Multi-Class Queueing Networks with Blocking

阻塞网络在性能研究中经常出现 计算机系统、分布式系统和通信系统。 尽管它们具有潜在的适用性，但使用并不广泛。 这主要是由于缺乏全面的分析工具， 这样的系统。 本研究旨在纠正这种情况， 开发用于分析多类神经网络的算法 用blocking。 到目前为止，这样的多类网络还没有 在公开文献中被研究过。 为了实现这一目标， 先分析一下。 这些配置将采用合并的形式， 分裂和串联多类阻塞网络。 的 从这些分析中获得的结果和经验 然后将整合配置，以获得解决方案 任意配置的技术。 最后，这些解决方案 技术将被纳入现有的算法， 加强网络，以进一步提高其效用。

Queueing networks with blocking frequently arise in performance studies of computer systems, distributed systems, and communication systems. Despite their potential applicability, their use is not widespread. This is mainly due to the absence of comprehensive tools for analyzing such systems. This research aims at rectifying this situation by developing algorithms for the analysis of multi-class queueing networks with blocking. So far, such multi-class queueing networks have not been studied in the open literature. In order to achieve this objective, small configurations will be analyzed first. These configurations will be in the form of merge, split, and tandem multi-class queueing networks with blocking. The results and experience obtained from the analysis of these configurations will then be consolidated in order to obtain solution techniques for arbitrary configurations. Finally, these solution techniques will be incorporated into existing algorithms for general queueing networks, in order to further increase their usefulness.