CN105446794A

CN105446794A - Disc operation method, apparatus and system based on virtual machine

Info

Publication number: CN105446794A
Application number: CN201410522778.9A
Authority: CN
Inventors: 张灿群
Original assignee: Beijing Kingsoft Cloud Network Technology Co Ltd; Beijing Kingsoft Cloud Technology Co Ltd
Current assignee: Beijing Kingsoft Cloud Network Technology Co Ltd; Beijing Kingsoft Cloud Technology Co Ltd
Priority date: 2014-09-30
Filing date: 2014-09-30
Publication date: 2016-03-30

Abstract

Embodiments of the present invention disclose a disc operation method, apparatus and system based on a virtual machine, relate to the technical field of a computer and are applied to a client server. The method comprises: acquiring metadata information corresponding to a virtual disk from a metadata server; saving the metadata information in the client server; accessing the virtual disk in the client server, and acquiring first metadata information corresponding to the to-be-accessed virtual disk from one or more pieces of metadata information stored in the client server; and according to the acquired first metadata information, sending a file operation instruction to a storage server, so that the storage server executes a file operation corresponding to the file operation instruction. When the scheme provided by the embodiments of the present invention is applied to carry out the file operation, the file operation delay can be reduced and the user experience can be improved.

Description

A kind of disk operating method, Apparatus and system based on virtual machine

Technical field

The present invention relates to field of computer technology, particularly a kind of disk operating method, Apparatus and system based on virtual machine.

Background technology

Distributed file system refers to that the physical memory resources of file system management is not necessarily connected directly between on local memory device, but can be connected with the memory device of network side by computer network.Distributed file system is generally the pattern based on client end/server end.

Common distributed file system is made up of client-server, meta data server, storage server etc. usually, client-server, is connected by network between meta data server and storage server.Traditional distributed file system adopts meta data server to carry out the metadata information of stored file in each storage server of unified management.

In addition, virtual machine technique is widely applied in recent years, and is applied in distributed file system.In prior art, based on the operation of the disk of virtual machine, can be understood as based on the file operation in distributed file system, when virtual machine technique is applied to distributed file system, one or more virtual machine can be disposed in client-server, and virtual machine can one or more virtual disk corresponding, each virtual disk can be understood as a file.When user by the equipment of oneself log on virtual machine end carry out file operation time, need first to send metadata request information to meta data server, therefrom obtain virtual disk metadata information corresponding in storage server, and then send file operation instruction according to obtained metadata information to storage server, after storage server receives file operation instruction, carry out corresponding file operation according to this file operational order.Application said method, user can carry out file operation at virtual machine end to the file be stored in storage server, but, when carrying out file operation at every turn, all need first to the corresponding metadata information of meta data server request, due to factors such as Internet Transmissions, file operation is existed and postpones, affect Consumer's Experience.

Summary of the invention

The embodiment of the invention discloses a kind of disk operating method, Apparatus and system based on virtual machine, to reduce the delay carrying out file operation at virtual machine end, improve Consumer's Experience.

For achieving the above object, the embodiment of the invention discloses a kind of disk operating method based on virtual machine, be applied to client-server, described method comprises:

Metadata information corresponding to virtual disk is obtained from meta data server; By described metadata information, be stored in described client-server;

When accessing the virtual disk in described client-server, from one or more metadata informations that described client-server stores, obtain the first metadata information that the virtual disk that will access is corresponding;

According to the first obtained metadata information, send file operation instruction to storage server, make described storage server perform file operation corresponding to described file operation instruction.

Optionally, describedly obtain metadata information corresponding to virtual disk from meta data server, comprising:

When creating virtual disk, obtain metadata information corresponding to created virtual disk from meta data server.

Optionally, the described disk operating method based on virtual machine also comprises:

The described storage server receiving the transmission of described storage server performs the object information of file operation corresponding to described file operation instruction.

Optionally, after the described storage server of described reception described storage server transmission performs the object information of file operation corresponding to described file operation instruction, also comprise:

The object information that described storage server performs file operation corresponding to described file operation instruction is sent, to make described meta data server according to this information spanned file Operation Log information to described meta data server.

Optionally, in described client-server, comprise at least one virtual machine, at least one virtual disk can be created in described virtual machine;

Described by described metadata information, be stored in described client-server, comprise:

By described metadata information, be stored in virtual disk corresponding to described metadata information, wherein, described virtual disk is arranged in the virtual machine that described client-server is installed.

When nullifying the virtual disk in client-server, the second metadata information that the virtual disk that nullify is corresponding is together deleted.

Optionally, when described file operation instruction is operating writing-file instruction, described according to the first obtained metadata information, send file operation instruction to storage server, make described storage server perform file operation corresponding to described file operation instruction, comprising:

The memory address corresponding in storage server according to the virtual disk that will access comprised in the first obtained metadata information, sends written document request to described storage server;

Send the file data of described file to be written to described storage server, make described storage server perform the file operation writing described file to be written.

Optionally, the described file data sending described file to be written to described storage server, makes described storage server perform the file operation writing described file to be written, comprising:

According to the Divide File rule preset, by the file data of described file to be written, be divided into multiple data block;

Call at least one process, send to described storage server and divide file data corresponding to the data block that obtains, described storage server is performed and writes the file operation of described file to be written, wherein, file data corresponding to one or more data block is sent by a process.

