Adaptive fault tolerance

Abstract

The goal of adaptive fault tolerance(AFT) is to expand the envelope of dependable system operation in distributed, real-time systems. Such systems often experience substantial run-time changes in the types and distributions of faults, in the availability of resources, in data distribution, and in users' requirements for dependability and performance. Preliminary examples, such as Adaptable Distributed Recovery Blocks (Kim) and distributed crash recovery, illustrate how adaptive fault tolerance can provide useful tradeoffs among service properties such as error-recovery latency, throughput, and precision, over a wide range of operating conditions. A general methodology for AFT system design must address issues of (1) rapid, incremental diagnosis/estimation of environmental and internal state, (2) safe and effective control, and (3) efficient, parametric or multimode fault-tolerant implementations. A major challenge is to achieve the additional flexibility without excessive complexity, both for performance and reliability concerns. Reflective architecture, a form of meta-design, is an attractive framework for AFT system design and for adaptive systems in general. It provides for the monitoring and redefinition of system behavior in a hierarchical manner that may be integrated with conventional "uses"-based hierarchical design.

Fault Tolerance in Transform-Domain Adaptive Filters Operating With Real-Valued Signals

“…This structure employs a novel fault-tolerance mechanism known as adaptive fault tolerance (AFT) to overcome fixed fault conditions in the adaptive filter coefficients. It should be noted that the AFT mechanism discussed here is based on the inherent adaptive process of the system and is different from the AFT mechanism used in [21]. In addition to faults in filter coefficients, faults may also arise at the outputs of adders, multipliers, and registers that hold intermediate values.…”

“…The occurrence of a fault in the th coefficient and removal of th row and column in is equivalent to removing the th row and column of each of the submatrices in . Hence, can be written as (21) The operation in (21) is equivalent to finding the eigenvalues of a circulant matrix having the same form as in (19). This will result in zero eigenvalues and eigenvalues each equal to two.…”

Comparison of Model Predictive and Reinforcement Learning Methods for Fault Tolerant Control

“…Adaptivity: The controller should be able to remain functional under previously unseen faults (Goldberg et al (1993)). Speedy convergence: The agents behaviour should be responsive to faults as they occur.…”

Architecting resilient computing systems: A component-based approach for adaptive fault tolerance

“…AFT is gaining more importance with the increasing concern for lowering the amount of energy consumed by cyber-physical systems and the amount of heat they generate [25]. Conceptual frameworks for adaptive fault tolerance (AFT) describing algorithms and target systems are presented in [26,27]. Several CORBA-based middleware exist [28,9,8] but evolution is tackled differently: adaptation has a parametric form (e.g., number of replicas) or it is performed off-line or, if done on-line, has a coarse-grained nature.…”