ABSTRACT
High Performance Computing (HPC) is used to solve a number of complex questions in computational and data-intensive sciences. By any measure, the past 15 years have witnessed dramatic increases in both the use and scale of HPC. The most prominent HPC benchmark is the TOP500 Linpack benchmark. TOP500 has qualities that make it valuable and successful: easily scaled problem size, straightforward validation, an open source implementation, almost 20 years of historical data, a large audience familiar with the software, and well managed guidelines and submission procedures. Aside from standard benchmarks, another indicator of growth in HPC is their performance on real scientific problems. In 2007, the Green500 benchmark and list were created in order to recognize the growing importance of energy efficiency in HPC. Although HPC systems share many hardware components with servers in enterprise and data centers, the HPC software stack is dramatically different from an enterprise or cloud software stack and is unique to HPC.
