Haohan Li

Ph.D. Candidate
Department of Computer Science
Campus Box 3175, Sitterson Hall
The University of North Carolina at Chapel Hill
Chapel Hill, NC 27599-3175 USA

Office: FB132, Brooks Building
Phone: 919-962-1935
Email: lihaohan#cs.unc.edu

I am going to finish my Ph.D. by May 2013. I am currently on the academic job market. My job application package is available here.

Short Bio

I am currently a Ph.D. candidate in the Department of Computer Science at University of North Carolina at Chapel Hill, under the supervision of Professor Sanjoy K. Baruah. I am a member of the Real-Time Systems Group. I received my B.E. at the Department of Computer Science and Engineering at Shanghai Jiao Tong University in 2008.

Research Overview

My research focuses on the design and analysis of scheduling policies and resource allocation mechanisms in computation-intensive and time-sensitive cyber-physical embedded systems. My dissertation, Scheduling Mixed-Criticality Real-Time Systems, addresses the following question: In a multitasking system where every task requires a real-time response while the estimation of the worst-case execution times of these tasks is difficult, how do we assign task priorities so that 1) the safety-critical tasks are asserted to be completed within a specified length of time, and 2) the non-critical tasks are also guaranteed to be completed within a predictable length of time if no task is actually consuming time at the worst case?

The mixed-criticality real-time system model is invented to answer this question. It defines multiple worst-case execution scenarios, and demands a scheduling policy to provide provable timing guarantees to each level of critical tasks with respect to each type of scenario. Two scheduling algorithms are proposed and experimentally implemented to serve this model:

  • OCBP Algorithm: It is aimed at discrete one-shot tasks. This algorithm is applicable to arbitrary number of criticality levels and is proved to optimally minimize the percentage of computational resource waste. This algorithm can be extended to recurrent tasks with acceptable run-time complexity.
  • EDF-VD Algorithm: It is aimed at recurrent tasks where each task is required to finish within its period. This algorithm is applicable to two criticality levels (safety-critical and non-critical) and is also proved to optimally minimize the percentage of computational resource waste. This algorithm has very low run-time complexity, and can be extended to both global schedulers and partitioned schedulers on multiprocessor platforms.

Publications

Journal Papers

Conference Papers

Workshop Papers

  • Sanjoy Baruah, Vincenzo Bonifaci, Gianlorenzo D'Angelo, Haohan Li, Alberto Marchetti-Spaccamela, Nicole Megow and Leen Stougie. Mixed-criticality scheduling. 10th Workshop on Models and Algorithms for Planning and Scheduling Problems (MAPSP), Nymburk, Czech Republic. June 2011.

Teaching

  • Spring 2013: Sole instructor of COMP110-003: Introduction to Programming.
  • Fall 2009: Teaching assistant of COMP116: Introduction to Scientific Programming.
  • Spring 2009: Teaching assistant of COMP541: Digital Logic and Computer Design.
  • Fall 2008: Teaching assistant of COMP110: Introduction to Programming (WWW based).

Awards

Selected Coursework

Miscellaneous

Locations of visitors to this page