Asynchronous communication method and device

Abstract

The invention discloses an asynchronous communication method, and the method comprises the steps: receiving a task message sent by a client; processing a resource request for the task message, and obtaining a part of obtained results; generating identification information according to the task message, and enabling the part of obtained results and the identification information to be transmitted to the client; continuously processing unfinished resource request, obtaining the part of obtained results, and temporarily storing the part of obtained results; and transmitting the temporarily stored result to the client when the identification information transmitted by the client is received. According to the invention, through the stepped processing of the resource request for the task message, the method processes the resource requests of all parts through the time difference between two requests, obtains a part of results, and transmits the part of results, divides a complex task with longer time consumption into simple tasks with shorter time consumption, respectively completes the simple tasks through employing occupied threads, effectively reduces the processing pressure, can avoid the long-time waiting of the client, and especially can avoid request failure caused by no response under the high parallel requests.

Description

technical field [0001] The present invention relates to the technical field of communication, in particular to an asynchronous communication method and device. Background technique [0002] In the most basic way of information interaction in the existing technology, the client initiates a resource request to the server purposefully, and is in a state of waiting for the result information. After receiving the resource request from the client, the server searches for the corresponding resource according to the content of the request , get the result after certain calculation and processing, and respond to the client. During the entire interaction process, the client needs to be in a synchronous waiting state all the time. Once the server-side processing overload causes the processing capacity to decline, the waiting time of the client will be extended, and the entire request process will fail due to long-term no response. The client needs to re-initiate the request Executing ...

AI Technical Summary

Problems solved by technology

This solution improves the efficiency of responding to client requests through caching and transparent retry mechanisms, and relieves the processing pressure on the back-end server. However, in essence, it is still a synchronous request / response mode. When there are too many requests from the front-end client At the same time, due to the limited cache capacity of the proxy service, most requests will still penetrate to the back-end server. Once the amount of calculation required for the requested resource increases, it will also bring huge pressure to the server and slow down the response. The continuous backlog of processing requests not only causes the client to hang up for a long time, but even causes the proxy service to go down, without fundamentally solving the problem

Images