An intelligent task dynamic scheduling system and method

Abstract

The invention relates to an intelligent scheduling system. The intelligent scheduling system comprises a load data acquisition module used for periodically collecting and reporting index data according to a specific time interval through an acquisition terminal program; a unified data management module which is used for storing load data and statistical data and providing a query interface; an intelligent algorithm module which is used for calculating service available values of all servers distributed in different clusters, and classifying and screening out a corresponding optimal IP list; And an IP distribution module which is used for selecting the most matched IP list and returning the most matched IP list to the client. The invention further relates to a method for achieving intelligent dynamic scheduling of the tasks. By adopting the system and the method, the high-availability requirement of a production environment is met. In the aspect of high concurrency, high concurrency processing capacity is achieved, and meanwhile the cache module and the IP distribution module both support horizontal expansion and are very flexible. In the aspect of effectiveness, the real load information of the server can be more comprehensively and accurately acquired, and more intelligent and more effective load balancing service can be provided for an access end.

Description

technical field [0001] The invention relates to the field of computer application technology, in particular to the field of intelligent dynamic load balancing, in particular to an intelligent task dynamic scheduling system and a method thereof. Background technique [0002] Load balancing is a front-end basic component, that is, the load balancer evenly distributes the traffic to multiple servers in the cluster according to a certain load balancing algorithm, so as to achieve the effective load sharing of the entire cluster and prevent the load caused by uneven load. Various problems to ensure the stability and high availability of the cluster. [0003] From the perspective of application scenarios, common load balancing models include global load balancing and intra-cluster load balancing. From the perspective of product form, it can be divided into hardware load balancing and software load balancing. Global load balancing is generally implemented through DNS. By resolving...

AI Technical Summary

Problems solved by technology

[0004] Most of the load balancing software uses optional load algorithms for task scheduling, which has some shortcomings. Generally, the running load and business load of the server are not considered comprehensively, and the real-time load status of the server cannot be accurately reflected.

In addition, load balancing software generally only focuses on the load balancing of the cluster, without considering factors such as the network environment and geographical location of the client, resulting in poor access performance of the optimal machine obtained by the scheduling system for the client, due to the lack of a feedback mechanism. The scheduling system cannot perceive this problem, and it will eventually affect the use of services

Images