题目: Bringing the Hardware Resource Revolution to Safety-Critical Real-Time Embedded Systems

时间:2014年8月12日，上午9点

地点:东五楼210

Abstract:

The recent hardware resource revolution (e.g., multicore and heterogeneous accelerators) is significantly enhancing system capability in various domains, except for safety-critical real-time embedded systems that often require certification. The reason is a lack of timing predictability associated with utilizing multiple shared hardware components. An extensive set of provably strong methods exists that allows CPU resources to be efficiently allocated in a timing predictable way. Unfortunately, when systems utilize additional shared hardware resources, present technology makes very pessimistic assumption regarding the utilization of resources to certify that timing violations do not occur. The resulting processing capacity loss can easily neutralize the impact of any additional resources. In this talk, I will illustrate the fundamental reason why handling additional resources besides CPUs in a timing predictable way is difficult: when applications access resources other than CPUs during execution, they will be suspended by the operating system. This introduces unpredictable suspension delays that negatively impact timing predictability. By identifying this principle problem, I will show that the timing predictability problems associated with utilizing different categories of resources can all be mapped (obviously with their own semantics) to the problem of scheduling and analyzing real-time application systems that may experience suspension delays. I will further describe our proposed new set of efficient suspension-aware methods that can solve this timing predictability problem associated with utilizing multiple shared hardware resources.

Bio:

Cong Liu is currently a tenure-track assistant professor in the Department of Computer Science at the University of Texas at Dallas, after obtaining his Ph.D in Computer Science from the University of North Carolina at Chapel Hill in summer 2013. His current research focuses on Real-Time and Embedded Systems, Battery-Powered Cyber-Physical Systems, and Mobile and Cloud Computing. He is the author and co-author of over 40 papers in premier journals and conferences such as RTSS, ICCPS, ECRTS , RTAS, EMSOFT, ICNP, INFOCOM. He received the Best Student Paper Award at the 30th IEEE Real-Time Systems Symposium, the premier real-time and embedded systems conference; he also received the best papers award at the 17th RTCSA. He is a member of ACM and IEEE