Downloads & Free Reading Options - Results

Emulation techniques are proposed as a solution to a difficulty arising in the analysis of the reliability of highly reliable computer systems for future commercial aircraft. The difficulty, viz., the lack of credible precision in reliability estimates obtained by analytical modeling techniques are established. The difficulty is shown to be an unavoidable consequence of: (1) a high reliability requirement so demanding as to make system evaluation by use testing infeasible, (2) a complex system design technique, fault tolerance, (3) system reliability dominated by errors due to flaws in the system definition, and (4) elaborate analytical modeling techniques whose precision outputs are quite sensitive to errors of approximation in their input data. The technique of emulation is described, indicating how its input is a simple description of the logical structure of a system and its output is the consequent behavior. The use of emulation techniques is discussed for pseudo-testing systems to evaluate bounds on the parameter values needed for the analytical techniques.

This document surveys some of the more important theoretical results about the structure of the reliability function of a k out of n system, and indicates how these results may be used to obtain easily calculable bounds for the reliability of a specified k out of n system. (Author)

This document surveys some of the more important theoretical results about the structure of the reliability function of a k out of n system, and indicates how these results may be used to obtain easily calculable bounds for the reliability of a specified k out of n system. (Author)

The randomization technique for computing transient probabilities of Markov processes is presented. The technique is applied to a Markov process model of a simplified fault tolerant computer system for illustrative purposes. It is applicable to much larger and more complex models. Transient state probabilities are computed, from which reliabilities are derived. An accelerated version of the randomization algorithm is developed which exploits ''stiffness' of the models to gain increased efficiency. A great advantage of the randomization approach is that it easily allows probabilities and reliabilities to be computed to any predetermined accuracy.

The randomization technique for computing transient probabilities of Markov processes is presented. The technique is applied to a Markov process model of a simplified fault tolerant computer system for illustrative purposes. It is applicable to much larger and more complex models. Transient state probabilities are computed, from which reliabilities are derived. An accelerated version of the randomization algorithm is developed which exploits ''stiffness' of the models to gain increased efficiency. A great advantage of the randomization approach is that it easily allows probabilities and reliabilities to be computed to any predetermined accuracy.

A hardware unit has been designed that reduces the cost, in terms of performance and power consumption, for implementing N-modular redundancy (NMR) in a multiprocessor device. The innovation monitors transactions to memory, and calculates a form of sumcheck on-the-fly, thereby relieving the processors of calculating the sumcheck in software

In this paper, a new modeling methodology to characterize failure processes in Time-Sharing systems due to hardware transients and software errors is summarized. The basic assumption made is that the instantaneous failure rate of a system resource can be approximated by a deterministic function of time plus a zero-mean stationary Gaussian process, both depending on the usage of the resource considered. The probability density function of the time to failure obtained under this assumption has a decreasing hazard function, partially explaining why other decreasing hazard function densities such as the Weibull fit experimental data so well. Furthermore, by considering the Kernel of the Operating System as a system resource, this methodology sets the basis for independent methods of evaluating the contribution of software to system unreliability, and gives some non obvious hints about how system reliability could be improved. A real system has been characterized according to this methodology, and an extremely good fit between predicted and observed behavior has been found. Also, the predicted system behavior according to this methology is compared with the predictions of other models such as the exponential, Weibull, and periodic failure rate. (Author)

In this paper a new modeling methodology to characterize failure processes in Time-Sharing systems due to hardware transients and software errors is summarized. The basic assumption made is that the instantaneous failure rate of a system resource can be approximated by a deterministic function of time plus a zero-mean stationary Gaussian process, both depending on the usage of the resource considered. The probability density function of the time to failure obtained under this assumption has a decreasing hazard function, partially explaining why other decreasing hazard function densities such as the Weibull fit experimental data so well. Furthermore, by considering the Kernel of the Operating System as a system resource, this methodology sets the basis for independent methods of evaluating the contribution of software to system unreliability, and gives some non obvious hints about how system reliability could be improved. A real system has been characterized according to this methodology, and an extremely good fit between predicted and observed behavior has been found. Also, the predicted system behavior according to this methology is compared with the predictions of other models such as the exponential, Weibull, and periodic failure rate. (Author)