PSRR: a scheme for time-bounded fault tolerance in distributed object-based systems

Abstract

An extension of the conventional object structuring approach, called the RTO.k object structuring approach, has been established as a unified scheme for object-oriented structuring of both real-time applications and non-real-time applications while enabling the system designer to provide design-time guarantees of timely service capabilities of the objects designed. In another area, the DRB/PSP scheme has been established as a concrete scheme for achieving scalable time-bounded fault tolerance in distributed and parallel computer systems. We present a new scheme called the primary-shadow (PS)-RTO.k replication (PSRR) scheme that integrates the RTO.k object structuring scheme and the basic principle of the DRB/PSP scheme. A partial validation of the PSRR scheme has been performed through incorporation of a simple version of the scheme into a defense application running on a PC LAN. This paper first introduces a new structuring rule that can be imposed on the RTO.k object structuring scheme in order to further simplify the task of the system designer in providing design-time guarantee of timely service capabilities of application systems. Thereafter, the core of the PSRR scheme, the basic operational rules and the basic structuring rules, are discussed.