HPX - High Performance ParalleX



The STE||AR Group (pronounced as stellar) stands for "Systems Technology, Emergent Parallelism, and Algorithm Research". We are an international group of faculty, researchers, and students working at different organizations. The goal of the STE||AR Group is to promote the development of scalable parallel applications by providing a community for ideas, a framework for collaboration, and a platform for communicating these concepts to the broader community.

All of our work is centered around building technologies for scalable parallel applications. HPX, our general purpose C++ runtime system for parallel and distributed applications, is no exeption. We use HPX for a broad range of scientific applications, helping scientists and developers to write code which scales better and shows better performance compared to more conventional programming models such as MPI.

HPX is based on ParalleX which is a new (and still experimental) parallel execution model aiming to overcome the limitations imposed by the current hardware and the way we write applications today. Our group focuses on two types of applications - those requiring excellent strong scaling, allowing for a dramatic reduction of execution time for fixed workloads and those needing highest level of sustained performance through massive parallelism. These applications are presently unable (through conventional practices) to effectively exploit a relatively small number of cores in a multi-core system. By extention, these application will not be able to exploit high-end computing systems which are likely to employ hundreds of millions of such cores by the end of this decade.

Critical bottlenecks to the effective use of new generation high performance computing (HPC) systems include:

The ParalleX model has been devised to address these challenges by enabling a new computing dynamic through the application of message-driven computation in a global address space context with lightweight synchronization. The work on HPX is centered around implementing the concepts as defined by the ParalleX model. HPX is currently targetted at conventional machines, such as classical Linux based Beowulf clusters and SMP nodes.

We fully understand that the success of HPX (and ParalleX) is very much the result of the work of many people. To see a list of who is contributing see our tables below.

HPX Contributors

Table 33. Contributors

Contributors to this Document

Table 34. Documentation Authors


Thanks also to the following people who contributed directly or indirectly to the project through discussions, pull requests, documentation patches, etc.

In addition to the people who worked directly with HPX development we would like to acknowledge the NSF, DoE, DARPA, Center for Computation and Technology (CCT), and Department of Computer Science 3 - Computer Architecture who fund and support our work. We would also like to thank the following organizations for granting us allocatons of thier compute resources: LSU HPC, LONI, XSEDE and the Gauss Center for Supercomputing.

HPX is currently funded by the following grants: