CN106484322A - A kind of method of carry file system, device and equipment - Google Patents
A kind of method of carry file system, device and equipment Download PDFInfo
- Publication number
- CN106484322A CN106484322A CN201610819478.6A CN201610819478A CN106484322A CN 106484322 A CN106484322 A CN 106484322A CN 201610819478 A CN201610819478 A CN 201610819478A CN 106484322 A CN106484322 A CN 106484322A
- Authority
- CN
- China
- Prior art keywords
- communication connection
- target storage
- node
- storage node
- client
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/061—Improving I/O performance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0614—Improving the reliability of storage systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0629—Configuration or reconfiguration of storage systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/067—Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
Abstract
The embodiment of the invention discloses a kind of method of carry file system, including：Client obtains the mark of multiple target storage node, and client sets up the multiple communication connections with multiple target storage node according to the mark of multiple target storage node, and client is using multiple communication connection carry file system.So, in the mount point set up by single client, employ a plurality of communication connection that is set up with multiple memory nodes, enable a client to the multiple memory nodes of concurrent access, so as to improve the transfer rate between client and storage system, while also improving the operation stability that high bandwidth requires business.
Description
Technical field
The present invention relates to the communications field, more particularly to a kind of method of carry file system, device and equipment.
Background technology
The network storage realizes data transfer based on computer network with standard network protocol, is Windows/Linux/Mac OS in network etc.
The client of various different operating systems provides the backup of file-sharing data.Multiple client is by connecting same storage system
(i.e. the network storage) is so as to realize the different business demand of different clients.File in such as basic reading storage system,
Or file etc. is write toward in storage system.As requirement of the storage system to memory capacity is larger, so storage system is general
It is made up of multiple memory nodes.
During multiple file of the client in storage system is accessed, it is (logical that single client is accomplished by very high stable bandwidth
Often>2GB/s).But, with the demand more and more higher of people, the higher more stable bandwidth ability of increasing service needed
Enough realizations, such as video production, a high-performance calculation group of planes (English full name：High Performance Computing, English contracting
Write：) etc. HPC business.And in the prior art, based on standard NFS protocol, single client can only be single with storage system
Memory node sets up connection.When the steady point bandwidth of the single memory node in storage system can not meet separate unit work station demand,
Business can be then affected normally to run, such as video traffic, it is possible that interim card situation.Therefore, such as in existing connection mechanism,
, when the business that high bandwidth is required is run, as single-unit point bandwidth is unstable, the bandwidth for providing is not enough so that visitor for single client
Transfer rate between family end and storage system is relatively low, so as to cause the normal operation of this kind of business.
Content of the invention
This application provides a kind of method of carry file system, device and equipment, for improving client with storage system
Transfer rate between system, improves the operation stability that high bandwidth requires business.
The application first aspect provides a kind of method of carry file system, and this document system is used for managing in storage system
The file of storage, the method include：
The storage system includes multiple memory nodes, and client selects multiple memory nodes as target from storage system
Memory node, the target storage node need to set up the memory node of communication connection for client, and client obtains multiple targets
The mark of memory node, the mark can be the unique identifier of memory node, such as Internet protocol (English full name：
Internet Protocol, english abbreviation：Abbreviation IP) address, or the sequence number of memory node, or the node of memory node
Title etc., client set up multiple communication connections according to the mark of multiple target storage node with multiple target storage node, visitor
Family end reuses the plurality of communication connection carry file system, creates mount point.
Client obtains the mark of multiple target storage node, and client is built according to the mark of multiple target storage node
Vertical multiple communication connections with multiple target storage node, client is using multiple communication connection carry file system.So, single
In the mount point set up by individual client, a plurality of communication connection that is set up with multiple memory nodes is employed so that client
Can the multiple memory nodes of concurrent access, so as to improve the transfer rate between client and storage system, while also improving
High bandwidth requires the operation stability of business.
In a kind of possible implementation, before the mark that client obtains multiple target storage node, the method
Can also include：
In the memory node that client includes, the storage of in be stored with file destination any data block is determined from storage system
Node is used as the target storage node.
Within the storage system, single memory node typically only stores documentary data block, in the implementation, client
End prioritizing selection is set up the memory node of connection and is stored with the data block of file destination that the client needs to obtain, such as stores
Node includes first node, Section Point and the 3rd node, and file destination is divided into two parts data block and is stored respectively in
In one node and Section Point.And client with memory node set up communication connection before, prioritizing selection be stored with target text
The first node of part data block and Section Point, so that client can be directly from first node when file destination is obtained
With obtain file destination in Section Point, reduce the memory node connected by client and obtain targets text from other memory nodes
The time of part data block, so as to improve the data transmission efficiency between client and storage system.
In alternatively possible implementation, client according to the mark of multiple target storage node, set up with multiple
Multiple communication connections of target storage node are specifically as follows：
Client determines the connection of multiple target storage node according to the mark of target storage node from join index storehouse
Index, client set up the multiple communication link with multiple target storage node according to the join index of multiple target storage node
Connect.
In the implementation, client preserves a join index storehouse, preserves each and deposit in the join index storehouse
Storage node sets up, with the client, the join index used during communication connection first.Need again to deposit with this so as to next client
When storage node sets up communication connection, only need to depositing the mark of the target storage node accessed by client and join index storehouse
The mark of storage node is mated, then can find corresponding join index, so as to set up communication connection.Without the need for client again with
Memory node is transmitted far call request and returns a series of preparation operations for setting up communication connection such as entrance handle, improves
Client sets up the efficiency of communication connection with memory node.
