Optimization of load balancing algorithm based on Informer long time series prediction

Since the ribbon load balancing algorithm is not dynamic enough to respond to load fluctuations in real time, this paper proposes a load balancing algorithm based on informer long time series prediction, which can better adapt to the dynamic load environment. By predicting the number of CPU cores, network utilization, memory utilization, and I/O utilization in the future, you can adjust the server weight in advance to achieve a more balanced load distribution and improve the resource utilization and response speed of the system. Jmeter is used to conduct a test experiment, and the Ribbon load balancing algorithm is compared with the algorithm proposed in this paper. The experimental results show that the algorithm used in this paper can be better applied to load balancing in microservice system under the condition of large concurrency.