Load balance and node state monitoring method in high performance computing

Abstract

The invention relates to a load balance and node state monitoring method in high performance computing. The method is suitable for reducing resources which occupy a clustering system, increasing the utilization rate of the resources, effectively improving the performance of server clustering and providing high-quality service for a user. The method concretely comprises the steps that firstly, according to the operation state and load and performance parameters of server nodes, computing is performed to obtain load weights of all the nodes, and an alternative node set for next task distribution is selected; afterwards, according to load difference values and distribution probabilities, the probabilities of all node distribution tasks in the alternative node set are computed, and new requests are distributed to the selected nodes in a random probability distribution mode to solve the problem of uneven load distribution; finally, based on a load correction formula, node loads of the distribution tasks are corrected to improve the load balance effect and improve the reliability and stability of the cluster system.

Description

technical field [0001] The invention relates to the field of computer network load balance processing, in particular to a load balance and node state monitoring method in high-performance computing. Background technique [0002] With the increasing popularity of network applications and the rapid growth of Internet services, whether in enterprise networks, campus networks or WANs, the development of business volumes has exceeded past estimates. Enterprises are increasingly dependent on the network, and the application requirements for distributed systems with scalability and reliability are also increasing. When an enterprise provides web services to users, with the rapid increase in the number of visitors, the web server needs to be able to provide a large number of concurrent access services. Its data traffic and calculation intensity are so large that a single device cannot bear it at all; on the other hand, how to achieve a reasonable traffic distribution among multiple...

AI Technical Summary

Problems solved by technology

[0011] Specifically: according to the running status of the server node and its load and performance parameters, the load weight of each node is obtained through calculation, and the set of nodes to be selected for the next allocation of tasks is selected; the candidate nodes are calculated according to the load difference and distribution probability The probability of assigning tasks to each node in the set, and use the random probability assignment method to assign new requests to the selected nodes to solve the problem of uneven load distribution; finally, use the load correction formula to adjust the load of the assigned tasks. Amended to improve the load balancing effect and enhance the reliability and stability of the cluster system

Images