Optionally, the described Divide File rule according to presetting, by the file data of described file to be written, is divided into multiple data block, comprises:

Obtain the quantity being in the process of idle condition;

According to the obtained quantity being in the process of idle condition, by the file data of described file to be written, be divided into multiple data block.

Optionally, described in call at least one process, send to described storage server and divide file data corresponding to the data block that obtains, make described storage server perform the file operation writing described file to be written, comprising:

Obtain the quantity dividing the data block obtained;

Call several processes to send to described storage server and divide file data corresponding to the data block that obtains, described storage server is made to perform the file operation writing described file to be written, wherein, the quantity of institute's calling process is identical with the quantity dividing the data block obtained.

Optionally, when described file operation instruction is for reading file operation instruction, described according to the first obtained metadata information, send file operation instruction to storage server, make described storage server perform file operation corresponding to described file operation instruction, comprising:

The memory address corresponding in storage server according to the virtual disk that will access comprised in the first obtained metadata information, send to described storage server and read file operation instruction, make described storage server perform described in read file operation instruction corresponding read file operation.

For achieving the above object, the embodiment of the invention discloses a kind of disk operating device based on virtual machine, be applied to client-server, described device comprises:

First metadata information obtains module, for obtaining metadata information corresponding to virtual disk from meta data server;

Metadata information memory module, for by described metadata information, is stored in described client-server;

Second metadata information obtains module, during for accessing the virtual disk in described client-server, from one or more metadata informations that described client-server stores, obtains the first metadata information that the virtual disk that will access is corresponding;

File operation instruction sending module, for according to the first obtained metadata information, sends file operation instruction to storage server, makes described storage server perform file operation corresponding to described file operation instruction.

Optionally, described first metadata information obtains module, specifically for when creating virtual disk, obtains metadata information corresponding to created virtual disk from meta data server.

Optionally, the described disk operating device based on virtual machine also comprises:

Object information receiver module, the described storage server sent for receiving described storage server performs the object information of file operation corresponding to described file operation instruction.

Object information sending module, for after described object information receiver module receives described object information, the object information that described storage server performs file operation corresponding to described file operation instruction is sent, to make described meta data server according to this information spanned file Operation Log information to described meta data server.

Metadata information memory module, specifically for by described metadata information, be stored in virtual disk corresponding to described metadata information, wherein, described virtual disk is arranged in the virtual machine that described client-server is installed.

Metadata information removing module, for when nullifying the virtual disk in client-server, deletes the second metadata information that the virtual disk that will nullify is corresponding simultaneously.

Optionally, when described file operation instruction is operating writing-file instruction, described file operation instruction sending module, comprising:

Written document request sends submodule, for the memory address corresponding in storage server according to the virtual disk that will access comprised in the first obtained metadata information, sends written document request to described storage server;

File data sends submodule, for sending the file data of described file to be written to described storage server, makes described storage server perform the file operation writing described file to be written.

Optionally, described file data sends submodule, comprising:

File data division unit, for according to the Divide File rule preset, by the file data of described file to be written, is divided into multiple data block;

File data transmitting element, for calling at least one process, send to described storage server and divide file data corresponding to the data block that obtains, described storage server is made to perform the file operation writing described file to be written, wherein, the file data that one or more data block is corresponding is sent by a process.

Optionally, described file data division unit, comprising:

Number of processes obtains subelement, for obtaining the quantity of the process being in idle condition;

File data divides subelement, for according to the obtained quantity being in the process of idle condition, by the file data of described file to be written, is divided into multiple data block.

Optionally, described file data transmitting element, comprising:

Data block quantity obtains subelement, for obtaining the quantity dividing the data block obtained;

File data sends subelement, to send to described storage server divide file data corresponding to the data block that obtains for calling several processes, described storage server is made to perform the file operation writing described file to be written, wherein, the quantity of institute's calling process is identical with the quantity dividing the data block obtained.

Optionally, when described file operation instruction is for reading file operation instruction,

Described file operation instruction sending module, specifically for the memory address corresponding in storage server according to the virtual disk that will access comprised in the first obtained metadata information, send to described storage server and read file operation instruction, make described storage server perform described in read file operation instruction corresponding read file operation.

For achieving the above object, the embodiment of the invention discloses a kind of disc operating system (DOS) based on virtual machine, described system comprises:

The client-server of the disk operating device based on virtual machine that meta data server, storage server and application are above-mentioned;

Wherein, described meta data server, for storing the metadata information that in described client-server, virtual disk is corresponding, receive the metadata request information for virtual disk that described client-server sends, and send the metadata information corresponding with this metadata request information according to this metadata request information to described client-server, in client-server described in described storage server requests, the memory address of the file that virtual disk is corresponding, is communicated with described client-server and storage server by network;

Described storage server, for receiving the address request information for virtual disk in described client-server that described meta data server sends, and send the memory address of virtual disk respective file in described client-server to described meta data server according to this address request information, receive the file operation instruction that described client-server sends, perform the file operation that described file operation instruction is corresponding.

