Multiple timing faults, although detectable individually, can hide each other’s faulty behavior making the faulty system indistinguishable from a non-faulty one. A set of graph augmentations are introduced for single timing faults. The fault detection capability of the augmentations is analyzed in the presence of multiple timing faults and shown that multiple occurrences of a class of timing fa...

A connected hypercube with faulty links and/or nodes is called an injured hypercube. To enable any non-faulty node to communicate with any other non-faulty node in an injured hypcrcube, the information on component failures has to be made available to non-faulty nodes so as to route messages around the faulty components. We propose first a distributed adaptive fault-tolerant routing scheme for ...

A fault-free cycle in the n-dimensional hypercube Qn with f faulty vertices is long if it has length at least 2 − 2f . If all faulty vertices are from the same bipartite class of Qn, such length is the best possible. We prove a conjecture of Castañeda and Gotchev [2] asserting that fn = ( n 2 ) −2 where fn is the largest integer such that for every set of at most fn faulty vertices, there exist...

Nodes in distributed systems can become faulty due to multiple reasons: software or hardware bug, node misconfiguration or node being compromised by an attacker. Often such faulty nodes can be difficult to detect and localize. An approach to solve this problem is to hold each node accountable for its actions i.e. each node is required to provide verifiable evidence for its actions and detect a ...