HPX
Scientific computing program
From Wikipedia, the free encyclopedia
HPX, short for High Performance ParalleX, is a runtime system for high-performance computing. It is currently under active development by the STE||AR group[2] at Louisiana State University. Focused on scientific computing, it provides an alternative execution model to conventional approaches such as MPI. HPX aims to overcome the challenges MPI faces with increasing large supercomputers by using asynchronous communication between nodes and lightweight control objects instead of global barriers, allowing application developers to exploit fine-grained parallelism.[3][4][5]
LSU Center for Computation and Technology
| HPX | |
|---|---|
| Developers | The STEllAR Group, archived from the original on 2019-04-03, retrieved 2019-04-03 LSU Center for Computation and Technology |
| Initial release | 2008 |
| Stable release | 1.10.0
/ May 29, 2024 |
| Written in | C++ |
| Operating system | Microsoft Windows Linux Mac OS X |
| Type | Partitioned global address space Parallel programming Runtime System |
| License | Boost Software License[1] |
| Website | hpx |
| Repository | github |
HPX is developed in idiomatic C++ and released as open source under the Boost Software License, which allows usage in commercial applications.
Applications
Though designed as a general-purpose environment for high-performance computing, HPX has primarily been used in
- Astrophysics simulation, including the N-body problem,[6] neutron star evolution,[7] and the merging of stars[8]
- LibGeoDecomp,[11][12][13] A Library for Geometric Decomposition codes
- Simulation crack and fractures utilizing Peridynamics[14]
- Phylanx,[15][16][17][5] A Library for Distributed Array Processing