As seen from the above, in the scheme that the embodiment of the present invention provides, before the virtual disk in access client server, first obtain metadata information corresponding to virtual disk from meta data server, then this metadata information is stored in client-server.After receiving the file operation requests from user, according to the file operation requests of user, during virtual disk in access client server, from client-server, directly obtain the first metadata information that the virtual disk that will access is corresponding, and send file operation instruction according to the first obtained metadata information to each storage server, and then carry out corresponding file operation by each storage server according to this file operational order.Compared with prior art, in the scheme that the embodiment of the present invention provides, the metadata information corresponding due to the virtual disk in client-server has been stored in client-server, therefore, during virtual disk in access client server, the metadata information that the virtual disk that will access is corresponding can be obtained from client-server, without the need to connecting meta data server again, reduce the burden of meta data server, without the need to passing through Internet Transmission, save the time, file operation can be reduced and postpone, improve Consumer's Experience.

Accompanying drawing explanation

In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, be briefly described to the accompanying drawing used required in embodiment or description of the prior art below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.

A kind of concrete structure schematic diagram of the distributed file system that Fig. 1 a provides for the specific embodiment of the invention;

Fig. 1 b is the signaling process schematic diagram based on the disk operating method of virtual machine in prior art;

The signaling process schematic diagram of the disk operating method based on virtual machine that Fig. 2 provides for the embodiment of the present invention;

The first schematic flow sheet of the disk operating method based on virtual machine that Fig. 3 provides for the embodiment of the present invention;

The second schematic flow sheet of the disk operating method based on virtual machine that Fig. 4 provides for the embodiment of the present invention;

The third schematic flow sheet of the disk operating method based on virtual machine that Fig. 5 provides for the embodiment of the present invention;

4th kind of schematic flow sheet of the disk operating method based on virtual machine that Fig. 6 provides for the embodiment of the present invention;

The first structural representation of the disk operating device based on virtual machine that Fig. 7 provides for the embodiment of the present invention;

The second structural representation of the disk operating device based on virtual machine that Fig. 8 provides for the embodiment of the present invention;

The third structural representation of the disk operating device based on virtual machine that Fig. 9 provides for the embodiment of the present invention;

The structural representation of a kind of disc operating system (DOS) based on virtual machine that Figure 10 provides for the embodiment of the present invention.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.

First simply introduce the technical term related in present specification, with reference to Fig. 1 a, Fig. 1 a is a kind of concrete structure schematic diagram of distributed file system.

Distributed file system: be not necessarily connected directly between on local memory device at the physical memory resources of distributed file system file system management, but can be connected with the memory device of network side by computer network, distributed file system is generally based on the pattern of client end/server end.

The distributed file system of the application is made up of client-server, meta data server, storage server etc., client-server, is connected by network between meta data server and storage server.

Wherein, meta data server, for the metadata information of storage file, the metadata information of file comprises: memory address of the size of file, the attribute of file, file etc.

Storage server, for storage file, for read-write on client side file provides relevant information.In distributed file system, can comprise multiple stage storage server, can comprise one or more hard disk in every platform stores service, wherein, a storage server is called a memory node.

Virtual machine: refer to by software simulation have complete hardware system function, the complete computer operated in a completely isolated environment.

Virtual disk: can be understood as the disk fictionalized inside a computing machine inside a remote computer.

In distributed file system, in client-server, at least one virtual machine can be installed, at least one virtual disk can be fictionalized in each virtual machine, namely multiple virtual disk can be there is in each virtual machine in client-server, for ease of simplified characterization, multiple virtual disk can be there is being simply described as in client-server as follows.Wherein, the virtual disk in virtual machine can be understood as the mapping of file on client-server of specific size in storage server, during a certain virtual disk in user's access client server, and the corresponding document namely in access storage server.

Because virtual disk can be understood as the file of specific size in storage server, and meta data server is for storing metadata information corresponding to this file system file in distributed file system, then the metadata information that each virtual disk in client-server is corresponding all should be stored in meta data server.

In addition, virtual disk in the application in virtual machine it is also understood that as being bipartite, a part corresponds to the file of specific size in storage server, the i.e. storage space of specific size, another part corresponds to the storage space of specific size in client-server, such as, the storage space of specific size in the storage space of specific size or hard disk in storage server buffer memory, above-mentioned two parts can be interrelated, not separate, certainly, also can be separate.

In a kind of specific implementation of the application, by prior art, above-mentioned two parts can be presented to user with the form of a disk, and without the need to indicating that to user this disk is bipartite.When presenting this disk to user, before the part corresponding to client-server can being positioned over the part corresponding to storage server, certainly, after also the part corresponding to client-server can being positioned over the part corresponding to storage server, the application does not limit this.

In the another kind of specific implementation of the application, when presenting virtual disk to user, only can present virtual disk corresponding to the part in storage server, and do not present the part that virtual disk corresponds to client-server, but associated by forms such as links between the two, because virtual disk corresponds to the part of client-server for storing metadata information, user can be prevented like this due to deletion metadata informations such as maloperations.

More than comprehensive, based on the operation of the disk of virtual machine, can be understood as based on the file operation in distributed file system.

From the angle of carrying out contrasting with prior art, the present invention is described in detail below.

Fig. 1 b is the signaling process schematic diagram based on the disk operating method of virtual machine in prior art.

Because the metadata information that virtual disk is corresponding is stored in meta data server, in prior art, when at every turn carrying out file operation in based on the distributed file system of virtual machine, need first to send metadata obtaining request to meta data server, the metadata information (S101) that the virtual disk will accessed with acquisition is corresponding, meta data server inquires about metadata information corresponding to the virtual disk that will access wherein according to received metadata obtaining request, and the metadata information inquired is sent to client-server (S102), client-server sends file operation instruction (S103) according to received metadata information to storage server, storage server performs file operation corresponding to this file operational order (S104) according to this file operational order.

