Availability of commercial parallel systems

Abstract

The objective of this panel is to discuss the advanced techniques commercial systems have implemented to achieve high availability and outlining issues on which the research community should be focused to enhance availability in commercial systems. Panelists from some of the commercial parallel processing companies have been invited to impart their experience and expertise in designing and implementing highly available parallel systems. Scalability has been addressed by many commercial parallel processing systems. The general question is how to address capacity growth with linear scalability and at the same time provide continuous availability for customers' mission-critical applications. Other issues that need to be addressed include how to handle system management, workload balancing, ease of parallel processing and providing a computing environment with near-continuous availability.