The High Performance Software Foundation (HPSF) is pleased to welcome HPX as a new established project. HPX brings a modern approach to parallel and distributed computing in C++, expanding the foundation’s portfolio of technologies advancing scalable, high performance systems.
HPX, short for High Performance ParalleX, is a general purpose C++ runtime system designed for parallel and distributed applications of any scale. It provides a unified programming model that enables developers to efficiently utilize available computing resources, from single nodes to large scale distributed systems.
Originally developed within the research community, HPX is the first open source implementation of the ParalleX execution model, an approach focused on overcoming key barriers to scalability such as latency, overhead, and resource contention.
A Foundation for Scalable C++ Applications
HPX extends the C++ standard library for concurrency and parallelism, enabling developers to write asynchronous, fine grained parallel applications using familiar C++ constructs. It builds on established standards while introducing capabilities that support both shared memory and distributed environments.
This approach allows developers to move beyond traditional parallel programming models by embracing techniques such as message driven computation and adaptive resource management. The result is improved scalability and performance across a wide range of use cases, including scientific computing and large scale simulations.
HPX has been applied in areas such as astrophysics and other compute intensive domains, demonstrating its ability to handle complex workloads efficiently while maintaining developer productivity.
Strengthening the HPSF Ecosystem
By joining HPSF, HPX becomes part of a growing ecosystem focused on advancing open technologies for high performance computing. Its emphasis on standards-aligned C++ development and scalable execution models complements existing HPSF projects and strengthens the foundation’s ability to support next generation workloads.
HPX’s inclusion reflects the continued importance of open collaboration in addressing the challenges of modern computing, particularly as systems grow more complex and distributed.
“Becoming a member of HPSF is a badge of honor—it means you’re accepted by the community and that your project brings worldwide benefit,” said Hartmut Kaiser, a professor of computer science and engineering with a joint appointment in the LSU Center for Computation & Technology, who leads the LSU team. “This leads to visibility and credibility, but also, increased interest and support from industry, and potential funding for LSU students who will have more ability to find jobs by gaining experience and being part of a recognized HPSF project.”
Get Involved
As an open source project, HPX welcomes contributions from developers, researchers, and organizations interested in shaping the future of parallel computing. Whether contributing code, sharing use cases, or participating in the community, there are many ways to get involved.