You are here

Cloud Computing; High-Performance Computing


Cloud computing is characterized by the allocation of internet-based distributed computing resources on an as-needed basis across a shared platform. Multiple computing cores may be dedicated in parallel to achieve the required computing performance levels for a specific task. Similarly, high performance computing usually relies on the use of a large number of co-located or distributed cores running in parallel. To achieve the highest levels of performance, massively parallel supercomputers may employ many thousands of cores.
This subtopic focuses on innovations that result in substantial improvements to cloud computing or high-performance computing platforms. These improvements may be in terms of computing power, computing efficiency, energy management, network storage requirements, the use of hybrid clouds, latency, data integrity and availability, cost, or any other factor of importance in such platforms, and may result from software- or hardware-based innovations. Included in this subtopic are innovations that will enable continued improvement in the performance of HPC platforms in the so-called post- Moore’s Law era.
Examples of applications that typically require the levels of computing power available through cloud computing or high-performance computing include (but are not limited to): stock market analysis and prediction; cryptanalysis; weather forecasting; fluid dynamic modelling, acoustic modelling and other computationally intensive engineering modelling; advanced audio and video signal processing.

US Flag An Official Website of the United States Government