Data Intensive Architecture. (DIVA)

As part of the DARPA Information Technology Office (ITO) Data Intensive Systems Architecture (DIVA) project, we are developing algorithmic benchmarks. The benchmarks include algorithms for solving linear and non-linear optimization problems, database search tasks, terrain masking problems that arise in a wide variety of data processing and fusion problems. The benchmarks are being used by the University of Southern California's Information Sciences Institute (ISI) to evaluate design alternatives for a parallel computing system built around so-called processor in-memory chips that reduce the bandwidth limitations induced by the physically separated processor and memory hardware typically found in computers today.