In alternatively possible implementation, the method can also include：
After file system described in client carry, client determines at least one communication link from multiple communication connections
Connect, client accesses corresponding target storage node based at least one communication connection to obtain at least one file destination.
If being transmitted with file as granularity between client and storage system, then client obtains multiple mesh at the same time
During mark file, then multiple file destinations can be obtained by the multiple target storage node of multiple communication connection concurrent accesses.Than
As client so, can improve the business of client with three memory nodes of concurrent access so as to obtain three file destinations
Operating rate.
The application second aspect provides a kind of device of carry file system, and this document system is used for managing in storage system
The file of storage, the device can be realized by software, it is also possible to realized by way of soft or hard is combined, in a kind of implementation
In, the device includes：
Acquiring unit, for obtaining the mark of multiple target storage node, storage system includes multiple memory nodes, the mark
Know to be the unique identifier of memory node, such as IP address, or the sequence number of memory node, or the section of memory node
Point title etc.；
Connection unit, for the mark according to multiple target storage node, sets up multiple with multiple target storage node
Communication connection；
Carry unit, for using multiple communication connection carry file system.
In another kind of implementation, the device includes：Transceiver, memory and processor, transceiver, memory with
And processor is connected by bus, memory internal memory contains computer instruction, and processor is by executing the computer in memory
The method of the carry file system provided by instruction realization such as above-mentioned first aspect.
The third aspect provides a kind of equipment, and the equipment includes at least one processor, memory and communication interface.Described at least
One processor, the memory and the communication interface are all connected by bus；The memory storage computer is executed and is referred to
Order；The computer executed instructions of memory storage described at least one computing device so that the equipment is by described
Various may the design that communication interface carries out data interaction to execute above-mentioned first aspect or first aspect with storage system carries
For carry file system method, or cause the equipment to carry out data interaction by the communication interface and storage system
To realize the device of the carry file system in second aspect.
The application fourth aspect provides a kind of computer-readable recording medium, and be stored with computer-readable recording medium meter
Calculation machine execute instruction, when at least one computing device computer executed instructions of equipment, equipment executes above-mentioned first party
Face or the various methods that may design the carry file system for providing of first aspect.
The 5th aspect of the application provides a kind of computer program, and the computer program includes that computer is executed and refers to
Order, the computer executed instructions are stored in a computer-readable storage medium.At least one processor of equipment can be from calculating
Machine readable storage medium storing program for executing reads the computer executed instructions, and at least one computing device computer executed instructions cause equipment
Implement the various methods that may design the carry file system for providing of above-mentioned first aspect or first aspect.
Description of the drawings
Fig. 1 is a system architecture schematic diagram of carry file system in inventive embodiments；
Fig. 2 is a schematic diagram of the method for carry file system in the embodiment of the present invention；
Fig. 3 is a schematic diagram of the device of carry file system in the embodiment of the present invention；
Fig. 4 is a schematic diagram of equipment in the embodiment of the present invention.
Specific embodiment
Embodiments provide a kind of method of carry file system, device and equipment, for improve client with
Transfer rate between storage system, improves the operation stability that high bandwidth requires business.
With reference to shown in Fig. 1, Fig. 1 is a system architecture schematic diagram of carry file system in the embodiment of the present invention.At this
In system architecture, including client, storage system.Wherein, storage system is made up of multiple memory nodes, and multiple memory nodes lead to
Optical fiber connection switch is crossed, and the switch connects the client by optical fiber, so as to multiple memory nodes be realized with client
Interconnection.Wherein, optical fiber can be 10 gigabit Ethernets (English full name：Gigabit Ethernet, english abbreviation：GE) optical fiber
Line, switch can be the high rate data transmission switch of 10GE.The client that is stored with memory node needs the target text for obtaining
The whole file of part or the data block of file, can carry out data and mutually pass between multiple memory nodes, client can pass through
Access some memory node to obtain the file destination for being stored in the memory node or being not stored on the memory node.When
When client needs the file destination for obtaining to be not stored in the memory node, mark time of the memory node by the file destination
Other memory nodes that goes through in whole storage system, so as to obtain the data of the file destination from other memory nodes.Single
The method of the carry file that client can also be provided by the present invention is logical with the multiple memory nodes foundation in the storage system
Letter connection, obtains file destination so as to realize the multiple memory nodes of concurrent access, improves between client and storage system
Message transmission rate.
It should be noted that the client can be personal computer (English full name：Personal Computer, English
Abbreviation：) or the equipment with computing capability such as server PC；Specifically, on the equipment, operation can access storage system
Client-side program comes so that the equipment becomes client.Storage system can be network attached storage (English full name：Network
Attached Storage, english abbreviation：NAS), direct-connected storage (the English full name of open system：Direct Access
Storage english abbreviation：DAS), storage area network (English full name：Storage Area Network, english abbreviation：SAN)
Deng.The hardware medium of memory node can isomorphism or isomery.For the ease of illustrating, the embodiment of the present invention is carried out by taking NAS storage as an example
Illustrate, citing restriction not to the present invention program.
