On-board Hybrid Heterogeneous Distributed Computing Resource Virtualization

It is necessary to develop a real-time fault-tolerant high-performance computing architecture to enhance the efficiency and performance of on-board applications, the architecture would provide a shared, unified, and scalable computing platform utilizing various hybrid heterogeneous resources such as CPU, DSP, FPGA, and GPU. In this paper, we propose an abstract method for modeling the energy consumption of heterogeneous computing resources and establish performance models for each resource. By enabling dynamic allocation and elastic expansion of on-board hybrid heterogeneous distributed computers based on user needs, we aim to achieve maximum computing energy efficiency and service quality.