Type

Text

Type

Dissertation

Advisor

Balasubramanian, Aruna | Das, Samir | Gill, Phillipa | Sekar, Vyas | Gopalakrishnan, Vijay.

Date

2015-12-01

Keywords

Computer science

Department

Department of Computer Science.

Language

en_US

Source

This work is sponsored by the Stony Brook University Graduate School in compliance with the requirements for completion of degree.

Identifier

http://hdl.handle.net/11401/77311

Publisher

The Graduate School, Stony Brook University: Stony Brook, NY.

Format

application/pdf

Abstract

Networks today rely on network functions or middleboxes (e.g. | firewalls, WAN optimizers) to provide critical performance, security, and policy compliance capabilities. However, today the management of these middleboxes is hard. First, these middleboxes are implemented as dedicated hardware appliances, making it difficult to dynamically scale resources. Second, operators need to carefully plan the network topology, manually set up rules to route traffic through the desired sequence of middleboxes, and implement safeguards for correct operation in the presence of failures and overload. We can overhaul today’s network infrastructure to address these problems by introducing flexibility in routing and the implementation of these middleboxes. However, from network operator’s perspective a key question is whether we can address these problems in a minimally disruptive manner, e.g. | which require minimal changes to existing middlebox implementations and routing mechanisms. In this thesis, I describe two case studies for introducing more flexibility in middlebox management with minimal changes to existing middlebox implementations and routing mechanisms. In the first part of the thesis, I describe SIMPLE, a Software-Defined Networking (SDN) based efficient middlebox traffic steering solution which works with existing middlebox implementations and uses existing SDN APIs. In SIMPLE, I address algorithmic and system design challenges to demonstrate the feasibility of using SDN to simplify middlebox traffic steering. In the second part of the thesis, I describe KLEIN a cellular core re-design that uses Network Function Virtualization (NFV) and smart resource management, stays within the confines of current cellular standards and uses legacy routing in the core network. I address key challenges w.r.t. scalability, responsiveness and in realizing KLEIN via backwards-compatible orchestration mechanisms. | Networks today rely on network functions or middleboxes (e.g. | firewalls, WAN optimizers) to provide critical performance, security, and policy compliance capabilities. However, today the management of these middleboxes is hard. First, these middleboxes are implemented as dedicated hardware appliances, making it difficult to dynamically scale resources. Second, operators need to carefully plan the network topology, manually set up rules to route traffic through the desired sequence of middleboxes, and implement safeguards for correct operation in the presence of failures and overload. We can overhaul today’s network infrastructure to address these problems by introducing flexibility in routing and the implementation of these middleboxes. However, from network operator’s perspective a key question is whether we can address these problems in a minimally disruptive manner, e.g. | which require minimal changes to existing middlebox implementations and routing mechanisms. In this thesis, I describe two case studies for introducing more flexibility in middlebox management with minimal changes to existing middlebox implementations and routing mechanisms. In the first part of the thesis, I describe SIMPLE, a Software-Defined Networking (SDN) based efficient middlebox traffic steering solution which works with existing middlebox implementations and uses existing SDN APIs. In SIMPLE, I address algorithmic and system design challenges to demonstrate the feasibility of using SDN to simplify middlebox traffic steering. In the second part of the thesis, I describe KLEIN a cellular core re-design that uses Network Function Virtualization (NFV) and smart resource management, stays within the confines of current cellular standards and uses legacy routing in the core network. I address key challenges w.r.t. scalability, responsiveness and in realizing KLEIN via backwards-compatible orchestration mechanisms. | 138 pages

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.