Implementing Efficient Fault Containment for Multiprocessors

This article describes our experience with implementing efficient fault containment in a prototype operatingsystem, called Hive [3]. Hive is a version of UNIX that is targeted to run on the Stanford FLASH multiprocessor(Figure 1). Novel algorithms in Hive work together with unique hardware features provided by FLASH to limitthe impact of both hardware faults and operating system bugs. Hive is based on Silicon Graphics IRIX 5.2 (aversion of UNIX SVR4) and preserves binary compatibility with...