Intelligent fitting global real‐time task scheduling strategy for high‐performance multi‐core systems

With the development of high-performance computing, it is possible to solve large-scale computing problems. However, irregularity and access characteristics problems bring challenges realisation performance optimisation. Improving a single core makes challenging maintain Moore's law, multi-core processors emerge. A chip brings together multiple universal processor cores equal status has same structure supported by an isomorphic processor. In granularity tasks leads complexity scheduling strategies. Satisfying high system performance, load balancing fault tolerance at minimum cost key task in field, especially specific hardware architecture. this study, global real-time implemented system. The adopts hybrid among clusters intelligent fitting within implement strategy. cluster policy, are allowed preempt with low priority, priority that memory dynamically improved, higher than all without access. An method also proposed. When data read cache ability value reasonable threshold, promoted highest preempting task. results show intelligently strategy for systems better nuclear utilisation rate schedulability.