SYNERGISTIC COMPUTING ON HETEROGENEOUS MULTIPROCESSOR

With the emerging demand for computations on mobile devices, heterogeneous multi-processors are dominating the mobile computing landscape. Heterogeneous multi-processors usually comprise of various components, including CPUs with different performance-power characteristics and application-specific accelerators (GPUs, DSPs, FPGAs, NPUs, etc.). The presented heterogeneity enables delicate matching of computational kernels on to the processors that are best suited to perform the computation. This leads to substantial improvements in performance and energy-efficiency to enable next generation mobile computing. While architectural heterogeneity is promising, software development efforts are required to fully benefit from this architectural advancement. The goal of this dissertation is to embrace the heterogeneity by synergistic computing on multiple components in different scenarios to unleash the full potential of the heterogeneous multiprocessors towards high-performance energy-efficient mobile computing.

Author dc:creator

• WANG SIQI

• heterogeneity, mobile systems, system-on-chips, GPGPU, thermal management, machine learning