ACM Transactions on Computer Systems (TOCS)
Dependability modeling of a heterogeneous VAX-cluster system using stochastic reward nets
Hardware and software fault tolerance in parallel computing systems
Availability of CAUTRA, a Subset of the French Air Traffic Control System
IEEE Transactions on Computers
Distributed Shared Memory: Concepts and Systems
Distributed Shared Memory: Concepts and Systems
EDCC-2 Proceedings of the Second European Dependable Computing Conference on Dependable Computing
SPNP: Stochastic Petri Net Package
PNPM '89 The Proceedings of the Third International Workshop on Petri Nets and Performance Models
LOGISIRE, a Safe Computer System for Process Automation
Fehlertolerierende Rechensysteme / Fault-Tolerant Computing Systems, 3. Internationale GI/ITG/GMA-Fachtagung
Towards Dependability Modeling of FT-CORBA Architectures
EDCC-4 Proceedings of the 4th European Dependable Computing Conference on Dependable Computing
SAFECOMP '01 Proceedings of the 20th International Conference on Computer Safety, Reliability and Security
Performance Validation of Fault-Tolerance Software: A Compositional Approach
DSN '01 Proceedings of the 2001 International Conference on Dependable Systems and Networks (formerly: FTCS)
| Hi-index | 0.00 |
This paper deals with the dependability evaluation of a Multipurpose, Multiprocessor System (MMS) under investigation by a system manufacturer. MMS is symmetric with a DSM and a CC-NUMA. As the system is scalable, we consider two architectures: a reference one composed of four nodes and an extended one with an additional spare node. A set of dependability and performability measures is evaluated for these architectures with two categories of application users, accepting different service degradation levels. Modeling is based on GSPN to which reward rates are added to evaluate performability measures. The originality of our approach is the clear separation between the architectural and the service concerns. The results allow us to quantify the influence of the main parameters and to compare the dependability of the systems under consideration. They can thus be used for supporting the manufacturer design choices as well as the potential user configuration choices.