With reference to shown in Fig. 2, Fig. 2 is one embodiment of the method for carry file system in the embodiment of the present invention, wherein, institute
File system is stated for the file stored in storage system is managed, the method is specially：
101st, client obtains the mark of multiple target storage node, and the storage system includes multiple memory nodes.
In order to improve the message transmission rate between client and storage system, in embodiments of the present invention, client is needed
Multiple communication connections to be set up with the multiple memory nodes in storage system.And before communication connection is set up, client needs
The mark of multiple target storage node is obtained, the mark can be the IP address of memory node.Specifically, client obtains storage
The IP address of node can be：Client is first to domain name system (English full name：Domain Name System, english abbreviation：
DNS) server is initiated storage system and initiates Floating IP address inquiry, and the Floating IP address is the virtuality that can be mated with multiple physical IP
IP, wherein, each Floating IP address corresponds to a memory node.Dns server is returned comprising floating IP address in storage system
List.It should be noted that the mark of target storage node can be the unique identifier of memory node, it is not limited in IP
Address, can also make the sequence number of memory node, or the nodename of memory node etc..Dns server belongs to the storage system,
It is responsible for the domain name mapping of the storage system.
Optionally, client can obtain the mark of all memory nodes in storage system, equal with all memory nodes
Set up communication connection.Or, client can also obtain the mark of the part memory node in storage system, with part storage section
Point sets up communication connection.In the mode for setting up communication connection with part memory node, client can be saved according to all storages
The connection loading condition of point is selected, the less memory node of prioritizing selection connection load, and sets up communication connection therewith.
In alternatively possible implementation, position that client can also be located by file destination to be obtained come
Select memory node.Specifically, in the first memory node included from storage system of client, determining in the file destination that is stored with and appointing
The memory node of one data block is used as target storage node.Within the storage system, single memory node is typically only stored with file
Data block, be seldom stored with whole file, therefore, when client needs to obtain certain complete file destination, then needs
The memory node connected by the client obtains other data blocks of the file destination from other memory nodes, so that objective
Family end obtains the whole file destination from the memory node for being connected.Therefore, in the implementation, client first selects to build
The memory node of vertical connection is stored with the data block of file destination that the client needs to obtain, and such as memory node includes first
Node, Section Point and the 3rd node, file destination are divided into two parts data block and are stored respectively in first node and second
In node.And client is before communication connection is set up with memory node, prioritizing selection is stored with the of file destination data block
One node and Section Point, so that client can be directly from first node with Section Point when file destination is obtained
Obtain file destination, reduce memory node that client connected from other memory nodes obtain file destination data blocks when
Between, so as to improve the data transmission efficiency between client and storage system.
102nd, client is set up and the plurality of target storage node according to the mark of the plurality of target storage node
Multiple communication connections.
Client after the mark for obtaining multiple target storage node, i.e., the Floating IP address ground of above-mentioned target storage node
Behind location, client then can set up communication connection with multiple target storage node, and the concrete mode for communicating to connect of setting up is：Client
Memory node request carry of the end according to corresponding to the list to any one Floating IP address (such as first) of floating IP address connects
Connect, the mode that for example asks can send file system mounted far call request, such as MOUNT RPC call request；
Memory node corresponding to first Floating IP address then can return file system root mesh by carry module (such as mountd module)
The entrance handle of record, such as ROOT FILEHANDLE；Client travels through floating IP address list, corresponding deposits with multiple respectively
NFS (the English full name of storage node：Network File System, english abbreviation：NFS) service module is (for example
NFS Server) set up communication connection.The communication connection can connect for socket (Socket), and client is all set up
After Socket connection, it is also possible that the connection status for having Socket connection is stored, the Socket of each memory node connects
The state of connecing includes that the Socket connection status of the memory node is normal or the Socket connection of the abnormal and memory node
The information such as loading condition.Communication connection is set up in client and multiple target storage node, be i.e. after Socket connection, then can enter
Row instantiation, using the plurality of communication connection carry file system.
In the mode of above-mentioned foundation communication connection, client is the Floating IP address by accessed multiple memory nodes
Address, obtains the entrance handle of the plurality of memory node, in real time so as to set up communication connection.And in another kind of implementation,
Client, can also be from the join index storehouse for prestoring really after the floating IP address for determining multiple target storage node
The join index of fixed the plurality of target storage node, sets up communication link further according to the plurality of join index with multiple memory nodes
Connect.
Will not usually be changed due to the entrance handle of each memory node and floating IP address, therefore client exists
After getting the entrance handle of memory node and the floating IP address of the memory node first, then will set up with the client logical
Letter connection, and the join index by the communication connection, the i.e. floating IP address of entrance handle and the memory node are saved in company
Connect in index database, when needing again to set up communication connection with the memory node so as to next client, only need to be by acquired in client
To floating IP address mated with the IP address in join index storehouse, then can find corresponding join index, so as to build
Vertical communication connection.Far call request is transmitted again with memory node without the need for client and returns that entrance handle etc. is a series of to be built
The preparation operation of vertical communication connection, improves the efficiency that client and memory node set up communication connection.If it should be noted that
The current desired memory node to be connected of client is First Contact Connections, then then do not preserve the storage in join index storehouse
The join index of node, needs by logical with memory node foundation the step of above-mentioned real-time foundation communication connection so as to client
Letter connection, after communication connection is set up, the join index of the memory node is stored in the join index storehouse.
