On-line adaptive algorithms in autonomic restart control

Authors:
Hiroyuki Okamura;Tadashi Dohi;Kishor S. Trivedi
Affiliations:
Department of Information Engineering, Graduate School of Engineering, Hiroshima University, Higashi-Hiroshima, Japan;Department of Information Engineering, Graduate School of Engineering, Hiroshima University, Higashi-Hiroshima, Japan;Department of Electrical and Computer Engineering, Duke University, Durham, NC
Venue:
ATC'10 Proceedings of the 7th international conference on Autonomic and trusted computing
Year:
2010

Citing 18
Cited 1

Analysis of Preventive Maintenance in Transactions Based Software Systems

IEEE Transactions on Computers
Introduction to Reinforcement Learning

Introduction to Reinforcement Learning
Achieving Fault-Tolerant Software with Rejuvenation and Reconfiguration

IEEE Software
Software Rejuvenation: Analysis, Module and Applications

FTCS '95 Proceedings of the Twenty-Fifth International Symposium on Fault-Tolerant Computing
Simulation-Based Optimization: Parametric Optimization Techniques and Reinforcement Learning

Simulation-Based Optimization: Parametric Optimization Techniques and Reinforcement Learning
A Dynamic Checkpointing Scheme Based on Reinforcement Learning

PRDC '04 Proceedings of the 10th IEEE Pacific Rim International Symposium on Dependable Computing (PRDC'04)
Analysis and Algorithms for Restart

QEST '04 Proceedings of the The Quantitative Evaluation of Systems, First International Conference
A Comprehensive Model for Software Rejuvenation

IEEE Transactions on Dependable and Secure Computing
Performance Assurance via Software Rejuvenation: Monitoring, Statistics and Algorithms

DSN '06 Proceedings of the International Conference on Dependable Systems and Networks
The Fast and the Fair: A Fault-Injection-Driven Comparison of Restart Oracles for Reliable Web Services

QEST '06 Proceedings of the 3rd international conference on the Quantitative Evaluation of Systems
Performability analysis of clustered systems with rejuvenation under varying workload

Performance Evaluation
Analysis of Restart Mechanisms in Software Systems

IEEE Transactions on Software Engineering
A Reinforcement Learning Approach to Automatic Error Recovery

DSN '07 Proceedings of the 37th Annual IEEE/IFIP International Conference on Dependable Systems and Networks
Non-parametric Predictive Inference of Preventive Rejuvenation Schedule in Operational Software Systems

ISSRE '07 Proceedings of the The 18th IEEE International Symposium on Software Reliability
A Faster Estimation Algorithm for Periodic Preventive Rejuvenation Schedule Maximizing System Availability

ISAS '07 Proceedings of the 4th international symposium on Service Availability
Simulation-Based Optimization Approach for Software Cost Model with Rejuvenation

ATC '08 Proceedings of the 5th international conference on Autonomic and Trusted Computing
A measurement study of the interplay between application level restart and transport protocol

ISAS'04 Proceedings of the First international conference on Service Availability
Analysis of a service degradation model with preventive rejuvenation

ISAS'06 Proceedings of the Third international conference on Service Availability

Analysis of revenue improvements with runtime adaptation of service composition based on conditional request retries

ESOCC'12 Proceedings of the First European conference on Service-Oriented and Cloud Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Restarts or retries are typical control schemes to meet a deadline in real-time systems, and are regarded as significant environmental diversity techniques in dependable computing. This paper reconsiders a restart control studied by van Moorsel and Wolter (2006), and refines their result from theoretical and statistical points of views. Based on the optimality principle, we show that the time-fixed restart time is best even in non-stationary control setting under the assumption of unbounded restart opportunities. Next we study statistical inference for the restart time interval and develop on-line adaptive algorithms for estimating the optimal restart time interval via non-parametric estimation and reinforcement learning. Finally, these algorithms are compared in a simulation study.