From foregoing description, in prior art, when carrying out file operation in based on the distributed file system of virtual machine, even if obtained the metadata information that the virtual disk that will access is corresponding in file operation before, carry out file operation had still needed first to obtain metadata information corresponding to the virtual disk that will access by network from meta data server at every turn.And in the file operation method of the distributed file system based on virtual machine that the embodiment of the present invention provides, before virtual disk in access client server, metadata information corresponding for this virtual disk is stored in client-server, if when carrying out file operation in a certain virtual disk of user in client-server afterwards, only need obtain corresponding metadata information from client-server, and without the need to obtaining from meta data server again.

Why the distributed file system based on virtual machine of prior art adopts the legacy systems structure be placed on by the metadata information of file in meta data server, be because previous distributed file system is designed to this structure, and do not consider the feature of virtual machine self.Old-fashioned system architecture has still been continued to use after the distributed file system based on virtual machine occurs.And this old-fashioned system architecture is relative to the distributed file system based on virtual machine, suitable not.First, the metadata information of each virtual disk is all kept in a meta data server, and the heavier loads of meta data server, needs the demand ceaselessly responding each user; During the data stored in user's accesses virtual disk, no matter be write operation or read operation, all will through meta data server, this will cause network to take seriously, and the time that whole distributed file system makes response to user's request is longer.

Concrete, the signaling process schematic diagram of the disk operating method based on virtual machine that Fig. 2 provides for the embodiment of the present invention.

When the disk operating method of the distributed file system based on virtual machine that the application embodiment of the present invention provides carries out file operation, from the metadata information that client-server has stored, first obtain the metadata information (S201) that the virtual disk that will access is corresponding, again according to obtained metadata information, send file operation instruction (S202) to storage server, storage server performs file operation corresponding to this file operational order (S203) according to the file operation instruction received.

Visible, when the scheme that the application embodiment of the present invention provides carries out file operation, only need prestore metadata information corresponding to virtual disk in client-server, such as, can when creating virtual disk, corresponding metadata information is obtained from meta data server, and when carrying out file operation in user's each virtual disk in client-server, corresponding metadata information is obtained from client-server, with obtained the metadata information of the virtual disk that will access by network mode compared with, system delay can be reduced.

By instantiation, the present invention is described in detail more below.

The first schematic flow sheet of the disk operating method based on virtual machine that Fig. 3 provides for the embodiment of the present invention, the method is applied to client-server, comprising:

S301: obtain metadata information corresponding to virtual disk from meta data server.

In practical application, multiple virtual machine can be installed in client-server, and one or more virtual disk can be created in each virtual machine.

In a preferred embodiment of the invention, when creating virtual disk, metadata information corresponding to created virtual disk can be obtained from meta data server.

When creating magnetic disk of virtual machine, distributed file system is by meta data server or storage space corresponding to the whole virtual disk of storage server predistribution, and obtain metadata information corresponding to virtual disk from meta data server, then this metadata information is stored in virtual machine.Receive the file operation requests from user, and according to the file operation requests of user, during virtual disk in access client server, directly can obtain the first metadata information that the virtual disk that will access is corresponding from virtual machine.

After can certainly creating virtual disk in client-server, any moment of accessing before this virtual disk obtains the metadata information of this virtual disk from meta data server.

S302: by above-mentioned metadata information, be stored in client-server.

Concrete, there is at least one virtual machine owing to creating in client-server, multiple virtual disk may have been created in each virtual machine, therefore by above-mentioned metadata information, be stored in each virtual disk of each virtual machine of client-server.

S303: during virtual disk in access client server, from one or more metadata informations that client-server stores, obtains the first metadata information that the virtual disk that will access is corresponding.

In a kind of concrete implementation, user can pass through network entry client-server, accesses wherein each virtual disk.In practical application, access rights of each virtual disk etc. in user's access client server can also be set.

S304: according to the first obtained metadata information, sends file operation instruction to storage server, makes the file operation that storage server execute file operational order is corresponding.

Can be known by aforementioned description, the first metadata information obtained comprises the storage address information of virtual disk in storage server that will access, therefore, after acquisition first metadata information, the storage address information that can comprise according to this information, sends file operation instruction to storage server.

Wherein, file operation instruction roughly can be divided into: read file operation instruction, operating writing-file instruction or other operations for file.Can comprise in file operation instruction: the information such as the version information of the mark of file, the mark of file operation, file, certainly, the application does not limit the information comprised in file operation instruction.The mark of above-mentioned file operation reads file operation, operating writing-file or other operations for file for representing that this file is operating as.

In a preferred embodiment of the invention, the method can also comprise:

The object information of the file operation that the storage server execute file operational order that reception storage server sends is corresponding.

Wherein, can running time etc. of include file the operation whether successfully mark of the mark of information, file operation, file, the memory location of file, the version information of file and file operation in the object information of file operation, certainly, the application does not limit the information comprised in the object information of file operation.

After client-server receives the object information for file operation, according to this information spanned file Operation Log information in client-server, with the file operation of recording user in virtual disk, daily servicing can be convenient to.

Further, in another preferred embodiment of the invention, after the object information receiving file operation corresponding to storage server execute file operational order that storage server sends, the object information of file operation corresponding to storage server execute file operational order can also be sent to meta data server, to make meta data server according to this information spanned file Operation Log information, prevent because client-server failure and other reasons causes the file operation log information recorded in client-server to be lost.

