Traditional large-scale enterprise cloud system has the problem of large delay. Therefore, a delay tolerant data congestion avoidance algorithm based on modular computing is designed. The resource allocation rules of rbfr strategy in traditional enterprise cloud are improved to change the traditional interactive overall mode and make it modular.Using modular thinking to construct the data congestion control algorithm, mainly in the control layer to achieve the model, through the introduction of ant colony algorithm in modularization, the design of delay tolerant data congestion avoidance algorithm is realized. In order to verify the algorithm’s congestion avoidance effect, comparative experiments are designed. Experimental results show that the packet loss rate of large enterprise cloud system under this algorithm is lower than that of traditional delay tolerant data congestion avoidance algorithm.

传统的大型企业云系统存在延迟大的问题。为此，设计了一种基于模块化计算的延迟容忍数据拥塞避免算法。改进了传统企业云中rbfr策略的资源分配规则，改变了传统的交互整体模式，使其模块化。 采用模块化思维构建数据拥塞控制算法，主要在控制层实现模型，通过引入蚁群算法在模块化的基础上，实现了延迟容忍数据拥塞避免算法的设计。为了验证算法的拥塞避免效果，设计了对比实验。