time guarantees the
load balancing on link resources utilization. We consider both the
static and dynamic versions and propose solution algorithms. Experiments
show our algorithms can reduce the TCAM space usage and network control
traffic by 20% to 80% on different network topologies.
