IMPROVING PERFORMANCE IN HPC SYSTEM UNDER POWER CONSUMPTIONS LIMITATIONS
Muhammad Usman Ashraf · International Journal of Advanced Research in Computer Science · 2019
Today's High-Performance Computing (HPC) systems require significant usage of "supercomputers" and extensive parallel processing approaches for solving complicated computational tasks at the Petascale level of performance (10 15 calculations per second). The next breakthrough in the computing revolution is the Exascale level of performance that is 10 18 calculations per second-a remarkable achievement in computing that will have a fathomless influence towards everyday life. Current supercomputers can't achieve such a high level of performance under power dissipation constraints. Even though the Exascale performance can be achieved by multiplying the number of cores according to Exascale computing system constraints, the challenge of power consumption still persists. However, the primary focus of this study is to analyse how to enhance performance under power consumption limitations for emerging technologies. Leading to objectives, the current study presents a comprehensive analysis of existing strategies that can be considered to enhance performance and reducing power for emerging Exascale computing system. Consequently, we have suggested a massive parallel programming mechanism which is promising to achieve HPC Exascale system goals.