103rd, client is using the plurality of communication connection carry file system.
After client is that multiple target storage node set up communication connection, then can carry out instantiating local mount point,
File system local mount point is created.Mount point is actually entering for the disk file system in (SuSE) Linux OS
Mouth catalogue, similar to the C for accessing different subregions in Windows operating system:、D:、E:Deng drive.Client is in example
After changing mount point, then connected memory node can be accessed, so as to obtain file destination.
In the embodiment of the present invention, client obtains the mark of multiple target storage node, and the storage system includes described
Multiple memory nodes, client are set up and the plurality of target storage node according to the mark of the plurality of target storage node
Multiple communication connections, client using the plurality of communication connection carry file system.So, single client is set up
In mount point, employ a plurality of communication connection that is set up with multiple memory nodes so that client can concurrent access multiple
Memory node, so as to improve the transfer rate between client and storage system, while also improve high bandwidth to require business
Operation stability.
In conjunction with Fig. 2 embodiment, in one alternate embodiment, the method for the carry file system also includes：
After file system described in client carry, client determines that from the plurality of communication connection at least one communicates
Connection；
Client accesses corresponding target storage node based on described at least one communication connection to obtain at least one
File destination.
When client and multiple memory nodes set up multiple communication connections, and by these multiple communication connection carry files
After system, client then can access multiple memory nodes according to these communication connections, so as to obtain the mesh in memory node
Mark file.Optionally, client can from multiple communication connection in select one or more communicate to connect, so as to access one or
Multiple memory nodes, and then obtain one or more file destinations.Client obtains the mode of file destination can be by by text
Part is transmitted for granularity, or the data block that is stored in memory node with file destination is transmitted as granularity.Such as,
Client can select a communication connection to access a memory node, of file destination if the memory node is only stored with
Divided data block or the data block of the file destination that is not stored with, then the memory node is then according to the mark of file destination, such as file
Name from other memory nodes are searched for, so as to obtain all data blocks of the file destination, so as to by the institute of the whole file destination
Data block is had to be sent to client.Or, the different data block of the file destination that is stored with multiple memory nodes respectively, then
Client can access multiple memory nodes simultaneously using multiple communication connections, get file destination from different memory nodes
Different pieces of information block, client reintegrates different data blocks again, so as to generate complete file destination.
Optionally, the target storage node that client is also based on corresponding to the plurality of communication connection concurrent access is next
Obtain multiple file destinations.
If being transmitted with file as granularity between client and storage system, then client obtains multiple mesh at the same time
During mark file, then multiple file destinations can be obtained by the multiple target storage node of multiple communication connection concurrent accesses.Than
As client can be with three memory nodes of concurrent access so as to obtain three file destinations.Client obtains file destination and can wrap
Including client carries out the business operations such as read operation or write operation to file destination.
Optionally, above-mentioned client determines at least one communication connection from the plurality of communication connection, specifically may be used
Think：
Client determines at least one little communication connection of traffic load from the plurality of communication connection.
Illustrate in above-described embodiment that client after memory node sets up communication connection, then stores the company of memory node
State is connect, in the connection status, contains the loading condition of memory node.Based on load balancing principle, client is selecting storage
When node carries out data transmission, data biography can be carried out with the memory node of the little communication connection of prioritizing selection current traffic load religion
Defeated.It is specifically as follows：Client determines the first few memory node of communication connection number from the plurality of target storage node,
Again from the plurality of communication connection, determine with the communication connection of first memory node as the little communication link of traffic load
Connect.
It should be noted that when client needs to access plural memory node using plural communication connection
When, then the selected plural communication connection of client is the little communication connection of traffic load.
Optionally, when file destination is at least two, client can also first obtain at least two file destination
In each file size information；
Client each file transmission consumed time according to the transfer rate with storage system is calculated；
Client according to each file transmission consumed time with file as hondrometer calculate average described in each
The number of files that memory node need to be transmitted；
The client obtains at least two target by the multiple memory nodes of the number of files concurrent access
File.
It is 4 that such as client needs the file destination for obtaining, and client selects two communication connections to carry out 4 files
Transmission.Client sends the read request of 4 files first to storage system, and storage system then can be according to 4 files
File identification gets the size information of each file of 4 files, if wherein first file is 10G, second file
For 1G, the 3rd file is 3G, and the 4th file is 2G, then according to the mode that is transmitted as granularity with file, by two companies
Connect and be transmitted, if the transfer rate communicated to connect per bar is 1G/s, then first file of transmission needs 10s is spent, and passes
Defeated second file needs to spend 1s, and the 3rd file of transmission needs to spend 3s, and the 4th file of transmission needs to spend 2s.So
According to principle in a balanced way, first file can be transmitted by a communication line, and another communication link is another to transmit
Outer three files.Or, the size of the 4th file is 8G, then the total size of 4 files is 22G, is communicated by two
Transmitting, averagely every communication line can then transmit two files to circuit, be that first file and second file exist respectively
It is transmitted on one communication line, second file is transmitted on another article of communication line with the 3rd file.So,
So that the size of data load carried out data transmission between client and storage system is more balanced, client and storage is improve
Efficiency of transmission between system.
