Method and device for constructing mpp public cloud and local private cloud

Abstract

The invention discloses a method and device for constructing an MPP public cloud and a local private cloud. The method includes the following steps: planning server resources, and writing them into a background database table; receiving a work order application, and obtaining role information of an application object; The role information checks whether the application object has the application authority; if the application object has the application authority, the corresponding GP cluster or PG database will be automatically built according to the work order application. This solution is based on Greenplum (referred to as GP), and does not need to specify IP, server resources and other information to deploy GP, and the deployment process is simple; for the application of PG database, there is no need to specify IP, port and other information, and the PG database can be automatically applied for, and at the same time through The multi-authority level tenant system builds public cloud databases and private clouds at the same time, which facilitates multi-level database management.

Description

technical field [0001] The invention relates to the field of server resource deployment, in particular to a method and device for constructing an MPP public cloud and a local private cloud. Background technique [0002] MPP is a massively parallel processing database cluster, which is built on the basis of GP. Simply put, GP is an object-oriented relational database like Oracle, DB2, and PG. In essence, it is a relational database cluster, a logical database composed of several independent database services. GP adopts the Shared-Nothing architecture. The entire cluster consists of multiple data nodes and control nodes. Each data node can run multiple databases independently of each other. The biggest feature of GP is that it can still support large-scale high-performance data warehouses and business intelligence queries without high-end hardware support. In the application of data warehouse and business intelligence, it can efficiently handle the process of I / O data throug...

AI Technical Summary

Problems solved by technology

The disadvantage of this solution is: the application of GP is complicated, and manual deployment through IP, port, user, user password and other information requires DBA to have strong experience in network, system, database, etc.; PG application is also relatively complicated and requires DBA Assistance again; in the whole process, the communication is through email, which is inefficient and the error rate of information is greatly increased; the lack of systematic monitoring is not conducive to database error troubleshooting and performance optimization

The disadvantages of this solution are: the entire process is communicated by email, which is inefficient, and the error rate of information is greatly increased; every time it is needed, the DBA still needs to log in to the server for remote batch operations; the monitoring is a messy script, which is not easy to view, and the log Still need DBA to extract

Images