Method for building distributed memory disk cluster storage system

Abstract

The present invention relates to an operation method of distributed memory disk cluster storage system, the distributed memory storage system is adopted thereby satisfying four desired improvements including the expansion of network bandwidth, the expansion of hard disk capacity, the increasing of TOPS speed, and the increasing of memory I/O transmitting speed. Meanwhile, the system can be cross-region, cross-datacenter and cross-WAN operated, so the user's requirements can be collected through the local memory disk cluster for providing the corresponding services, the capacity of the memory disk cluster can also be gradually expanded for further providing cross-region or cross-country data service.

What is claimed is: 
     
         1 . A method for building a distributed memory disk cluster storage system, wherein the distributed memory disk cluster storage system comprises a plurality of computers, and each of the plurality of computers comprises a processor, a memory, and a network card, the method comprising:
 installing a virtual machine platform into the plurality of computers so that each of the plurality of computers has a plurality of virtual machines;   planning the memory of each of the plurality of computers as a plurality of chunk memory disks by means of the virtual machine platform;   integrating the plurality of computers to form a cluster memory disk; and   dividing each of a plurality of files of the plurality of chunk memory disks of the plurality of computers of the cluster memory disk into a plurality of data, copying the plurality of data to generate a plurality of copied data, and evenly distributing the plurality of copied data to all the plurality of chunk memory disks of the plurality of computers of the cluster memory disk,   wherein all the plurality of chunk memory disks of the plurality of computers of the cluster memory disk are connected so as to be jointly operated.   
     
     
         2 . The method as claimed in  claim 1 , wherein the step of planning the memory of each of the plurality of computers as the plurality of chunk memory disks by means of the virtual machine platform comprises:
 each of the plurality of computers sets a memory capacity assignment for the memory through a virtual machine operating system of the virtual machine platform and a program for setting the memory capacity assignment to plan the memory of each of the plurality of computers as the plurality of chunk memory disks.   
     
     
         3 . The method as claimed in  claim 1 , wherein the plurality of computers are integrated to form the cluster memory disk by means of a network layer interface virtualized by a virtual machine operating system of the virtual machine platform, a memory bus with multiple channels and the network card in each of the plurality of computers. 
     
     
         4 . The method as claimed in  claim 2 , wherein when the virtual machine operating system of the virtual machine platform accesses a file from a memory location of the virtual machine operating system, the file is computed by one of the plurality of virtual machines to generate computed data, and the computed data is stored back in the memory location, so that the plurality of data in each of the plurality of computers is processed in the plurality of chunk memory disks by parallel computing. 
     
     
         5 . The method as claimed in  claim 4 , wherein other data which is not stored in the plurality of computers is accessed by the plurality of chunk memory disks of other computers through the network card to connect to a connection port cluster link. 
     
     
         6 . The method as claimed in  claim 1 , wherein the plurality of computers are categorized to a first data center, at least one second data center and a backup center,
 wherein the first data center is controlled by a virtual cluster data control station, and each of at least one second data center and the backup center are controlled by a virtual cluster data backup station, wherein the virtual cluster data backup station is a backup of virtual cluster data control station,   wherein the first data center and at least one second data center jointly form a distributed memory file system of the distributed memory disk cluster storage system.   
     
     
         7 . The method as claimed in  claim 1 , wherein each of the plurality of chunk memory disks is electrically connected to at least one hard disk storage device, and the at least one hard disk storage device is configured to backup data in each of the plurality of chunk memory disks once every preset period of time. 
     
     
         8 . The method as claimed in  claim 1 , wherein the plurality of chunk memory disks of the plurality of computers use a continuous data protector for continuously backing up the data to a hard disk cluster array. 
     
     
         9 . The method as claimed in  claim 1 , wherein each of the plurality of computers further comprises a hard disk, a mother board, an I/O interface card, a connection cable and a housing. 
     
     
         10 . The method as claimed in  claim 1 , wherein each of the plurality of copied data is encrypted through mixing a plurality of encryption algorithm and stored in the memory. 
     
     
         11 . The method as claimed in  claim 1 , wherein an operation status of each of the plurality of chunk memory disks is monitored by a search engine, and the search engine restarts once an error is detected. 
     
     
         12 . The method as claimed in  claim 2 , wherein an assigned memory of each of the plurality of virtual machines is operated through a storage area network, the storage area network adopts a network layer interface virtualized by a virtual machine operating system of the virtual machine platform to connect all the chunk memory disks. 
     
     
         13 . The method as claimed in  claim 1 , wherein the virtual machine platform plans the processor, the memory and a hard disk as a resource pool, and transfers each of unused computer resources to other computers. 
     
     
         14 . The method as claimed in  claim 1 , wherein a maximum capacity of the cluster memory disk is larger than 52.5 TB(terabyte). 
     
     
         15 . The method as claimed in  claim 1 , wherein all the plurality of chunk memory disks of the plurality of computers access data synchronously. 
     
     
         16 . A method for building a distributed memory disk cluster storage system, wherein the distributed memory disk cluster storage system with comprises a plurality of computers, and each of the plurality of computers comprises a processor, a memory, and a network card, the method comprising:
 installing a virtual machine platform into the plurality of computers so that each of the plurality of computers has a plurality of virtual machines;   planning the memory of each of the plurality of computers as a plurality of chunk memory disks by means of the virtual machine platform;   planning all the chunk memory disks of each of the plurality of computers as a resource pool of each of a plurality of cluster memory disks respectively by means of a network layer interface virtualized by a virtual machine operating system of the virtual machine platform, a memory bus with multiple channels, and the network card in each of the plurality of computers, wherein all the plurality of chunk memory disks of the plurality of computers access data synchronously;   simulating the plurality of cluster memory disks as a common resource pool,   dividing each of a plurality of files of the plurality of chunk memory disks of the plurality of computers of plurality of cluster memory disks into a plurality of data, copying the plurality of data to generate a plurality of copied data, and evenly distributing the plurality of copied data to all the plurality of chunk memory disks of the plurality of computers of the plurality of cluster memory disks of the common resource pool,   wherein all the plurality of chunk memory disks of the plurality of computers of the cluster memory disk of the common resource pool are connected so as to be jointly operated.   
     
     
         17 . The method as claimed in  claim 16 , wherein the plurality of computers are categorized to a first data center, at least one second data center and a backup center,
 wherein the first data center is controlled by a virtual cluster data control station, and each of at least one second data center and the backup center are controlled by a virtual cluster data backup station, wherein the virtual cluster data backup station is a backup of virtual cluster data control station,   wherein the first data center and at least one second data center jointly form a distributed memory file system of the distributed memory disk cluster storage system.   
     
     
         18 . The method as claimed in  claim 17 , wherein the common resource pool of cluster memory disks is controlled by the virtual cluster data control station and the virtual cluster data backup station to evenly distribute amounts of processing data to all the chunk memory disks of the plurality of computers. 
     
     
         19 . The method as claimed in  claim 16 , wherein the plurality of cluster memory disks are connected with each other according to a physical internet transmission protocol, and the plurality of data is transmitted through an encryption algorithm;
 when one of network connections among the plurality of cluster memory disks is unable to be established, each of the plurality of cluster memory disks is independently operated; and   when one of the network connection is recovered, data is synchronized to each of the chunk memory disks of each of the plurality of cluster memory disks.   
     
     
         20 . The method as claimed in  claim 16 , wherein a maximum capacity of the common resource pool is larger than 16 ZB(zettabyte). 
     
     
         21 . The method as claimed in  claim 16 , wherein all the plurality of chunk memory disks of the plurality of computers access data synchronously.