基于区分业务优先级和提高系统时延性能的网络需求,提出了依托站点状态的两级轮询控制系统.系统在混合服务两级轮询模型的基础上,根据站点缓冲区状态采用并行调度方式仅对有数据分组的活动站点提供服务.该模型既能满足区分站点优先级的需求又能避免空闲查询,从而提高系统利用率、降低等待时延.采用嵌入式马尔科夫链和概率母函数的方法对该系统建立数学模型,对系统平均等待时延特性进行了精确解析.通过理论计算与仿真实验结果的对比验证了理论分析的正确性,与已有两级轮询系统相比,具有更好的时延性能.
Based on priority differentiation and the system efficiency, this paper proposes a station dependent two-level polling system. The mixed service two-level polling system is extended by using queue state-dependent routing, in which only active stations with information packets could be visited by server. The scheme meets the requirement not only for conflict free but also for priority differentiation and efficiency, and provides a lower latency. An embedded Markov chain framework is set up to drive the closed-form expression for the mean waiting time. Numerical examples demonstrate that theoretical and simulation results are identical and the new system has a better efficiency at both key station and normal station.