Transparent parallelism accelerates time to solution for difficult optimization problems

Comet™ features high-level abstractions for parallel and distributed computing. By reducing the distance between sequential and parallel programs for combinatorial optimization, Comet gives you the ability to leverage commodity multi-processors or clusters of multi-processors. Abstractions such as parallel loops, software interruptions, events, and thread pools provide a natural mapping between the Comet program and its underlying architecture. These high-level abstractions are implemented using source-to-source transformations, and use language concepts such as threads, processes, and synchronization primitives.

Comet's Abstraction Layer Creates Benefits of Parallelism

• Transparent parallelism, creating no additional work or development effort
• Minimizes the distance between sequential and parallel programs
• Implements a multi-start local search controller and hybrid evolutionary algorithms sequentially
• The ability to leverage cloud computing to speed time to solution