Certainly, in practical application, client-server also can receive only the object information of file operation corresponding to storage server execute file operational order that stores service sends, and this object information is sent to meta data server, by meta data server according to this object information spanned file Operation Log, and client-server is not according to this object information spanned file Operation Log information, so both can alleviate the work load of client-server, the centralized management to file operation information can have been realized again.

In another preferred embodiment of the present invention, when nullifying the virtual disk in client-server, delete the second metadata information that the virtual disk that will nullify is corresponding.

Concrete, the virtual disk that nullify can be stored in client-server.

Concrete, no matter metadata information is stored in the storage space of specific size in client-server corresponding to virtual disk, still be stored in any storage space preset in client-server, the virtual disk that metadata information is corresponding with it is all interrelated, not separate, when nullifying virtual disk, nullify virtual disk corresponding to the part in storage server, and delete metadata information corresponding to this virtual disk of storing in client-server, but not only nullify virtual disk corresponding to the part in storage server, effectively can avoid the situation of the storage resources too much taking client-server like this.

As seen from the above, in the scheme that the present embodiment provides, when creating magnetic disk of virtual machine, the space of the whole virtual disk of distributed file system predistribution, and obtain metadata information corresponding to virtual disk from meta data server, then this metadata information is stored in the virtual disk of client-server.After receiving the file operation requests from user, according to the file operation requests of user, during virtual disk in access client server, from the virtual machine of client-server, directly obtain the first metadata information that the virtual disk that will access is corresponding, and send file operation instruction according to the first obtained metadata information to each storage server, and then carry out corresponding file operation by each storage server according to this file operational order, and by file operation result feedback to the virtual machine of client-server.Compared with prior art, in the scheme that the present embodiment provides, the metadata information corresponding due to the virtual disk in client-server has been stored in client-server, therefore, after virtual machine creating, during virtual disk in access client server, the metadata information that the virtual disk that will access is corresponding can be obtained from client-server, without the need to connecting meta data server again, reduce the burden of meta data server, without the need to by Internet Transmission, save the time, file operation can be reduced postpone, improve Consumer's Experience.

Below respectively to perform operating writing-file and to read file operation in based on the distributed file system of virtual machine, be described respectively.

Concrete, see Fig. 4, provide the second schematic flow sheet of the disk operating method based on virtual machine, the present embodiment file operational order is operating writing-file instruction, wherein, according to the first obtained metadata information, send file operation instruction to storage server, make the file operation (S304) that storage server execute file operational order is corresponding, comprising:

S3041: the memory address corresponding in storage server according to the virtual disk that will access comprised in the first obtained metadata information, sends written document request to storage server.

Wherein, can carry the information such as the mark of file to be written and the version information of file to be written in written document request, certainly, the application does not limit information entrained in written document request.

Provide the definition of virtual machine above, virtual machine refers to have complete hardware system function by software simulation, operate in the complete computer in a completely isolated environment, since what virtual machine was corresponding is complete computer operating system, so those skilled in the art are understandable that, corresponding file management system is had in virtual machine, user is in virtual disk during written document, the relative memory address of writing in files in virtual disk can be stored in the file operation information list in virtual disk by file management system, think that follow-up file operation provides relevant information, again due to virtual disk corresponding be particular memory space in storage server, therefore, file operation list in virtual disk is the file operation list in the particular memory space in storage server corresponding to virtual disk.

Above-mentioned file operation information list, for the relative memory address of file in this virtual disk stored in storing virtual disk.

To in virtual disk during writing in files, two kinds of situations can be divided into:

A kind of situation is: any version that there is not file to be written in virtual disk, now, not about the relative storage address information of this file in virtual disk in file operation information list, in this case, the file management system that virtual disk is corresponding can be the relative memory address of file allocation to be written in virtual disk according to its file management rule, and this relative memory address is stored in file operation information list.

Another kind of situation is: the version that there is file to be written in virtual disk, now, exist about the relative storage address information of this file in virtual disk in file operation information list, in this case, from file operation information list, the relative storage address information of this file in virtual disk is obtained according to information such as the marks of file to be written.

With reference to foregoing description, in this step, the memory address that the virtual disk that will access that client-server comprises according to the first metadata information is corresponding in storage server, after sending written document request to storage server, storage server is according to information such as the marks of the file to be written carried in this written document request, in the file operation information list stored from the storage space that the virtual disk that will access is corresponding, obtain the relative memory address of file to be written in this storage space, the memory address of the virtual disk that will access that comprises in the first metadata information correspondence in storage server is obtained again according to obtained relative memory address and institute, calculate the memory address of file to be written in storage server, and then determine the address writing file to be written.

S3042: the file data sending file to be written to storage server, makes storage server perform the file operation writing file to be written.

When sending the file data of file to be written to storage server, can according to the one section one section transmission of certain data length, can certainly be not segmentation, the file data that whole file to be written is corresponding sends together.

As seen from the above, in the scheme that the present embodiment provides, in based on the distributed file system of virtual machine during written document, the metadata information that the virtual disk that will access is corresponding is obtained from client-server, then corresponding in storage server according to this virtual disk comprised in this metadata information memory address, send written document request to storage server, and complete operating writing-file.Compared with prior art, in the scheme that the present embodiment provides, during virtual disk in access client server, the metadata information that the virtual disk that will access is corresponding can be obtained from client-server, without the need to passing through Internet Transmission, save the time, file operation can be reduced and postpone, improve Consumer's Experience.

