Method, device, server and server cluster for realizing asynchronous call

Abstract

The embodiment of the present invention provides an asynchronous call implementation method, device, server and server cluster, wherein the asynchronous call implementation method is applied to the server cluster, the server cluster includes multiple servers, and the multiple servers can all access the preset shared database; There are a main process and a sub-thread running in the server. The method includes: the main process judges whether there is a target identification field in the shared database based on the resource operation request; wherein, the resource operation request is used to request a specified operation for a specified resource; The field that conflicts with the specified operation of the resource; if there is no target identification field, the main process generates at least one remote procedure call protocol RPC request based on the resource operation request, and adds the RPC request to the local memory queue; the child thread starts from the local memory queue Get the RPC request and send it. The embodiment of the present invention improves the efficiency of RPC calling.

Description

technical field [0001] The invention relates to the field of communication technology, in particular to a method, device, server and server cluster for realizing asynchronous calling. Background technique [0002] In a large-scale distributed system, there are often many nodes, and various RPC (Remote Procedure Call Protocol, remote procedure call protocol) calls are made between these nodes. In particular, for centralized service nodes, it is necessary to call a large number of remote services through RPC to achieve specific functions. For example, the relationship between neutron-server (sub-server) and agent (agent): neutron-server exposes rest API (Application Programming Interface, application programming interface), so that third-party services call rest API, and neutron-server sends RPC requests to the associated In neutron-agent, neutron-agent implements the functions corresponding to third-party services. [0003] Specifically, the process of the computing service...

AI Technical Summary

Problems solved by technology

[0004] However, the inventors found that by using the for loop on the server side to make RPC calls serially, the response time of the API is linearly related to the size of the cluster to a certain extent. For a large-scale distributed system cluster, this serial RPC call The method seriously affects the response time of the API. For example, if there are 100 RPC requests, the time spent on each RPC request is 0.01s, and the time spent on the RPC request is 1s.

Images