Time delay control method, system and related equipment

Abstract

The invention provides a time delay control method, a time delay control system and related equipment. The method comprises the following steps: a time delay control system receives a message sent by a producer, determines a time period to which the message generation time belongs according to the generation time of the message so as to determine a writable file group, and writes the message into the file group; obtaining a fluctuation index according to the message generation time, the message arrival time and the identifier of the message source, wherein the fluctuation index is used for describing the time delay fluctuation amplitude of the message source; and updating the state of the file group according to the fluctuation index, and outputting the file group in the consumable state to a consumer. According to the method, the process that a consumer caches the message in the processing process can be omitted, the occupancy rate of a memory and CPU computing resources is reduced, and meanwhile the response speed is increased.

Description

technical field [0001] The present application relates to the technical field of message buses, and in particular, to a time delay control method, system and related equipment. Background technique [0002] With the commercialization of the fourth and fifth generation communication technologies, the network throughput rate continues to rise, the complexity of the network structure increases significantly, and the data volume increases exponentially. More and more, network planning and optimization are faced with great challenges. For example, in the 5G Non-Standalone (NSA) scenario, network optimization planning often requires the processing and analysis of messages from different sources generated at the same time. [0003] In order to meet the above business scenarios, a message system with high throughput and high reliability is required. The existing messaging system Apache Kafka is a high-throughput distributed publish-subscribe messaging system. Producers provide mes...

AI Technical Summary

Problems solved by technology

The existing message system Apache Kafka is a high-throughput distributed publish-subscribe message system. Producers provide messages to Kafka in order, and consumers also take out messages in order and distribute them to their corresponding task threads, etc. The message in the task thread is processed when it is complete, which causes the task thread to always cache some low-latency messages and spend a long time waiting for high-latency messages, thus continuously occupying a large amount of random access memory (Random AccessMemory, RAM) resources (that is, continuously occupying a large amount of memory resources) and central processing unit (Central Processing Unit, CPU) resources greatly increase the complexity of processing logic, but also reduce the efficiency of message processing

Images