High-End Computing University Research Activity

High-performance computing is increasingly essential to progress in science and engineering. Contemporary high-end computing (HEC) systems often comprising of tens- to hundreds-of-thousands of processors allow researchers to study complex problems that were previously intractable. However, emerging data-intensive scientific challenges and opportunities demand more of HEC systems. For example, observation- and simulation-driven applications require higher throughput input/output (I/O) capabilities, large data storage capacities, and tools for efficiently finding, processing, organizing and moving data. Data-management challenges also include the need to access large volumes of data produced by different applications, in numerous locations, and in various formats.Although storage capacity and processing power are growing rapidly, increases in data bandwidth and access times are not keeping pace. In fact, the advent of multicore processors has resulted in a decrease in memory and bandwidth per core. The performance gap between HEC processing power and storage device performance demands advances in massively parallel I/O systems to maintain the throughput of applications. The ability to efficiently map I/O operations between millions of distributed memories and hundreds-of-thousands of storage devices is also a formidable problem that calls for research.Accordingly, in 2009, the High-End Computing University Research Activity (HECURA) program invites research and education proposals in the areas of I/O, file and storage systems design for efficient, high-throughput data storage, retrieval and management in cases where HEC systems comprise hundreds-of-thousands to millions of processors. Research areas of interest include, but are not limited to:I/O architectures and I/O middleware; archives/backups as extensions to file systems;file systems research and file systems-related protocols;metadata research;access methods; data management systems; security; novel storage devices for the I/O stack; Quality of Service;management, and reliability and availability at scale (RAS);hardware and software tools for design and simulation of I/O, file and storage systems; andefficient benchmarking, tracing, performance measurement and tuning tools of I/O, file and storage systems.