Remote procedure call server control system and method

Abstract

The present invention provides a remote procedure call server control system and method, the server is remotely called by the client, and the server control system includes: a task controller, a resource manager, a resource pool and a distributed collaborative service module , wherein: the task controller is configured to receive a service creation request from a user; the resource manager is configured to receive a service creation request from the task controller, and according to the service parameters in the service creation request, create And start several computing nodes that meet the service parameters; the resource pool is used to allocate computing resources of several computing nodes that meet the service parameters; the distributed collaborative service module is used to learn the running status, and share the running status of each computing node with the client. The invention can improve the scalability and fault tolerance of the remote procedure call service.

Description

technical field [0001] The invention relates to the technical field of the Internet, in particular to a remote procedure call server control system and method. Background technique [0002] Remote Procedure Call Protocol (RPC) is a way to request services from remote computer programs over the network without knowing the underlying network technology. The service based on remote procedure call is the cornerstone of various modern distributed systems, and its stability, reliability and efficiency are directly related to the availability and economy of the entire system. Therefore, major Internet companies generally develop their own remote procedure call systems. [0003] Usually, the server side of the remote procedure call system is deployed and installed on pre-determined static resources. For example, the number of machines (physical or virtual) is determined, the computing capabilities of the machines are determined, and the addresses of the machines are also determine...

AI Technical Summary

Problems solved by technology

For example, if there is a problem with the service provider's machine, the service user cannot automatically eliminate the problematic machine, which will cause related calls to fail to complete successfully

For another example, if the topology of the service provider's machine cluster changes, including but not limited to adding machines, deleting machines, changing machine addresses, etc., the service user cannot automatically perceive

In addition, changes in resource patterns cannot be predicted in a timely manner, for example, it may be too late to implement contingency measures when the service load increases

Images