16 November 2011
ET International has introduced its SWARM (SWift Adaptive Runtime Machine) technology, which is providing a commercial implementation consistent with the ParalleX execution model.
This open parallel execution model is designed to eliminate the primary constraints of conventional programming methods and more flexibly manage application algorithm parallelism to enable the development of scalable applications for many-core systems.
As HPC more fully integrates the use of hybrid and multicore architectures, development of parallel applications that can effectively scale for optimal execution on many-core systems has presented a daunting challenge. ParalleX represents a paradigm shift from traditional message-passing computing to a message-driven model that utilises dynamic scheduling and resource management instead of the static scheduling conventional of today’s programming models. This open execution model for parallel computation is designed to enable developers to exploit the full processing power of many-core systems with a high degree of parallelism through fine-grain, decoupled transaction processing with asynchronous event-driven execution.
In contrast to the standard execution model of communicating sequential processes, this design serves to expose multiple forms of parallelism, thus hiding latency and reducing overall overhead. Swarm is bringing ParalleX to conventional system platforms to help solve the massively parallel, data-intensive problems faced by scientific and industrial communities.