Abstract: The problem of online processing node fault detection in mesh-connected multicore and many-core VLSI multiprocessors is considered. A novel hardware-level approach to the multiprocessor test based on mutual inter-processor checking is presented which presupposes that a coordinated healthy/faulty decision is made for each processor core by applying the majority operator to the individual healthy/faulty tags calculated by the corresponding set of testing neighbors. Formal rules are defined for forming sets of testing and tested neighbors for each processor node of the mesh which are invariant to the location of the node within the mesh and to its dimension. The formulae to determine the number of testing neighbors for each node depending on the dimension of the mesh are given. A parallel hardware-level algorithm implementing the proposed test method is presented and its possible hardware implementation is discussed. The successful fault detection probability is evaluated in the case when the proposed approach is used, its dependencies on the individual test node reliability are investigated. The proposed approach is shown to provide increased successful fault detection probability compared to the traditional self-checking and neighbor-checking for all practically significant cases.
Jamil Al-Azzeh, Evgeny A. Titenko and Igor V. Zotov, 2017. A Distributed Majority-Operator-Based Built-In Mutual Inter-Node Test Method for Mesh-Connected VLSI Multiprocessors. Journal of Engineering and Applied Sciences, 12: 7112-7118.