Scheduling theory maintains that there are fundamental similarities in problems of sequence that transcend the characteristics of the particular tasks to be ordered or the resources to be used. Traditionally, scheduling policies are implemented using algorithms; we study scheduling algorithms to discover the various properties of the schedules they produce. To facilitate analysis the policies are typically limited to homogeneous task sets (e.g., all periodic tasks) and consider only one or very

more »

... few task attributes. In some cases the results are so attractive that the task sets of systems are made to fit the algorithm rather than using a policy more appropriate to the system. We therefore make the following observation: if scheduling policies are driven by how well they can be expressed and analyzed, then we need a more general framework for expressing scheduling policies. We introduce the Importance Abstraction as a general scheduling framework. The scheduling algorithm is invariant: choose the most important task at every point in time. Each task is described by a function, called an importance function, that profiles the task's importance to the system over time. The importance abstraction can express not only the traditional scheduling policies but a wide range of new policies based on how important individual tasks are to the system. Since the scheduling policy is described using functions rather than a single algorithm we can exploit the maturity of mathematical proof techniques when analyzing the schedule produced by the policy. Since this abstraction is applicable to any system of tasks and processors, we examine the communication subsystem as an example, and find that importance functions facilitate the expression of message discrimination policies as well as help unify scheduling across the operating system/ communication subsystem domain boundary. iv Acknowledgements I thank and acknowledge the people who have helped and encouraged me as I pursued this degree-there were of course more than I can name here. Over the last six years I have had the pleasure of working with and learning from one of the most talented collections of people in this department: the Computer Networks Laboratory, in particular, And everyone knows that 228 was the coolest office. Thanks for all the softball games, the cookouts, the parties, and the tightness of a circle of friends. I am grateful to Lee Cohen, my abiding best friend from college days, for providing an ear to bend, a place to hide, an excellent Bloody Mary to drink, and support as I plowed through this degree. v I owe a substantial portion of this degree to Carmen Pancerella who, through all of our ups and downs, has remained my best friend and most trusted confidant. She pulled me when I got stuck, pushed me when I got stubborn, and never let me forget that life is larger than graduate school. She kept my stomach full, my head from getting too large, my appointments from being missed, and my days from being boring. I would be nowhere without the love and support of my family. They have heard the phrase "I'm about a year away from finishing" often over the last three years, yet they always believed, sometimes more than I, that I'd finish. I very much appreciate the effort and guidance of my committee, Andrew Grimshaw, Sang Son, Jim Aylor, Bill Wulf, and Alf Weaver. When I first had some of the ideas presented in this dissertation, all I saw were dead-ends and limitations. I am forever grateful to Bill for seeing past these obstacles and sometimes having to convince me, rather than the other way around, that these ideas were good. My thank-you's will never repay all of the time Bill has spent with me. My greatest appreciation of all goes to Alf Weaver. From the beginning of my tenure at Virginia until the very last moments of our advisor/advisee relationship, Alf has given far beyond what was required. He has always treated me as a respected colleague, selflessly promoting me and all of his students above himself. I have learned a great deal from Alf, not just concerning research, but more importantly, how to make the most with what you've got, and that most of what you've got comes from the people around you. And all I know about wines, I learned from him. A person like Alf comes into one's life perhaps once, if one is really lucky. Alf is more than a teacher and advisor; Alf is my friend.