A High-Performance Memory-Mapped Data Multi-Level Allocation Approach for Real-Time Performance Adaptation
Qi Guo, Tianyu Guo, Haiping Guo, Yuanhong Lu, Jie Zhang, Libin Huang, Jingyue Zhang, Hui Luo · 2024
Conventional high-performance memory mapping data multi-layer allocation methods mainly use SOM (Self organizing maps) improved self-organizing mapping algorithm to extract data allocation characteristics, which is vulnerable to the impact of cluster topology, resulting in low data allocation utilization. Therefore, a high-performance memory mapping data multi-layer allocation method is designed considering real-time performance adaptation. That is, a dynamic scheduling and allocation model for high-performance memory mapping data is constructed, and a multi-layer allocation algorithm for high-performance memory mapping data is designed considering real-time performance adaption, thus completing the multi-layer allocation of high-performance memory mapping data. The experimental results show that the designed real-time performance adaptive high-performance memory mapping data multi-layer allocation method has a high data allocation utilization rate, good allocation effect, meets the real-time requirements of data allocation, and makes some contributions to the realization of dynamic data resource allocation.