Collaborative Research: System Software for Scalable Applications

点击翻译按钮获取中文摘要

This award facilitates scientific research using the large new computational resource named Blue Waters being developed by IBM and scheduled to be deployed at the University of Illinois. As hardware complexity increases in HPC systems, it is difficult for applications to take full advantage of the available system resources and avoid potential bottlenecks. The purpose of this project is to improve the performance and productivity of key system software components on petascale platforms such as Blue Waters. The researchers from University of Chicago, University of Illinois and Argonne National Laboratores propose to study four different classes of system software: message passing libraries, parallel I/O, data visualization, and operating system. They will use a computational chemistry application (NWChem) and a climate modeling application (CCSM) on Blue Waters to study and validate the performance and scalability of their system software components. Through rigorous experimentation, analysis, and design cycles, the researchers will dramatically improve the capabilities of not only systems being deployed in the near term, but of all systems pushing scalability limits in the near future.The benefits of this research will be available to all applications on a variety of large petascale systems in the form of revised system software. The results will be disseminated to system vendors as well as to other researchers and students in the form of tutorials, workshop and conference presentations, and seminars.