On the theory of system administration
Science of Computer Programming
Feedback Control of Computing Systems
Feedback Control of Computing Systems
Configurable immunity for evolving human-computer systems
Science of Computer Programming - Methods of software design: Techniques and applications
LISA '98 Proceedings of the 12th USENIX conference on System administration
Seeking Closure in an Open World: A Behavioral Agent Approach to Configuration Management
LISA '03 Proceedings of the 17th USENIX conference on System administration
Experience in Implementing an HTTP Service Closure
LISA '04 Proceedings of the 18th USENIX conference on System administration
Experience implementing an IP address closure
LISA '06 Proceedings of the 20th conference on Large Installation System Administration
AIMS '08 Proceedings of the 2nd international conference on Autonomous Infrastructure, Management and Security: Resilient Networks and Services
Dynamics of Resource Closure Operators
AIMS '09 Proceedings of the 3rd International Conference on Autonomous Infrastructure, Management and Security: Scalability of Networks and Services
An approach to understanding policy based on autonomy and voluntary cooperation
DSOM'05 Proceedings of the 16th IFIP/IEEE Ambient Networks international conference on Distributed Systems: operations and Management
Combining Learned and Highly-Reactive Management
MACE '09 Proceedings of the 4th IEEE International Workshop on Modelling Autonomic Communications Environments
On the combined behavior of autonomous resource management agents
AIMS'10 Proceedings of the Mechanisms for autonomous management of networks and services, and 4th international conference on Autonomous infrastructure, management and security
| Hi-index | 0.01 |
We present a resource management algorithm based upon guided "walks" within a system state space. Walks are guided via simple predictions of optimum behavior whose accuracy increases as system state approaches a predicted optimum. Optimum behavior is defined as maximizing payoff, which is the difference between value of provided service and cost of providing the service. Feedback between prediction, movement in the state space, and direct observation of behavior allows the algorithm to track optimum payoff, even though there is no detailed model of system behavior. Efficiency of the algorithm is defined as the ratio between observed and optimum payoffs, and can be estimated without reference to a detailed model. We demonstrate by simulation that, under commonly encountered conditions, our algorithm can achieve near-optimal behavior. Our strategy is thus a potentially viable alternative to management based upon closed control loops in many practical situations.