Optionally, the security multi-link in order to improve XNFS, so that storage system is for NAS storage as an example, can also be by NAS
Unified certification module in storage is transplanted in client, and client and NAS storage solidify a pair of identical account and close respectively
Code, encrypting storing, then used as Authentication Client, NAS is stored as certificate server client, using above-mentioned unified certification machine
System completes two-way authentication, can just carry out file system mounted and file access in two-way authentication by rear.
With reference to shown in Fig. 3, in the embodiment of the present invention, one embodiment of the device of carry file system includes：
Acquiring unit 201, for obtaining the mark of multiple target storage node, the storage system includes multiple storage sections
Point；
Connection unit 202, for the mark according to the plurality of target storage node, is set up and is stored with the plurality of target
Multiple communication connections of node；
Carry unit 203, for using the plurality of communication connection carry file system.
Optionally, described device includes：
Determining unit 204, for, before the mark that the acquiring unit obtains multiple target storage node, depositing from described
In the memory node that storage system includes, determine that the memory node of any data block in the file destination that is stored with is deposited as the target
Storage node.
Optionally, the connection unit 202 specifically for：
According to the mark of the target storage node, the company of the plurality of target storage node is determined from join index storehouse
Connect index；
According to the join index of the plurality of target storage node, set up multiple logical with the plurality of target storage node
Letter connection.
Optionally, the determining unit 204 is additionally operable to, after file system described in 203 carry of carry unit, from institute
State in multiple communication connections and determine at least one communication connection；
The acquiring unit 201 is additionally operable to, and accesses corresponding target storage section based on described at least one communication connection
Put to obtain at least one file destination.
Optionally, the determining unit 204 specifically for：
At least one little communication connection of traffic load is determined from the plurality of communication connection.
Optionally, the determining unit 204 specifically for：
From the plurality of target storage node, the first few memory node of communication connection number is determined；
From the plurality of communication connection, determine little as traffic load with the communication connection of first memory node
Communication connection.
Optionally, the acquiring unit 201 is additionally operable to：
After file system described in carry, based on the target storage node corresponding to the plurality of communication connection concurrent access
To obtain multiple file destinations.
Specific descriptions with regard to unit in Fig. 3 embodiment are referred to the alternative embodiment of Fig. 2 embodiment and Fig. 2
Provided in carry file system method detailed description, do not repeat herein.
The embodiment of the present invention also provides a kind of storage system, and the storage system is the storage system described in above-described embodiment
System, the method for realizing above-mentioned carry file system for coordinating with client.
The embodiment of the present invention also provides a kind of memory node, and the memory node belongs to the storage system described in above-described embodiment
System, the memory node are the memory node described in above-described embodiment, and the memory node is used for coordinating to realize with client
The method of the carry file system that states.
Shown in Figure 4, the embodiment of the present invention additionally provides a kind of equipment 300, and equipment 300 includes at least one processor
301st, memory 302 and communication interface 303；At least one processor 301, the memory 302 and the communication interface
303 are all connected by bus 304；
The memory 302, for storing computer executed instructions；
At least one processor 301, for executing the computer executed instructions of the storage of the memory 302 so that
The equipment 300 carries out data interaction by the communication interface 303 and storage system to execute the offer of said method embodiment
Carry file system method, or cause the equipment 300 to carry out data by the communication interface 303 and storage system
The interactive some or all of function to realize the device of carry file system.
At least one processor 301, can include different types of processor 301, or the process including same type
Device 301；Processor 301 can be any one following：Central processing unit (Central Processing Unit, referred to as
CPU), arm processor, field programmable gate array (Field Programmable Gate Array, abbreviation FPGA), special
Processor etc. is with the device for calculating disposal ability.A kind of optional embodiment, at least one processor 301 can also collect
Become many-core processor.
Memory 302 can be any one following or any one combination：Random access memory (Random Access
Memory, abbreviation RAM), read-only storage (read only memory, abbreviation ROM), nonvolatile memory (non-
Volatile memory, abbreviation NVM), solid state hard disc (Solid State Drives, abbreviation SSD), mechanical hard disk, disk,
The storage mediums such as disk permutation.
Communication interface 303 is used for equipment 300 carries out data friendship with other equipment (memory node in such as storage system)
Mutually.Communication interface 303 can be any one following or any one combination：Network interface (such as Ethernet interface), wireless network
The device with network access facility such as card.
The bus 304 can include address bus, data/address bus, controlling bus etc., and for ease of representing, Fig. 4 is thick with one
Line represents the bus.Bus 304 can be any one following or any one combination：Industry standard architecture (Industry
Standard Architecture, abbreviation ISA) bus, peripheral component interconnection (Peripheral Component
Interconnect, abbreviation PCI) bus, EISA (Extended Industry Standard
Architecture, abbreviation EISA) wired data transfer such as bus device.
Those skilled in the art can be understood that, for convenience and simplicity of description, the system of foregoing description,
Device and the specific work process of unit, may be referred to the corresponding process in preceding method embodiment, will not be described here.
In several embodiments provided herein, it should be understood that disclosed system, apparatus and method are permissible
Realize by another way.For example, device embodiment described above is only schematic, for example, the unit
Divide, only a kind of division of logic function, when actually realizing, can have other dividing mode, for example multiple units or component
Can in conjunction with or be desirably integrated into another system, or some features can be ignored, or not execute.Another, shown or
The coupling each other for discussing or direct-coupling or communication connection can be the indirect couplings of device or unit by some interfaces
Close or communicate to connect, can be electrical, mechanical or other forms.