In another specific embodiment of the present invention, see Fig. 5, provide the third schematic flow sheet of the disk operating method based on virtual machine, compared with embodiment illustrated in fig. 4, in the present embodiment, send the file data of file to be written to storage server, make storage server perform the file operation (S3042) writing file to be written, comprising:

S30421: according to the Divide File rule preset, by the file data of file to be written, be divided into multiple data block.

Concrete, according to the Divide File rule preset, by the file data of file to be written, be divided into multiple data block, can according to following several at least one carry out:

According to the memory node that various piece in file data is corresponding, by the file data of file to be written, be divided into multiple data block;

According to the quantity of idle process, by the file data of file to be written, be divided into multiple data block;

According to fixing data length, by the file data of file to be written, be divided into multiple data block;

According to the size of the storage space that can use in memory device, by the file data of file to be written, be divided into multiple data block.

The application is just described for above-mentioned, and in practical application, the mode of dividing data block is not limited in above-mentioned several.

In a kind of specific implementation, according to the Divide File rule preset, by the file data of file to be written, be divided into multiple data block, can carry out according to following steps:

Obtain the quantity being in the process of idle condition;

According to the obtained quantity being in the process of idle condition, by the file data of file to be written, be divided into multiple data block.

In addition, those skilled in the art are understandable that, in distributed file system, a file to be written can be divided into one or more part.When file to be written is divided into multiple part, when storing this file to be written, the various piece of this file can be stored in multiple different memory node, also can be stored in same memory node, therefore, also can according to memory node dividing data block corresponding to file various piece to be written, such as, according to the positional information of various piece in file data in the file data of file to be written, to determine in file data that position is adjacent and file data that the is part correspondence that memory node information is identical is data block etc.

Further, after the file data treating writing in files divides, can arrange its data block sequence number in file data for each data block, wherein, data block sequence number is used for the relative position of identification data block in the file data of file to be written.

Such as, suppose that file to be written obtains data block 1, data block 2 and data block 3 after dividing, the sequence number of each data block is: 1,2 and 3, then the relative position that can be understood as each data block in file to be written is: data block 1, data block 2 and data block 3 from file header.

In addition, after the file data treating writing in files divides, the information such as the length of each data block, the initial memory address of data block can also be obtained.

S30422: call at least one process, sends to storage server and divides file data corresponding to the data block that obtains, make storage server perform the file operation writing file to be written.

It should be noted that, in this step, file data corresponding to one or more data block is sent by a process.

In prior art, usually a process sends file data from file to be written to storage server is only called, and along with the fast development of computer technology, multi-process technology obtains to be applied more and more widely, when there is idle process in client-server, the file data that a process sends file to be written to storage server if only call fully cannot call computer resource.

Concrete, call at least one process, to storage server send divide file data corresponding to the data block that obtains time, first can obtain the quantity dividing the data block obtained, call several processes again to send to storage server and divide file data corresponding to the data block that obtains, make storage server perform the file operation writing file to be written, wherein, the quantity of institute's calling process is identical with the quantity dividing the data block obtained.

In addition, call at least one process, to storage server send divide file data corresponding to the data block that obtains time, first can also obtain the quantity dividing memory node corresponding to the data block that obtains, call several processes again to send to storage server and divide file data corresponding to the data block that obtains, storage server is performed and writes the file operation of file to be written, wherein, the quantity of the memory node that the quantity of institute's calling process is corresponding with the data block that division obtains is identical.

As seen from the above, in the scheme that the present embodiment provides, the file data of file to be written is divided into multiple data block, and calls at least one process and send file data corresponding to each data block to storage server.Compared with calling a process in prior art and to send to storage server the file data of file to be written, can computer resource be made full use of, the speed sending file to be written to storage server can be improved.

In another specific embodiment of the present invention, see Fig. 6, provide the 4th kind of schematic flow sheet of the disk operating method based on virtual machine, the present embodiment file operational order is for reading file operation instruction, wherein, according to the first obtained metadata information, send file operation instruction to storage server, make the file operation (S304) that storage server execute file operational order is corresponding, comprising:

S3043: the memory address corresponding in storage server according to the virtual disk that will access comprised in the first obtained metadata information, to storage server send read file operation instruction, make storage server perform read file operation instruction corresponding read file operation.

Wherein, read to carry the information such as the mark of file to be read and the version information of file to be read in file operation instruction, certainly, the application does not limit and reads information entrained in file request.

Concrete, the memory address that client-server is corresponding in storage server according to the virtual disk that will access comprised in the first obtained metadata information, after reading file operation instruction to storage server transmission, in the file operation information list that storage server can first store from storage space corresponding to the virtual disk that will access, obtain the relative memory address of file to be read in this storage space, the memory address of the virtual disk that will access that comprises in the first metadata information correspondence in storage server is obtained again according to obtained relative memory address and institute, calculate the address of file to be read in storage server, and then according to above-mentioned result of calculation, perform read file operation instruction corresponding read file operation.

As seen from the above, in the scheme that the present embodiment provides, when reading file in based on the distributed file system of virtual machine, the metadata information that the virtual disk that will access is corresponding is obtained from client-server, then corresponding in storage server according to this virtual disk comprised in this metadata information memory address, send to storage server and read file request, and completely read file operation.Compared with prior art, in the scheme that the present embodiment provides, during virtual disk in access client server, the metadata information that the virtual disk that will access is corresponding can be obtained from client-server, without the need to passing through Internet Transmission, save the time, file operation can be reduced and postpone, improve Consumer's Experience.

