Our need for more calculation has never been satiated by devices to aid in computing. Every new breakthrough in hardware seems only to inspire more demanding uses for computers. Often this progress hardware is incremental, however, in the history of large-scale simulation, there have been several disruptive changes brought about by changes in hardware, typically driven by forces outside of scientific computing. Whether incremental or dramatic, these changes necessitate changes in algorithms to realize the performance benefits of the new machines. A disruptive transition is currently underway, and we will discuss some of the anticipated changes in both supercomputers and the algorithms deployed on them.
Back to Science at Extreme Scales: Where Big Data Meets Large-Scale Computing Tutorials