The unit that illustrates as separating component can be or may not be physically separate, aobvious as unit
The part for showing can be or may not be physical location, you can be located at a place, or can also be distributed to multiple
On NE.Some or all of unit therein can be selected according to the actual needs to realize the mesh of this embodiment scheme
's.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it is also possible to
It is that unit is individually physically present, it is also possible to which two or more units are integrated in a unit.Above-mentioned integrated list
Unit both can be realized in the form of hardware, it would however also be possible to employ the form of SFU software functional unit is realized.
If the integrated unit is realized and as independent production marketing or use using in the form of SFU software functional unit
When, can be stored in a computer read/write memory medium.Based on such understanding, technical scheme is substantially
The part in other words prior art being contributed or all or part of the technical scheme can be in the form of software products
Embody, the computer software product is stored in a storage medium, use so that a computer including some instructions
Equipment (can be personal computer, server, or network equipment etc.) executes the complete of each embodiment methods described of the present invention
Portion or part steps.And aforesaid storage medium includes：USB flash disk, portable hard drive, read-only storage (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), magnetic disc or CD etc. are various can store journey
The medium of sequence code.
The above, above example only in order to technical scheme to be described, rather than a limitation；Although with reference to front
State embodiment to be described in detail the present invention, it will be understood by those within the art that：Which still can be to front
State the technical scheme described in each embodiment to modify, or equivalent is carried out to which part technical characteristic；And these
Modification is replaced, and does not make the essence of appropriate technical solution depart from the spirit and scope of various embodiments of the present invention technical scheme.
Claims (15)
1. a kind of method of carry file system, it is characterised in that the file system is used for managing storage in storage system
File；Methods described includes：
The mark of multiple target storage node is obtained, the storage system includes multiple memory nodes；
According to the mark of the plurality of target storage node, the multiple communication connections with the plurality of target storage node are set up；
Using the plurality of communication connection carry file system.
2. method according to claim 1, it is characterised in that the mark for obtaining multiple target storage node it
Before, methods described includes：
In the memory node included from the storage system, determine that the memory node of any data block in the file destination that is stored with is made
For the target storage node.
3. method according to claim 1 and 2, it is characterised in that the mark according to the plurality of target storage node
Know, foundation is included with multiple communication connections of the plurality of target storage node：
According to the mark of the target storage node, the connecting strand of the plurality of target storage node is determined from join index storehouse
Draw；
According to the join index of the plurality of target storage node, the multiple communication link with the plurality of target storage node are set up
Connect.
4. the method according to any one of claims 1 to 3, it is characterised in that methods described also includes：
After file system described in carry, determine that at least one communicates to connect from the plurality of communication connection；
Corresponding target storage node is accessed based on described at least one communication connection to obtain at least one file destination.
5. method according to claim 4, it is characterised in that described determine from the plurality of communication connection described at least
One communication connection, specially：
At least one little communication connection of traffic load is determined from the plurality of communication connection.
6. method according to claim 5, it is characterised in that described determine traffic load from the plurality of communication connection
At least one little communication connection, specially：
From the plurality of target storage node, the first few memory node of communication connection number is determined；
From the plurality of communication connection, determine with the communication connection of first memory node as the little communication of traffic load
Connection.
7. the method according to any one of claims 1 to 3, it is characterised in that methods described also includes：
After file system described in carry, obtained based on the target storage node corresponding to the plurality of communication connection concurrent access
Take multiple file destinations.
8. a kind of device of carry file system, it is characterised in that the file system is used for managing storage in storage system
File；Described device includes：
Acquiring unit, for obtaining the mark of multiple target storage node, the storage system includes multiple memory nodes；
Connection unit, for the mark according to the plurality of target storage node, sets up and the plurality of target storage node
Multiple communication connections；
Carry unit, for using the plurality of communication connection carry file system.
9. device according to claim 8, it is characterised in that described device includes：
Determining unit, for before the mark that the acquiring unit obtains multiple target storage node, from the storage system
Including memory node in, determine the memory node of any data block in file destination of being stored with as target storage section
Point.
10. device according to claim 8 or claim 9, it is characterised in that the connection unit specifically for：
According to the mark of the target storage node, the connecting strand of the plurality of target storage node is determined from join index storehouse
Draw；
According to the join index of the plurality of target storage node, the multiple communication link with the plurality of target storage node are set up
Connect.
11. devices according to any one of claim 8 to 10, it is characterised in that
The determining unit is additionally operable to, after file system described in the carry unit carry, from the plurality of communication connection
Determine at least one communication connection；
The acquiring unit is additionally operable to, and accesses corresponding target storage node based on described at least one communication connection to obtain
At least one file destination.
12. devices according to claim 11, it is characterised in that the determining unit specifically for：
At least one little communication connection of traffic load is determined from the plurality of communication connection.
13. devices according to claim 12, it is characterised in that the determining unit specifically for：
From the plurality of target storage node, the first few memory node of communication connection number is determined；
From the plurality of communication connection, determine with the communication connection of first memory node as the little communication of traffic load
Connection.