Corresponding with the above-mentioned disk operating method based on virtual machine, the embodiment of the present invention additionally provides a kind of disk operating device based on virtual machine.

The first structural representation of the disk operating device based on virtual machine that Fig. 7 provides for the embodiment of the present invention, this application of installation, in client-server, comprising: the first metadata information obtains module 701, metadata information memory module 702, second metadata information obtains module 703 and file operation instruction sending module 704.

Wherein, the first metadata information obtains module 701, for obtaining metadata information corresponding to virtual disk from meta data server;

Metadata information memory module 702, for by described metadata information, is stored in described client-server;

Second metadata information obtains module 703, during for accessing the virtual disk in described client-server, from one or more metadata informations that described client-server stores, obtains the first metadata information that the virtual disk that will access is corresponding;

File operation instruction sending module 704, for according to the first obtained metadata information, sends file operation instruction to storage server, makes described storage server perform file operation corresponding to described file operation instruction.

Preferably, the first metadata information obtains module 701, specifically for when creating virtual disk, obtains metadata information corresponding to created virtual disk from meta data server.

Preferably, can also should comprise based on the disk operating device of virtual machine: object information receiver module (not shown).

Wherein, object information receiver module, the described storage server sent for receiving described storage server performs the object information of file operation corresponding to described file operation instruction.

Preferably, can also should comprise based on the disk operating device of virtual machine: object information sending module (not shown).

Wherein, object information sending module, for after described object information receiver module receives described object information, the object information that described storage server performs file operation corresponding to described file operation instruction is sent, to make described meta data server according to this information spanned file Operation Log information to described meta data server.

Concrete, comprise at least one virtual machine in described client-server, at least one virtual disk can be created in described virtual machine;

Metadata information memory module 702, specifically for by described metadata information, be stored in virtual disk corresponding to described metadata information, wherein, described virtual disk is arranged in the virtual disk that described client-server is installed.

Preferably, can also should comprise based on the disk operating device of virtual machine: metadata information removing module.

Wherein, metadata information removing module, for when nullifying the virtual disk in client-server, deletes the second metadata information that the virtual disk that will nullify is corresponding.

Concrete, the virtual disk that nullify is stored in the virtual machine of client-server.

As seen from the above, in the scheme that the present embodiment provides, before virtual disk in access client server, first obtain metadata information corresponding to virtual disk from meta data server, and this metadata information is stored in client-server, during virtual disk in user's access client server, the first metadata information that the virtual disk that will access is corresponding is obtained from client-server, and send file operation instruction according to the first obtained metadata information to storage server, and then carry out corresponding file operation by storage server according to this file operational order.Compared with prior art, in the scheme that the present embodiment provides, the metadata information corresponding due to the virtual disk in client-server has been stored in client-server, therefore, during virtual disk in access client server, the metadata information that the virtual disk that will access is corresponding can be obtained from client-server, without the need to passing through Internet Transmission, save the time, file operation can be reduced and postpone, improve Consumer's Experience.

In one particular embodiment of the present invention, see Fig. 8, provide the second structural representation of the disk operating device based on virtual machine, the present embodiment file operational order is operating writing-file instruction, wherein, file operation instruction sending module 704, comprising: written document request sends submodule 7041 and file data sends submodule 7042.

Wherein, written document request sends submodule 7041, for the memory address corresponding in storage server according to the virtual disk that will access comprised in the first obtained metadata information, sends written document request to described storage server;

File data sends submodule 7042, for sending the file data of described file to be written to described storage server, makes described storage server perform the file operation writing described file to be written.

Concrete, written document request sends submodule 7041 and can comprise: file relative address obtains unit, file memory address computing unit and written document request transmitting unit (not shown).

In another specific embodiment of the present invention, see Fig. 9, provide the third structural representation of the disk operating device based on virtual machine, compared with embodiment illustrated in fig. 8, the present embodiment file data send submodule 7042, comprising: file data division unit 70421 and file data transmitting element 70422.

Wherein, file data division unit 70421, for according to the Divide File rule preset, by the file data of described file to be written, is divided into multiple data block;

File data transmitting element 70422, for calling at least one process, send to described storage server and divide file data corresponding to the data block that obtains, described storage server is made to perform the file operation writing described file to be written, wherein, the file data that one or more data block is corresponding is sent by a process.

Concrete, file data division unit 70422 can comprise: number of processes obtains subelement and file data divides subelement (not shown).

Wherein, number of processes obtains subelement, for obtaining the quantity of the process being in idle condition;

Concrete, file data transmitting element 70422 can comprise: data block quantity obtains subelement and file data sends subelement (not shown).

As seen from the above, in the scheme that the present embodiment provides, the file data of file to be written is divided into multiple data block, and call at least one process and send file data corresponding to each data block to storage server, compared with calling a process in prior art and to send to storage server the file data of file to be written, can computer resource be made full use of, the speed sending file to be written to storage server can be improved.

In another specific embodiment of the present invention, when file operation instruction is for reading file operation instruction, file operation instruction sending module 704, specifically for the memory address corresponding in storage server according to the virtual disk that will access comprised in the first obtained metadata information, send to described storage server and read file operation instruction, make described storage server perform described in read file operation instruction corresponding read file operation.

