Reliability of fault-tolerant systems with parallel task processing

Abstract

The paper considers performance and reliability of fault-tolerant software running on a hardware system that consists of multiple processing units. The software consists of functionally equivalent but independently developed versions that start execution simultaneously. The computational complexity and reliability of different versions are different. The system completes the task execution when the outputs of a pre-specified number of versions coincide. The processing units are characterized by different availability and processing speed. It is assumed that they are able to share the computational burden perfectly and that execution of each version can be fully parallelized. The algorithm based on the universal generating function technique is used for determining the distribution of system task execution time. This algorithm allows analysts to evaluate complex hardware-software reliability and performance indices such as expected task execution time and probability that the task is completed within a given time. Illustrative examples are also presented. The algorithm based on the universal generating function technique is used for determining the distribution of system task execution time. This algorithm allows analysts to evaluate complex hardware-software reliability and performance indices such as expected task execution time and probability that the task is completed within a given time. Illustrative examples are presented.

An efficient algorithm for the multi-state two separate minimal paths reliability problem with budget constraint

“…This way, we obtain all the system solutions in a decreasing lexicographic order. For instance, assuming d 1 þ d 2 ¼ 15 with 0 r d 1 r 10 and 0 r d 2 r 12, the first solution is (10,5) and the other ones are subsequently (9,6), (8,7), (7,8), (6,9), (5,10), (4,11), and (3,12). The most benefit of using this approach to solve the system is that we obtain all the system solutions in a decreasing lexicographic order.…”

“…Fig. 2 shows an almost large benchmark ARPANET example [31,32] with M ¼ (13,12,14,8,12,8,11,12,9,7,14,12,8,13,12,6,13,15,14,10,12,6,7,10,11,14,13,12,14,9,15,15), L¼ (3,8,7,7,4,7,6,2,7,6,3,5,7,8,3,6,9,8,7,…”

“…Nowadays, network reliability theory has extensively been applied to a variety of real-world systems such as transportation [2], mobile ad hoc wireless [3,4], power transmission and distribution [5,6], grid computing [7], manufacturing [8][9][10], and computer and communication [9,11]. Moreover, in some optimization problems such as maximizing system reliability [6,8] or optimal design of a network subject to reliability constraint [12], there is an increasingly significant need for efficiently computing or estimating the system reliability.…”

Deriving Job Completion Reliability and Job Energy Consumption for a General MapReduce Infrastructure from Single-Job Perspective

“…Levitin et al [7] estimated the task execution time and reliability of a multi-processing-unit hardware system. However, the two systems were not organized as a master-slave infrastructure, and the redundant schemes they utilized were not cold-standby schemes.…”

Optimal Transmission Line Assignment Considering Reliabilities and Assignment Cost in the Multi-State Network

“…As network reliability plays an important role in ensuring the normal output of data, many researches focus on reliability assurance or reliability improvement. For a given network, in order to guarantee the network reliability, lots of studies focus on the network backup paths or the spare paths [19]- [24], the network tolerance [25], [26], the network resilience [27], [28], etc.…”