Nov 29, 2017
Jolley Hall, Room 309
"Real-Time Communication in Cloud Environments"
Adviser: Chenyang Lu
Real-time communication is critical to emerging cloud applications from smart cities to 5G core networks. The new class of latency-critical applications requires latency differentiation and performance isolation in a highly scalable fashion in a virtualized cloud environment. This thesis research aims to develop novel cloud architecture and services to support real-time communication at both the platform and infrastructure layers. At the platform layer, we are building RTM, a real-time messaging middleware that features (1) latency differentiation through prioritization, (2) performance isolation through rate limits, and (3) scalability through load distribution among message brokers. A key contribution of RTM lies in the exploitation of the complex interactions among performance isolation and load distribution in message broker clusters. At the infrastructure layer, we are developing VATC, a virtualization-aware traffic control framework in virtualized hosts. VATC provides a network I/O layer that achieves differentiated packet processing, rate limits, and scalability through multiple-core support. The research will be evaluated in a cloud testbed in the context of smart city and Internet of Things applications.