14. devices according to any one of claim 8 to 10, it is characterised in that the acquiring unit is additionally operable to：
After file system described in carry, obtained based on the target storage node corresponding to the plurality of communication connection concurrent access
Take multiple file destinations.
15. a kind of equipment, it is characterised in that including at least one processor, memory and communication interface；At described at least one
Reason device, the memory and the communication interface are all connected by bus；
The memory, for storing computer executed instructions；
At least one processor, for executing the computer executed instructions of the memory storage so that the equipment leads to
Crossing the communication interface and storage system carries out data interaction and comes the carry file system described in perform claim 1 to 7 any one of requirement
The method of system.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610819478.6A CN106484322A (en) | 2016-09-12 | 2016-09-12 | A kind of method of carry file system, device and equipment |
| PCT/CN2017/097681 WO2018045860A1 (en) | 2016-09-12 | 2017-08-16 | File system mounting method, apparatus and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610819478.6A CN106484322A (en) | 2016-09-12 | 2016-09-12 | A kind of method of carry file system, device and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106484322A true CN106484322A (en) | 2017-03-08 |
Family
ID=58273487
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610819478.6A Pending CN106484322A (en) | 2016-09-12 | 2016-09-12 | A kind of method of carry file system, device and equipment |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN106484322A (en) |
| WO (1) | WO2018045860A1 (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018045860A1 (en) * | 2016-09-12 | 2018-03-15 | 华为技术有限公司 | File system mounting method, apparatus and equipment |
| CN109460182A (en) * | 2017-09-06 | 2019-03-12 | 杭州海康威视系统技术有限公司 | A kind of storage of data, read method and device |
| CN110045929A (en) * | 2019-04-19 | 2019-07-23 | 苏州浪潮智能科技有限公司 | A kind of mount point management method and relevant apparatus |
| CN111078658A (en) * | 2019-10-31 | 2020-04-28 | 苏州浪潮智能科技有限公司 | Method and device for mounting cluster file system |
| CN111859414A (en) * | 2020-06-18 | 2020-10-30 | 厦门亿联网络技术股份有限公司 | Mounting method and device of file system and storage medium |
| CN111988409A (en) * | 2020-08-20 | 2020-11-24 | 深圳市瑞驰信息技术有限公司 | Method and system for realizing external storage starting of cloud mobile phone mounting |
| CN112492011A (en) * | 2020-11-19 | 2021-03-12 | 苏州浪潮智能科技有限公司 | Distributed storage system fault switching method, system, terminal and storage medium |
| CN113157647A (en) * | 2021-03-29 | 2021-07-23 | 普联技术有限公司 | Method and device for accessing mounted file |
| CN113342281A (en) * | 2021-06-28 | 2021-09-03 | 深圳市瑞驰信息技术有限公司 | Method for realizing mounting in android system based on rbd storage |
| CN113835613A (en) * | 2020-06-24 | 2021-12-24 | 浙江宇视科技有限公司 | File reading method and device, electronic equipment and storage medium |
| CN114124651A (en) * | 2022-01-26 | 2022-03-01 | 苏州浪潮智能科技有限公司 | Control method and device for cloud platform management node and cloud platform server cluster |
| CN113157647B (en) * | 2021-03-29 | 2024-04-26 | 普联技术有限公司 | Method and device for accessing mounted file |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112291292B (en) * | 2019-07-26 | 2023-04-07 | 浙江宇视科技有限公司 | Data storage method, device, equipment and medium |
| CN110688073B (en) * | 2019-10-10 | 2023-07-14 | 深圳市网心科技有限公司 | Data storage method, electronic device, system and medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040078465A1 (en) * | 2002-10-17 | 2004-04-22 | Coates Joshua L. | Methods and apparatus for load balancing storage nodes in a distributed stroage area network system |
| CN101510223A (en) * | 2009-04-03 | 2009-08-19 | 成都市华为赛门铁克科技有限公司 | Data processing method and system |
| CN101567003A (en) * | 2009-05-27 | 2009-10-28 | 清华大学 | Method for managing and allocating resource in parallel file system |
| CN101699825A (en) * | 2009-10-28 | 2010-04-28 | 成都市华为赛门铁克科技有限公司 | Network file system access method, device and system |
| CN104050248A (en) * | 2014-06-05 | 2014-09-17 | 北京捷成世纪科技股份有限公司 | File storage system and storage method |
| CN105677703A (en) * | 2015-12-25 | 2016-06-15 | 曙光云计算技术有限公司 | NAS file system, and access method and apparatus thereof |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140222879A1 (en) * | 2013-02-01 | 2014-08-07 | Apple Inc. | Method and system for unmounting an inaccessible network file system |
| CN106484322A (en) * | 2016-09-12 | 2017-03-08 | 华为技术有限公司 | A kind of method of carry file system, device and equipment |
-
2016
- 2016-09-12 CN CN201610819478.6A patent/CN106484322A/en active Pending
-
2017
- 2017-08-16 WO PCT/CN2017/097681 patent/WO2018045860A1/en active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040078465A1 (en) * | 2002-10-17 | 2004-04-22 | Coates Joshua L. | Methods and apparatus for load balancing storage nodes in a distributed stroage area network system |
