Accelogic demonstrates the world’s first reconfigurable computing algorithms to achieve greater than 1,000x speedups for numerical global optimization software.

The world’s first prototype of a massively parallel reconfigurable computing algorithm demonstrating greater than 1,000x speedups for global optimization has been successfully implemented and benchmarked today. Preliminary results using the Pintér benchmarking functions indicate that, when implemented in a commercial reconfigurable computing system, the new algorithm outperforms (in both speed and accuracy) a 1,000 processor supercomputer programmed with a state-of-the-art parallel branch-and-bound algorithm. Accelogic’s solution is also two orders of magnitude less expensive.

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *