An empirical study of the reliability of UNIX utilities
Communications of the ACM
Handbook of software reliability engineering
Handbook of software reliability engineering
Measuring Software Dependability by Robustness Benchmarking
IEEE Transactions on Software Engineering
The Future of Systems Research
Computer
Workshop on Dependability Benchmarking
DSN '02 Proceedings of the 2002 International Conference on Dependable Systems and Networks
An approach towards benchmarking of fault-tolerant commercial systems
FTCS '96 Proceedings of the The Twenty-Sixth Annual International Symposium on Fault-Tolerant Computing (FTCS '96)
Comparing Operating Systems Using Robustness Benchmarks
SRDS '97 Proceedings of the 16th Symposium on Reliable Distributed Systems
A Framework for Assessing Dependability in Distributed Systems with Lightweight Fault Injectors
IPDS '00 Proceedings of the 4th International Computer Performance and Dependability Symposium
Measuring Robustness of a Fault-Tolerant Aerospace System
FTCS '95 Proceedings of the Twenty-Fifth International Symposium on Fault-Tolerant Computing
Towards availability benchmarks: a case study of software raid systems
ATEC '00 Proceedings of the annual conference on USENIX Annual Technical Conference
| Hi-index | 0.00 |
Benchmarks have historically guided research and development in computer systems performance. As availability becomes an increasingly important metric for evaluating computer systems, there is a natural demand for availability benchmarks. However, industry-wide availability benchmarks do not exist today. This paper proposes a framework for benchmarking the availability of computer systems. Our goal in this effort is to define a framework which can be used to implement availability benchmarks that are applicable across a wide range of systems. The proposed framework is a hierarchical approach which goes beyond direct benchmarking of availability in terms of number of outage minutes per year. This paper identifies the three key attributes impacting system availability (fault and maintenance rate, robustness, and recovery), and methods to benchmark these key attributes of availability.