On Quality of Service Optimization with Discrete QoS Options

We present a QoS management framework that enables us to quantitatively measure QoS, and to analytically plan and allocate resources. In this model, end users' quality preferences are considered when system resources are apportionedacross multipleapplicationssuch that the net utility that accrues to the end-users is maximized. In [23][24], we primarily worked with continuous QoS dimensions, and assumed that the utility gained by improvements along a QoS dimension were always representable by concave functions. In this paper, we relax both assumptions. One, we support discrete QoS operating points. Two, we make no assumptions about the concavity of the utilityfunctions. Using these as the basis, we tackle the problem of maximizing system utility by allocating a single finite resource to satisfy the QoS requirements of multiple applications along multiple QoS dimensions. We present two near-optimal algorithms to solve this problem. The first yields an allocation within a known boun...

• Type: Conference
• Year: 1999
• Where: RTAS
• Authors: Chen Lee, John P. Lehoczky, Ragunathan Rajkumar, Daniel P. Siewiorek
• Keywords: Continuous Qos Dimensions, Embedded Systems, Qos Dimensions, QoS Management Framework, RTAS 1999