In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

Yu Li

will defend his dissertation

Theoretical Fundamentals of Real-time Virtualization from the Resource Management Perspective

Abstract

A Virtual Machine Monitor (VMM) partitions a host physical machine into a group of Virtual Machines (VMs). Typically, a VM machine only preempts a part of a dedicated physical resource temporally or spatially. This fact greatly impacts the real-time task scheduling in VMs because most traditional real-time scheduling theories are based on dedicated resources. The real-time community has introduced some Hierarchical Real-Time Scheduling Models to address this issue. Among them, the Regularity-based Resource (RRP) Model is able to provide maximal transparency for task scheduling. However, the current theoretical results on the RRP Model are still far from the complete theoretical fundamentals required by a real-time VMM. At the resource level, only a naive algorithm has been found for resource partitioning. At the task level, only the Periodic Task Model is investigated, and even for this task model, only one simple case has been considered. This work explores the RRP Model at both the resource and task levels. On one hand, it is the first to solve the resource partitioning problem with both global and partitioned strategies. On the other hand, it solves the task scheduling problem with a strong result that the classic task scheduling problem in the RRP Model can be easily transformed into an equivalent problem on a dedicated resource. With these theory enhancement, a 2-layer real-time resource model is completely presented and the theoretical fundamentals of a real-time VMM are fully established from the resource management perspective.

Date: Wednesday, November 30, 2016
Time: 10:30 AM 
Place: PGH 550
Advisor: Dr. Albert M. K. Cheng

Faculty, students, and the general public are invited.