Service flow cutting method and device based on time sequence model

Abstract

The invention provides a service flow cutting method and device based on a time sequence model, which can be used in the technical field of artificial intelligence, and the method comprises the steps: receiving a flow cutting request transmitted by an operation and maintenance personnel terminal; obtaining server cluster information for processing a service, obtaining index data corresponding to a flow cutting index in a server cluster through the server cluster information according to the preset flow cutting index, and predicting the index data through a preset time sequence model to obtain a flow cutting index change trend; according to the flow switching control method and device, predicting the flow switching index change of the server cluster for service processing through the preset time sequence model to obtain the flow switching index change trend, so that the cost of performance testing is reduced; and the accuracy of predicting the change trend of the system flow switching index and the stability of server cluster service processing are improved.

Description

technical field [0001] The present invention relates to the field of information system operation and maintenance technology, in particular to the field of financial technology, and in particular to a method and device for switching business traffic based on a time series model. Background technique [0002] The size of business traffic is closely related to the running status of the server cluster for business processing. If the business traffic is too large and exceeds the load that the business operating environment of the server cluster can carry, business failures are prone to occur, resulting in business failure. At present, in production, the business flow is mainly limited by setting the business flow threshold. The threshold can obtain the maximum transaction volume that the transaction environment can carry through the transaction pressure test, and then set a ratio of 90% on the basis of the maximum transaction volume as the flow cut Trigger conditions, or through...

AI Technical Summary

Problems solved by technology

This method of predicting the cut-off value through production pressure testing needs to be operated artificially, and in order to prevent changes in the trading environment, or the demand optimization of the transaction itself leads to changes in the cut-off value, frequent pressure testing is often required , so this method is time-consuming and laborious, and the efficiency is relatively low

However, the usual big data and artificial intelligence algorithms cannot automatically adjust the cut-off value as the trading environment changes (server aging, performance degradation, and transaction optimization), nor can it predict the concurrent changes of non-periodic sudden transactions. Flow adjustment methods and strategies are relatively simple

Images