|
|  |
Washington University in St. Louis News & Information > Faculty Experts at Washington University in St. Louis >

Assistant Professor Of Computer Science & Engineering
Expertise: middleware, cybersecurity, changing environments, real-time behavior
Bio: Gill received his doctorate from Washington University. He is widely published and holds research interests in real-time, fault-tolerant, secure, and embedded middleware hybrid static/dynamic resource management, adaptive and reflective distributed systems, distributed object computing, real-time Java, distributed real-time Java, cybersecurity, network security, software security. Next-generation distributed and embedded real-time systems have requirements to respond adaptively to rapidly changing environmental conditions. Especially in security and surveillance contexts, these changes may occur over a range of time scales, and may require strict real-time constraints with limited resources. To build and maintain these next-generation systems feasibly and affordably, it is essential to capture the fundamental patterns, algorithms, protocols, and architectures that facilitate integration and hybridization of static and dynamic resource management, system configuration, and admission control strategies. In surveillance applications, these patterns of computational scheduling and system usage must be related to the dynamics of the world in order to optimize system performance for specific tasks.
WUSTL Contact Information:
|
|
| E-mail: | cdgill@wustl.edu |
| Address: | One Brookings Dr. Campus Box 1045 St. Louis, MO 63130
|
|
Education:
-
Sc.B. in English and Biology at Washington University
-
M.S. in Computer Science at University of Missouri-Rolla
-
Ph.D. in Computer Science at Washington University
Additional Background: As the demand for mission-critical distributed embedded real-time systems grows, the demand for predictable, robust, and efficient real-time software increases. Next-generation distributed and embedded real-time systems have requirements to respond adaptively to rapidly changing environmental conditions over a range of time scales, while still meeting strict real-time constraints with limited resources. To build and maintain these next-generation systems feasibly and affordably, it is essential to capture the fundamental patterns, algorithms, protocols, and architectures that facilitate integration and hybridization of static and dynamic resource management, system configuration, and admission control strategies. Gill's doctoral research has explored the benefits of hybridizing and optimizing static and dynamic scheduling strategies for real-time application quality of service. To support this, he has developed, analyzed, and integrated novel policies and mechanisms for admission control and scheduling that provide real-time behavior of adaptive middleware-centric scheduling under the direction of higher-level resource management. His research is now being used in a variety of industrial research systems, such as the Bold Stroke avionics mission- computing platform at Boeing, and is targeted for potential transition to many production systems in the future. His work has also influenced and augmented other research on resource management in middleware at Lockheed Martin, Honeywell, and BBN Technologies under the DARPA Quorum program, the Air Force Research Labs (AFRL) ASFD and ASTD programs, and the AFRL / Open Systems Joint Task Force (OS/JTF) WSOA program. His current and future research interests focus on the following topics:
- Discovering patterns and implementing frameworks for combining, hybridizing, and extending proven techniques for resource management, system configuration, and admission control, to offer new approaches to building adaptive distributed embedded real-time systems.
- Experimental development, performance measurement, empirical analysis, and concurrent visualization of distributed embedded real-time systems.
- Providing application-tailored middleware optimizations using advanced programming models and programming language techniques such as reflection, generics, aspect weaving, and domain-specific type systems, to constrain the complexity and cost of building these systems while retaining expressive power.
|
|  |
|