A method for self-adaptive adjustment of rpc timeout value between partitions

Abstract

The invention discloses a method for self-adaptive adjustment of the RPC timeout value between partitions. The timeout value range is divided into a conventional increase range and a decreaseable range; when the RPC timeout value is in the normal increase range, if a timeout occurs, the doubling method is adopted Increase the timeout value; after the RPC timeout value enters the range that can be reduced, the client communicates with the server to adjust the RPC timeout value through mutual negotiation. If it is busy, adjust the RPC timeout value to 3 times the current value, otherwise adjust to 2 times the current value; if the server detects that its CPU is idle, it will actively notify the client to lower the RPC timeout value. The present invention adjusts the timeout value between partitions, fully considers the situation when the server CPU is busy or idle to adjust the timeout value, and provides an adjustment method for reducing the timeout value at the same time, so that the timeout value adjustment has higher accuracy, adjustment speed and automatic adjustment. adaptability.

Description

technical field [0001] The invention relates to the technical field of network communication, in particular to a method for self-adaptive adjustment of timeout value of RPC (Remote Procedure Call, Remote Procedure Call) between partitions. Background technique [0002] In the process of communication between the client and the server of the computer network, if the client does not receive a response from the server within a certain period of time, the system calls this situation "timeout". The processing method is called "timeout mechanism". Timeout is one of the common phenomena in network services. Due to the complexity of the network, the timeout mechanism is widely used in network services. [0003] RPC protocol is an important communication technology in computer network. In RPC application, timeout mechanism is an indispensable means, which can be used to detect the failure of the system. In the RPC system, how to set a reasonable timeout value is one of the most imp...

AI Technical Summary

Problems solved by technology

When the client resubmits the task (or submits a new task), the server may still execute the task that was abandoned by the client last time, and the resubmitted task (or submitted new task) cannot be executed immediately

This leads to resubmitted tasks (or new submitted tasks) waiting in queue on the server side. When more and more tasks are waiting, they will accumulate on the server side, and the waiting time of tasks is also limited. When the task exceeds the set After the specified timeout value, the task may be resubmitted, which will further increase the probability of timeout phenomenon

Images