| CN101510223A (en) * | 2009-04-03 | 2009-08-19 | 成都市华为赛门铁克科技有限公司 | Data processing method and system |
| CN101567003A (en) * | 2009-05-27 | 2009-10-28 | 清华大学 | Method for managing and allocating resource in parallel file system |
| CN101699825A (en) * | 2009-10-28 | 2010-04-28 | 成都市华为赛门铁克科技有限公司 | Network file system access method, device and system |
| CN104050248A (en) * | 2014-06-05 | 2014-09-17 | 北京捷成世纪科技股份有限公司 | File storage system and storage method |
| CN105677703A (en) * | 2015-12-25 | 2016-06-15 | 曙光云计算技术有限公司 | NAS file system, and access method and apparatus thereof |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018045860A1 (en) * | 2016-09-12 | 2018-03-15 | 华为技术有限公司 | File system mounting method, apparatus and equipment |
| CN109460182A (en) * | 2017-09-06 | 2019-03-12 | 杭州海康威视系统技术有限公司 | A kind of storage of data, read method and device |
| CN110045929B (en) * | 2019-04-19 | 2020-06-19 | 苏州浪潮智能科技有限公司 | Mounting point management method and related device |
| CN110045929A (en) * | 2019-04-19 | 2019-07-23 | 苏州浪潮智能科技有限公司 | A kind of mount point management method and relevant apparatus |
| CN111078658B (en) * | 2019-10-31 | 2022-08-12 | 苏州浪潮智能科技有限公司 | Method and device for mounting cluster file system |
| CN111078658A (en) * | 2019-10-31 | 2020-04-28 | 苏州浪潮智能科技有限公司 | Method and device for mounting cluster file system |
| CN111859414A (en) * | 2020-06-18 | 2020-10-30 | 厦门亿联网络技术股份有限公司 | Mounting method and device of file system and storage medium |
| CN113835613A (en) * | 2020-06-24 | 2021-12-24 | 浙江宇视科技有限公司 | File reading method and device, electronic equipment and storage medium |
| CN113835613B (en) * | 2020-06-24 | 2024-03-29 | 浙江宇视科技有限公司 | File reading method and device, electronic equipment and storage medium |
| CN111988409A (en) * | 2020-08-20 | 2020-11-24 | 深圳市瑞驰信息技术有限公司 | Method and system for realizing external storage starting of cloud mobile phone mounting |
| CN111988409B (en) * | 2020-08-20 | 2023-06-13 | 深圳市臂云科技有限公司 | Method and system for realizing cloud mobile phone mounted external storage starting |
| CN112492011A (en) * | 2020-11-19 | 2021-03-12 | 苏州浪潮智能科技有限公司 | Distributed storage system fault switching method, system, terminal and storage medium |
| CN113157647A (en) * | 2021-03-29 | 2021-07-23 | 普联技术有限公司 | Method and device for accessing mounted file |
| CN113157647B (en) * | 2021-03-29 | 2024-04-26 | 普联技术有限公司 | Method and device for accessing mounted file |
| CN113342281A (en) * | 2021-06-28 | 2021-09-03 | 深圳市瑞驰信息技术有限公司 | Method for realizing mounting in android system based on rbd storage |
| CN114124651A (en) * | 2022-01-26 | 2022-03-01 | 苏州浪潮智能科技有限公司 | Control method and device for cloud platform management node and cloud platform server cluster |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2018045860A1 (en) | 2018-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106484322A (en) | A kind of method of carry file system, device and equipment | |
| CN102918808B (en) | The method of transfer of virtual LAN Information in the discovery of phantom station interface and configuration protocol | |
| US9330156B2 (en) | System and method for software defined network aware data replication | |
| US5321815A (en) | Route selection using cached partial trees in a data communications network | |
| US20200092084A1 (en) | System and methods for operating a blockchain network | |
| US9983813B2 (en) | Maintenance of a fabric priority during synchronous copy operations | |
| JP6275850B2 (en) | Method and apparatus for processing operation requests in a storage system | |
| CN107079060A (en) | The system and method optimized for carrier-class NAT | |
| CN106164898B (en) | Data processing method and device | |
| CN105493445A (en) | Regional firewall clustering in a networked computing environment | |
| CN112134741A (en) | Client-directed networking restrictions in distributed systems | |
| CN107819891A (en) | Data processing method, device, computer equipment and storage medium | |
| US20160337726A1 (en) | Data switching system, method for sending data traffic, and switching apparatus | |
| CN104092758B (en) | A kind of read method of distributed type high speed cloud storage service device group system | |
| CN104811392A (en) | Method and system for processing resource access request in network | |
| CN104954275A (en) | Systems and methods for multipath load balancing | |
| CN105867843A (en) | Data transmission method and device | |
| CN107707628A (en) | Method and apparatus for transmitting data processing request | |
| CN104508651A (en) | Connection mesh in mirroring asymmetric clustered multiprocessor systems | |
| US10313470B2 (en) | Hierarchical caching and analytics | |
| Long | Storage Networking Protocol Fundamentals | |
| US9544371B1 (en) | Method to discover multiple paths to disk devices cluster wide | |
| CN106686117A (en) | Distributed calculation cluster data storage processing system and method | |
| US9614910B2 (en) | Maintenance of a fabric priority among operations in related fibre channel exchanges | |
| CN104205730A (en) | Network element data access method, device and network management system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |