Automatically increasing the fault-tolerance of distributed systems
Gil Neiger, Sam Toueg · 1988
The design of fault-tolerant distributed systems is a costly and diflicult task.Its cost and difficulty increase dramatically with the severity of failures that a system must tolerate.We seek to simplify this task by developing methods to automatically translate protocols tolerant of "benign" failures to ones tolerant of more "severe" failures.This paper describes two new translation mechanisms for qr~hronous systems; one translates programs tolerant of crash failures into programs tolerant of general omission failures, and the other translates from gene& omiesion failures to arbitrary failures.Together these can be used to translate any program tolerant of the most benign failures to a program tolerant of the most severe.