Multi-agent grid resource dynamic scheduling and collaborative optimization method based on multi-objective optimization
By performing fine-grained data quantization and performance gap correction on multi-entity computing network resource scheduling, and combining bandwidth rates and latency estimation, an adaptive scheduling trigger signal is generated. This solves the problems of ineffective reconstruction and gain overestimation caused by high-frequency disturbances in multi-entity computing network environments, and achieves stable resource allocation and collaborative deployment.
CN122412093APending Publication Date: 2026-07-17STATE GRID HENAN INFORMATION & TELECOMM CO
View PDF 0 Cites 0 Cited by
Patent Information
- Application Number
- CN202610546692.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-23
- Publication Date
- 2026-07-17
Smart Images
Figure CN122412093A_ABST
Abstract
本申请涉及算网资源调度技术领域，具体公开了一种基于多目标优化的多主体算网资源动态调度与协同优化方法，其通过对细粒度业务请求与异构节点探针数据进行约束量化与特征聚合，形成任务约束表达与全局算网状态感知基础，在此基础上同步量化迁移重构代价与预期性能增益，并引入基于性能差距比率与最优迁移路径代价乘性耦合的指数型衰减因子对增益进行非线性折减修正，随后将修正后的增益与重构代价纳入博弈框架并施加动量平滑与滞后阈值判决，仅在净收益持续超越势垒时触发多目标协同求解，经帕累托前沿寻优与置信度加权择优输出最优分配方案，最终通过异构原语翻译与三段式无损时序完成跨域协同部署。
Need to check novelty before this filing date? Find Prior Art