基于云计算的航空飞行试验数据中心任务调度优化架构设计  

Optimization Architecture Design of Mission Scheduling for Aviation Flight Test Data Center Based on Cloud Computing

作　　者：许辰宏 于刘 XU Chenhong;YU Liu(Shanghai Aviation Industrial(Group)Co.,Ltd.,Shanghai 201206,China)

机构地区：[1]上海航空工业(集团)有限公司,上海201206

出　　处：《计算机测量与控制》2024年第2期168-173,共6页Computer Measurement &Control

摘　　要：为解决航空飞行试验数据中心任务调度行为明显滞后的问题,实现对航空飞行试验数据的实时调度,设计基于云计算的航空飞行试验数据中心任务调度优化架构;设置WiRo中心网络,联合试验数据预测器与飞行任务分配器,完善中心任务调度优化架构体系的基础应用结构设计;根据PSO优化度量值的取值范围,求解惯性权重指标与粒子编码条件,并按照云计算法则,推导函数表达式条件,实现基于云计算的航空飞行试验数据调度模型的构建;在动态数据权限的约束下,计算中心调度任务的资源占用率与长尾延迟参数,实现对任务调度架构的优化配置,联合WiRo中心网络及EMU调度结构,完成基于云计算的航空飞行试验数据中心任务调度优化架构的设计;实验结果表明,云计算技术作用下,长尾延迟最大值为1.11%,单位时间内的数据吞吐量达到了9.85 B/s,由数据吞吐量有限造成的中心任务调度行为滞后的问题得到较好解决,符合实时调度航空飞行试验数据的实际应用需求。To solve the significant lag in the task scheduling behavior of aviation flight test data centers and achieve the real-time scheduling of aviation flight test data,an optimization architecture based on cloud computing for the aviation flight test data center task scheduling is designed.Set up a WiRo central network,unite the test data predictors and flight task allocators,and improve the basic application structure design of the central task scheduling optimization architecture system.Based on the range of particle swarm optimization(PSO)optimization metrics,the inertia weight index and particle encoding conditions are solved,and the function expression conditions are derived according to cloud computing rules,the construction of an aviation flight test data scheduling model based on cloud computing is achieved.Under the constraints of dynamic data permissions,the resource occupancy rate and long tail delay parameters of the scheduling tasks in the computing center are calculated to optimize the configuration of the task scheduling architecture.The WiRo central network is Combined with the EMU scheduling structure to complete the cloud based optimization architecture design for the task scheduling of the aviation flight test data center.The experimental results show that under the influence of cloud computing technology,the maximum long tail delay is 1.11%,and the data throughput per unit time reaches 9.85 B/s.The delayed scheduling behavior of central tasks caused by the limited data throughput is well solved,which meets the practical application requirements for the real-time scheduling of the aviation flight test data.

关 键 词：云计算 航空飞行 试验数据 中心任务 调度架构 资源占用率 动态权限 延迟参数 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]