Title: Soft real-time processing with dynamic qos level resource management
Abstract: As desktop computers experience dramatic increases in computational power, it is becoming possible to run applications that have in the past been feasible only on dedicated systems. In particular, high throughput applications such as video display and processing, real-time hardware control, and virtual reality are being executed on desktop systems at rates formerly achievable only with special-purpose computers with real-time operating systems or no operating system at all. There are two significant differences between these desktop applications and traditional real-time applications: the applications are executed on top of a general purpose operating system and the real-time constraints are non-critical. However, some level of real-time support is still needed in order to provide acceptable levels of application performance. Some researchers have focused on developing soft real-time operating systems that provide real-time application support but allow the applications to miss some or all deadlines. This thesis focuses on an alternative and significantly more general method called Quality of Service Levels (originally proposed by Hideyuki Tokuda at Carnegie-Mellon University) in which applications are developed with multiple modes in which they can execute. Each mode provides a different output quality and consumes a different amount of resources. A centralized resource manager dynamically allocates the available resources among the running applications according to the levels that the applications have specified. This thesis significantly expands on Tokuda's ideas in exploring the QoS Level model of soft real-time application execution and presents results from and analysis of a working system based on this model.
Publication Year: 1999
Publication Date: 1999-01-01
Language: en
Type: article
Access and Citation
Cited By Count: 5
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot