September 2009 HPC User Forum Meeting Notes: HPC Application Scaling Issues

Abstract

This IDC update documents the 33rd High-Performance Computing (HPC) User Forum — which took place at the Omni Interlocken Resort Hotel in Broomfield, Colorado, on September 8–10, 2009 — in near-verbatim form.

The principal tasks of the meeting were to:

• Share information among HPC users, vendors, and IDC for improving the health of the HPC industry
• Continue showcasing instructive examples of HPC leadership and partnerships in government, industry, and academia
• Explore leading-edge trends, developments, and issues in the earth sciences, especially weather and climate
• Discuss the status quo and directions in front-end compiler technology and application scalability
• Deliver IDC' s latest HPC market update and forecast data

Table of Contents

• In This Update
• Technical Panel on HPC Application Scaling Issues, Requirements, and Trends
• Moderators
• Panelists
• Question 1: Are There More Appropriate Domain-Specific Performance Metrics for HPC Applications Available than "Percent of Peak" or "Parallel Efficiency and Scalability"?
• Question 2: Similar to HPC Disruptive Technologies Thought to Be Needed in Many Hardware Road Maps to Reach Exascale, Are There Looming Challenges Whose Resolution Will Be Game Changing Over the Next Decade?
• Question 3: What Is the Role of Local Node-Based Floating Point Accelerators in the Next Three to Five Years?
• Question 4: Should Applications Continue with Current Programming Models?
• Learn More
• Related Research