Corresponding with the above-mentioned disk operating method based on virtual machine, device, the embodiment of the present invention additionally provides a kind of disc operating system (DOS) based on virtual machine.

The structural representation of a kind of disc operating system (DOS) based on virtual machine that Figure 10 provides for the embodiment of the present invention, this system comprises: the client-server 1103 of the disk operating device based on virtual machine that any embodiment provides in meta data server 1101, storage server 1102 and application drawing 7 to Fig. 9;

Wherein, described meta data server 1101, for storing the metadata information that in described client-server 1103, virtual disk is corresponding, receive the metadata request information for virtual disk that described client-server 1103 sends, and send the metadata information corresponding with this metadata request information according to this metadata request information to described client-server 1103, the memory address of the file that virtual disk is corresponding in described client-server 1103 is asked to described storage server 1102, communicated with described client-server 1103 and storage server 1102 by network,

Described storage server 1102, for receiving the address request information for virtual disk in described client-server 1103 that described meta data server 1101 sends, and send the memory address of virtual disk respective file in described client-server 1103 to described meta data server 1101 according to this address request information, receive the file operation instruction that described client-server 1103 sends, perform the file operation that described file operation instruction is corresponding.

For device, system embodiment, because it is substantially similar to embodiment of the method, so description is fairly simple, relevant part illustrates see the part of embodiment of the method.

It should be noted that, in this article, the such as relational terms of first and second grades and so on is only used for an entity or operation to separate with another entity or operational zone, and not necessarily requires or imply the relation that there is any this reality between these entities or operation or sequentially.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thus make to comprise the process of a series of key element, method, article or equipment and not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise by the intrinsic key element of this process, method, article or equipment.When not more restrictions, the key element limited by statement " comprising ... ", and be not precluded within process, method, article or the equipment comprising described key element and also there is other identical element.

One of ordinary skill in the art will appreciate that all or part of step realized in said method embodiment is that the hardware that can carry out instruction relevant by program has come, described program can be stored in computer read/write memory medium, here the alleged storage medium obtained, as: ROM/RAM, magnetic disc, CD etc.

The foregoing is only preferred embodiment of the present invention, be not intended to limit protection scope of the present invention.All any amendments done within the spirit and principles in the present invention, equivalent replacement, improvement etc., be all included in protection scope of the present invention.

Claims

1. based on a disk operating method for virtual machine, be applied to client-server, it is characterized in that, described method comprises:

2. method according to claim 1, is characterized in that, describedly obtains metadata information corresponding to virtual disk from meta data server, comprising:

3. method according to claim 1 and 2, is characterized in that, described method also comprises:

4. method according to claim 3, is characterized in that, after the described storage server of described reception described storage server transmission performs the object information of file operation corresponding to described file operation instruction, also comprises:

5. method according to claim 1 and 2, is characterized in that,

Comprise at least one virtual machine in described client-server, at least one virtual disk can be created in described virtual machine;

6. method according to claim 1 and 2, is characterized in that, described method also comprises:

7. method according to claim 1 and 2, it is characterized in that, when described file operation instruction is operating writing-file instruction, described according to the first obtained metadata information, file operation instruction is sent to storage server, make described storage server perform file operation corresponding to described file operation instruction, comprising:

8. method according to claim 7, is characterized in that, the described file data sending described file to be written to described storage server, makes described storage server perform the file operation writing described file to be written, comprising:

9. method according to claim 8, is characterized in that, the described Divide File rule according to presetting, by the file data of described file to be written, is divided into multiple data block, comprises:

Obtain the quantity being in the process of idle condition;

10. method according to claim 8, it is characterized in that, described in call at least one process, send to described storage server and divide file data corresponding to the data block that obtains, make described storage server perform the file operation writing described file to be written, comprising:

Obtain the quantity dividing the data block obtained;

11. methods according to claim 1 and 2, it is characterized in that, when described file operation instruction is for reading file operation instruction, described according to the first obtained metadata information, file operation instruction is sent to storage server, make described storage server perform file operation corresponding to described file operation instruction, comprising:

12. 1 kinds, based on the disk operating device of virtual machine, are applied to client-server, it is characterized in that, described device comprises:

13. devices according to claim 12, is characterized in that, described first metadata information obtains module, specifically for when creating virtual disk, obtains metadata information corresponding to created virtual disk from meta data server.

14. devices according to claim 12 or 13, it is characterized in that, described device also comprises:

15. devices according to claim 14, is characterized in that, described device also comprises:

16. devices according to claim 12 or 13, is characterized in that,

17. devices according to claim 12 or 13, it is characterized in that, described device also comprises:

18. devices according to claim 12 or 13, it is characterized in that, when described file operation instruction is operating writing-file instruction, described file operation instruction sending module, comprising:

19. devices according to claim 18, is characterized in that, described file data sends submodule, comprising:

20. devices according to claim 19, is characterized in that, described file data division unit, comprising:

21. devices according to claim 19, is characterized in that, described file data transmitting element, comprising:

22. devices according to claim 12 or 13, is characterized in that, when described file operation instruction is for reading file operation instruction,

23. 1 kinds, based on the disc operating system (DOS) of virtual machine, is characterized in that, described system comprises:

Meta data server, storage server and application rights require the client-server of the disk operating device based on virtual machine according to any one of 12 to claim 22;