Control engineering for computing systems - industry experience and research challenges
Authors:
- D.M. TilburyDTD.M. Tilbury
- 1 December 2005
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems
- Vol. 25 (6) , 56-68
Abstract
Control theory has an important role to play in developing new computing systems, especially complicated software systems. Control problems that commonly occur in computing systems include regulating service levels, optimizing resources, and rejecting disturbances such as variable workloads. Developing control systems to address these problems involves a number of challenges. The article describes some of the challenges that arise in applying control theory to computing systems.Keywords
This publication has 28 references indexed in Scilit:
- A duality model of TCP and queue management algorithmsIEEE/ACM Transactions on Networking, 2003
- Feedback performance control in software servicesIEEE Control Systems, 2003
- Trends in software and controlIEEE Control Systems, 2003
- Dynamic power management for nonstationary service requestsIEEE Transactions on Computers, 2002
- Congestion control as a stochastic control problem with action delaysAutomatica, 1999
- A control-based middleware framework for quality-of-service adaptationsIEEE Journal on Selected Areas in Communications, 1999
- Differential priority-based adaptive rate service discipline for QoS guarantee of video streamComputer Communications, 1997
- Feedback control of congestion in packet switching networks: the case of a single congested nodeIEEE/ACM Transactions on Networking, 1993
- Achieving service rate objectives with decay usage schedulingIEEE Transactions on Software Engineering, 1993
- A fair share schedulerCommunications of the ACM, 1988