Transcript number dynamic adjustment method based on file heat and node load

Abstract

The invention discloses a transcript number dynamic adjustment method based on file heat and node load. The method comprises the following steps: 1) setting reliability parameters according to file reliability, and initializing the transcript number of the file; 2) performing statistics on the access times, and calculating the file heat hoti; 3) calculating an average access heat havg of the file,and dividing files in a set F into hotspot files, ordinary files and cold files; 4) calculating the average node Lavg of nodes, and dividing the nodes into heavy load nodes, intermediate load nodes and light load nodes; and 5) traversing the files in the set F I, calculating a theoretical transcript number Mi of the file according to the file heat and the node load, and dynamically adjusting thetranscript number rm of the file. According to the method, the current access amount and the previous file heat are counted, the influence of the file change rate to the heat is considered, the transcript number is dynamically adjusted via the average access heat of the files and the average load of the nodes in the cluster, thereby reducing the average response time and improving the overall performance of the cluster.

Description

technical field [0001] The invention relates to the technical field of computer cloud storage, in particular to a method for dynamically adjusting the number of copies based on file heat and node load. Background technique [0002] Cloud storage systems generally use distributed file systems to store and manage data. Copy technology is a traditional data management technology. In order to avoid data loss and improve data security and availability, most cloud storage systems use the strategy of creating copies to improve system reliability and performance. [0003] Generally speaking, there are two main types of replica creation technologies for cloud storage systems: static replica creation and dynamic replica creation. The number and location of the copies in the static copy creation method are determined in advance, that is, before the file is created in the system, the number and location of the copies to be created are first determined, and the number of copies will not...

AI Technical Summary

Problems solved by technology

[0004] Although relevant scholars have done a lot of research on replica creation and adjustment, these methods are not very applicable in the actual hybrid cloud environment

First of all, in terms of calculating file heat, it is only a simple count of the number of accesses within a period of time. Even considering the use of multiple frequencies to predict the access characteristics of files in the next cycle, it is difficult to take into account file heat and node load balancing. In a hybrid cloud environment In the large number of nodes, the performance of each node has obvious differences

When a lot of hot data gathers in some storage nodes with poor performance, it will cause hot spots and reduce the overall performance of the system

Images