A cluster load balancing scheduling method and system

Abstract

The present invention provides a cluster load balancing scheduling method and system. The method includes the following steps: obtaining the IP address information carried in the service request and processing the IP address information according to a preset processing strategy to obtain the binary data corresponding to the IP address. digit information; according to the preset initial level, obtain the current hash mask and calculate the binary digits corresponding to the current hash mask and use the binary digits as the folding length; according to the obtained folding length, the corresponding IP address The binary digits are folded to obtain the folding result; the folding result is subtracted from the current mask, and according to the operation result, the corresponding target server is determined according to the preset server selection strategy. The invention can greatly reduce the calculation amount of the hash operation in the cluster load balancing dispatching algorithm of the waterfall hash, thereby improving the efficiency.

Description

technical field [0001] The invention belongs to the field of load balancing, and in particular relates to a cluster load balancing scheduling method and system. Background technique [0002] Various load balancing algorithms are often used in cluster computing to solve the scalability and transparency problems of the system, that is, the load balancing scheduler efficiently distributes requests to different servers for execution, making a cluster system composed of multiple independent computers Becoming a virtual server, when the client application interacts with the cluster system, it is like interacting with a high-performance server. [0003] At present, common load balancing scheduling algorithms include round-robin scheduling, weighted round-robin scheduling, minimum connection scheduling, weighted minimum connection scheduling, locality-based least connection, locality-based least connection with replication, target address hash scheduling, source address Hash schedu...

AI Technical Summary

Problems solved by technology

[0004] However, in the cluster load balancing scheduling algorithm of waterfall hashing, the mask of the current level is hashed according to the source IP address multiplied by a large prime number. There are two problems with this algorithm. When it is less than the number of digits of the product of the IP address and the large prime number, the hash operation only uses the bits of the product of the IP address and the large prime number, which will cause high conflicts in the hash result, which needs to be avoided by adding layers. For example, when the mask is 100, the IP The last two digits of the product of the address and the large prime number determine the result of the hash. If there are two IP addresses, and the last two digits of the product of the IP address and the large prime number are the same, the same hash value is obtained. At this time, the waterfall level needs to be increased. to avoid conflicts, so that the efficiency is reduced; the second problem is that the multiplication operation of the IP address and a large prime number belongs to the multiplication operation of two large numbers, and the efficiency is low

Images