Programmable cyber networks for critical infrastructure

Kumar, Rakesh

Permalink

https://hdl.handle.net/2142/106221 Copy

Description

Title

Programmable cyber networks for critical infrastructure

Author(s)

Kumar, Rakesh

Issue Date

2019-11-27

Director of Research (if dissertation) or Advisor (if thesis)

Nicol, David M.

Doctoral Committee Chair(s)

Nicol, David M.

Committee Member(s)

• Bailey, Michael D
• Vaidya, Nitin H.
• Caesar, Matthew

Department of Study

Electrical & Computer Eng

Discipline

Electrical & Computer Engr

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

Critical Infrastructure Networking Reliability Resilience Security

Abstract

The operational integrity of the infrastructure systems is critical for a nation state’s economic and security interests. Such systems rely on automation to perform complex control tasks to avoid malfunction caused by human error. However, such automation requires coordination of their components. Such coordination relies on several guarantees of safety and performance from their underlying cyber networks. It is challenging to provide such guarantees using traditional networks due to their rigid feature set and distributed, opaque and non-standard control interfaces. The central goal of this dissertation is to develop a set of design tools that use network programmability to achieve end-to-end delay, access control and resiliency guarantees for critical infrastructure (CI) applications. We propose and evaluate the architecture and design of several tools to address these guarantees singularly and simultaneously. With the standardized control and data-planes, the computational analysis of the centralized network configurations has emerged as a powerful approach for solving a variety of problems. We used this approach in one of our analysis tools to simultaneously validate access control and resilience of networks. We also used an analytic approach to assess the resiliency of CI network with the use of metrics computed by using Monte Carlo methods. To that end, we built a data-plane simulator to enable such computation. Furthermore, it is now feasible to synthesize desired behavior in a programmable CI network to meet the performance and resilience goals for individual application flows. We used the synthesis approach to build a tool that uses efficient centralized algorithms to synthesize control-plane configuration resulting in flows meeting their end-to-end delay deadlines. We also demonstrate a framework that uses synthesis at the intersection of control and data planes to implement network coding to achieve seamless resiliency for network flows.

Graduation Semester

2019-12

Type of Resource

text

Permalink

http://hdl.handle.net/2142/106221

Copyright and License Information

Copyright 2019 Rakesh Kumar

Owning Collections