Load balancing method and device of database system based on high-performance cluster distribution

Abstract

The invention discloses a load balancing method and a device of a database system based on a high-performance cluster distribution. The method comprises: sorting priority of database operations in a database system; allocating corresponding operation time for database operations of different priorities; creating a table file to generate a task queue and an expired data processing task queue; in the data acquisition cycle, executing the data storage operation, that is, splitting a plurality of sub-table files according to the number of data nodes and the size of the table files, and storing thesub-table files on a plurality of data nodes; executing the table file generation operation in the table file generation cycle; at that expire data processing cycle, performing the expire data processing operation. The invention adjusts the distribution of the table on the database node by taking the data table as an operation unit, monitors the load distribution of the data, and properly makes load balancing adjustment, thereby effectively balancing the load of the database and preventing excessive load of the database.

Description

technical field [0001] The invention relates to the field of cluster servers, in particular to a load balancing method and device for a database system based on high-performance cluster distribution. Background technique [0002] In the current performance data collection system, these operations involving the database are not well planned in terms of execution timing. The storage of performance data is done after the performance data collection is completed, and periodic report generation with larger granularity and expired data are stored. Deletion is scheduled through the timer. Since the relevance and timeliness of these database operations are not considered, this scheduling method itself may be invalid. For example, when a 30-minute report needs to be generated, the data of the first 30 minutes is not completely saved. After completion, the correctness of the generated data is affected; at the same time, such scheduling may cause different database operations to superi...

AI Technical Summary

Problems solved by technology

[0002] In the current performance data collection system, these operations involving the database are not well planned in terms of execution timing. The storage of performance data is done after the performance data collection is completed, and periodic report generation with larger granularity and expired data are stored. Deletion is scheduled through the timer. Since the relevance and timeliness of these database operations are not considered, this scheduling method itself may be invalid. For example, when a 30-minute report needs to be generated, the data of the first 30 minutes is not completely saved. Complete, affecting the correctness of the generated data; at the same time, such scheduling may cause different database operations to overlap at the same time, resulting in excessive database load

In a relational database, the business logic of a distributed transaction often needs to be processed by multiple systems. In the end, it is necessary to ensure the consistency of transaction processing by multiple systems. It is also required to give priority to supporting open source database systems, but at the same time there are difficulties The problem of realizing distributed storage and transportation

[0003] In addition, in the prior art, the selection of the acquisition cycle often does not consider the time-consuming data acquisition and storage, which will also lead to the superposition of related database operations, causing the database load to be too large, making the database load a bottleneck of the system and affecting to the usability of the system itself, and the problem of invalid operation due to failure to consider timeliness

Images