Towards the support for many-task computing on many-core computing platforms

Abstract—Current software and hardware limitations prevent ManyTask Computing (MTC) from leveraging hardware acceler-ators (NVIDIA GPUs, Intel MIC) boasting Many-Core Comput-ing architectures. Some broad application classes that t the MTC paradigm are work ows, MapReduce, highthroughput computing, and a subset of high-performance computing. MTC emphasizes using many computing resources over short periods of time to accomplish many computational tasks (i.e. including both dependent and independent tasks), where the primary metrics are measured in seconds. MTC has already proved successful in Grid Computing and Supercomputing on MIMD architectures, but SIMD architectures of today’s accelerators pose many challenges in the efficient support of MTC workloads on accelerators. This work aims to address the programmability gap between MTC and accelerators, through an innovative middleware that will enable MIMD workloads to run on a SIMD architecture. This work will enable a broader class of applications to leverage the growing number of accelerated high-end computing systems