[PAST EVENT] Zijiang Hao, Computer Science - Ph.D. Oral Exam
Recent years have witnessed the development of distributed computing. New types of distributed application scenarios have emerged, including mobile-cloud computing, inter-vehicle networking, edge computing, and so on. They bring both challenges and opportunities.
In this oral exam, I will present three research projects based on these new distributed application scenarios. The first one is SMOC, a secure and highly usable platform for mobile-cloud computing. SMOC has two design fundamentals. First, applications can freely migrate between the user's mobile device and a backend cloud server. We design a file system extension to enable this feature, so users can freely choose to run their applications either in the cloud (for rich hardware resources and high security guarantees), or on their local mobile device (for better user experience). Second, in order to protect user data on the mobile device, we leverage the hardware virtualization technology, which isolates the data from the local mobile device operating system. Evaluation on our prototype system reveals that SMOC is efficient, easy-to-use, and secure.
The second one is EdgeEngine, an efficient and customizable framework for edge computing. EdgeEngine can greatly simplify the development of edge computing applications. In most cases, the developer only needs to implement the main logic of the application, while EdgeEngine can efficiently execute the application on the underlying edge infrastructure, by applying a group of default policies. For the special cases where the developer's knowledge is necessary for achieving satisfactory performance, EdgeEngine provides a flexible programming interface, through which the developer can implement user-specified policies with very little effort, helping the system efficiently execute the application on the edge infrastructure. Evaluation on our prototype system reveals that EdgeEngine works efficiently in the real world.
We also propose EdgePaxos, a storage system that achieves both data reliability and global data availability in edge environments. EdgePaxos provides high performance by organizing edge nodes into Paxos groups and carefully incorporating them with a backend cloud. We outline the main idea of EdgePaxos with convincing application scenarios.
Zijiang Hao is a Ph.D. candidate in the Computer Science Department at William & Mary. His advisor is Prof. Qun Li. His research interests include mobile-cloud computing, edge computing, geo-distributed storage systems, consensus algorithms, and machine learning. Before joining William & Mary, Zijiang Hao had been working in the industry for four years. He has received his M.S. from Tsinghua University (Beijing, China) in 2007 and his B.S. from Tsinghua University (Beijing, China) in 2004.