Storage Class Memory (SCM) is the term used to describe a new class of solid-state, nonvolatile memory technologies. There are several (≥10) such technologies currently under active research and development that are vying to become the SCM of choice. As a group, their performance is nearer memory than storage while their cost is nearer storage than memory. In the systems designer's eye, these performance and cost characteristics blur the historic distinction between memory and storage. Exploiting SCM will require significant changes in the design of memory and storage systems. These changes at the system level will force or enable changes at the application level as well as opening up new application opportunities.

Experts in advanced system technologies will predict the design of the best HPC Storage Systems in 2020. They will defend why they think the technology they select will be the winning technology 15 years from now. The panelists will pick one set of technology - not a list of possibilities - to define the system. They will define the performance and aspects of the technologies and explain why their system is the most likely to succeed.Besides questions and comments at the sessions, attendees will vote for the proposed systems of 2020 they think are most likely to succeed. The presentations, votes and attendee comments will all be sealed in a time capsule that will be opened in 2020, which will be used in 2020 to compare the predications to reality. The time capsule will also include an appropriate prize for the presenter who made the best prediction. The winner needs to be present to collect their prize.

A model for analyzing the reliability of gracefully degrading and standby-sparing computer systems is developed. The basis of the model is the identification of four distinct causes of crashes: time-domain multiple faults, resource exhaustion, space-domain multiple faults, and solitary faults. Expressions are developed for each of these crash-causing mechanisms and for their interrelationship. The model is then demonstrated by using it to analyze the reliability of a gracefully degrading computer system called PRIME.