Toward a Scalable Heterogeneous Runtime System for the Convey MX Architecture

Given the recent advent of the multicore era [1], research efforts in the area of high performance, low latency runtime systems have increased significantly....