Virtual symmetric multiprocessing virtual machine creation method, data processing method and system

Abstract

The invention provides a virtual symmetrical multi-processing virtual machine establishing method, a data processing method and a system. The virtual symmetrical multi-processing virtual machine establishing method includes the steps that multiple computer nodes are connected through an infinite bandwidth network, and the computer nodes are in communication connection, wherein the computer nodes comprise a main unit and multiple sub-units; operating systems of the computer nodes are integrated into a virtual operating system controlled by the main unit; hardware of the computer nodes is integrated into a hardware resource pool, and databases of the computer nodes are integrated into a database pool; the main unit directly manages and calls the hardware resource pool and the database pool through the virtual operating system. According to the virtual symmetrical multi-processing virtual machine establishing method, the data processing method and the system, data processing needing a large quantity of internal storage and computing resources can be achieved, and the advantages of being high in computing speed, simple in structure, low in cost, convenient to manage, high in flexibility, high in resource utilization rate and high in expandability are achieved.

Description

technical field [0001] The invention relates to the field of computer virtual machines, in particular to a virtual symmetric multiprocessing virtual machine creation method and a data processing method. Background technique [0002] People have never stopped pursuing the application of big data databases. The application of modern big data databases basically refers to the establishment of databases on the original calculator or server hard disk and the operation of index programs. The data volume is generally more than 100,000. Therefore, we generally do not call a system that runs less than 100,000 records or does not apply a big data index program a big data database, but only a database with a data volume greater than 100,000 records and runs an index program as a big data database system . The components of the big data database system are mainly composed of computing processors, hard disks, memory, operating systems and database software. In many database applications...

AI Technical Summary

Problems solved by technology

Obviously, to meet the I / O requirements of ultra-large-scale data calculation, search, and indexing, if traditional hard disks are used as carriers, it is difficult to achieve high-expansion, high-speed, and low-latency big data by relying solely on hard disk stack transmission. Database system design purpose

And in the current development direction of virtualization technology, research is directed towards the direction of distributed application virtualization, such as traditional VMware, Zen-desktop, Hyper-V or KVM and today's container and Docker technologies, the main focus is The resources required by the application are packaged, so that multiple application calculations can be performed on a single computer at the same time, and the calculation amount of the single computer can be maximized. While increasing the burden of I / O, when performing calculations that require a huge amount of memory, the single computer It also cannot provide enough memory for computing exchange, which affects the speed of data processing

Images