CN109976667A - A kind of mirror image management method, apparatus and system - Google Patents
A kind of mirror image management method, apparatus and system Download PDFInfo
- Publication number
- CN109976667A CN109976667A CN201910145316.2A CN201910145316A CN109976667A CN 109976667 A CN109976667 A CN 109976667A CN 201910145316 A CN201910145316 A CN 201910145316A CN 109976667 A CN109976667 A CN 109976667A
- Authority
- CN
- China
- Prior art keywords
- node
- mirror image
- metadata
- warehouse server
- image data
- 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.)
- Granted
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
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0608—Saving storage space on 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
- G06F3/0638—Organizing or formatting or addressing of data
-
- 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]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1095—Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/45562—Creating, deleting, cloning virtual machine instances
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/45587—Isolation or security of virtual machine instances
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Transfer Between Computers (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Processing Or Creating Images (AREA)
Abstract
The embodiment of the present invention provides a kind of mirror image management method, apparatus and system, is related to field of computer technology, can reduce the number of transmissions of mirror image data, and improve the utilization rate of memory space, to reduce the expense of docker system.The system includes: warehouse server and N number of node for being connected with the warehouse server, and the memory module of N number of node constitutes distributed memory system, and the mirror image data of mirror image is shared in the distributed memory system in N number of node；Wherein, first node is used for: generating the first mirror image, which includes the first metadata and the first mirror image data；First mirror image data is saved to the distributed memory system；And upload request is sent to the warehouse server, first metadata is carried in the upload request, which is any of N number of node；The warehouse server, is used for: receiving the upload request of first node transmission；And save first metadata in the upload request.
Description
Technical field
The present invention relates to field of computer technology more particularly to a kind of mirror image management methods, apparatus and system.
Background technique
Docker is an application hosting frame (application hosting framework), it can pass through class
Application is disposed, managed like the same container of virtual machine.
In current docker system, docker registry (i.e. warehouse server) of each host with distal end
It is connected, each host needs the mirror image needed for downloading (pull) in warehouse server when carrying out docker instantiation operation
(i.e. image), and be stored in ware-house here；Correspondingly, if host is modified mirror image, when obtaining new mirror image,
Also need to upload new mirror image (push) to warehouse server.
Specifically, mirror image can bedding storage be in warehouse server in the form of a tree, the branch of each node to root node
Road constitutes a mirror image, binary tree as shown in Figure 1, comprising: the mirror image 1 that node 1 and node 2 form, node 1,2 and of node
Multiple mirror images such as the mirror image 3 that the mirror image 2 that node 4 forms, node 1 and node 3 form, and mirror image number specifically can be used in each mirror image
It is described according to metadata, wherein mirror image data contains the data of a complete (SuSE) Linux OS, and metadata is logical
Often retouched by a kind of JSON (JavaScript Object Notation, data interchange format of a lightweight) file
It states, for characterizing the attributive character of the mirror image, such as the information such as ID of the title of mirror image, leaf node.
In docker system, the number of mirror image can be very more, therefore, when multiple main frames download mirror image, should download
The mirror image data of mirror image will also download the metadata of mirror image, and the expense of entire docker system is caused to increase, also, different main
Machine may use identical mirror image, and in existing docker system, it is required to take from warehouse using the host of identical mirror image
The mirror image is downloaded in business device, and is saved to local storage space, i.e., saves identical mirror image in a large amount of hosts, cause host
The waste of memory space.
Summary of the invention
The embodiment of the present invention provides a kind of mirror image management method, apparatus and system, can reduce the transmission time of mirror image data
Number, and the utilization rate of memory space is improved, to reduce the expense of docker system.
In order to achieve the above objectives, the embodiment of the present invention adopts the following technical scheme that
In a first aspect, the embodiment of the present invention provides a kind of docker system, the system include: warehouse server and with
The memory module of the connected N number of node of the warehouse server, N number of node constitutes distributed memory system, N number of node scope
The mirror image data of picture is shared in the distributed memory system, and N is the integer more than or equal to 2；Wherein, first node is used for:
The first mirror image is generated, which includes the first metadata and the first mirror image data；First mirror image data is saved to this
Distributed memory system；And upload request is sent to the warehouse server, first metadata is carried in the upload request,
The first node is any of N number of node；The warehouse server, is used for: receiving the upload of first node transmission
Request；And save first metadata in the upload request.
As can be seen that since the mirror image data of mirror image is shared in the distributed storage system in N number of node in docker system
In system, therefore, in mirror image upload procedure, first node only need to be to the metadata that warehouse server uploads the new mirror image generated
Can, without uploading the mirror image data of new mirror image, so as to reduce the number of transmissions of mirror image data in docker system, and improves and deposit
The utilization rate for storing up space, reduces the expense of entire docker system.
In a kind of possible design, which includes warehouse driving；Wherein, the first node, be also used to
The warehouse server sends first mirror image data；The warehouse server, is also used to: receive the first node transmission this first
Mirror image data；And warehouse driving is called to save first mirror image data to the distributed memory system.
In a kind of possible design, which includes mirror image management module, wherein the first node is also used to:
Downloading request is sent to the warehouse server by the mirror image management module, the mark of the first metadata is carried in downloading request
Know；Mould is managed by the mirror image and receives first metadata of warehouse server transmission, and searches the distributed memory system
In whether be stored with the first mirror image data corresponding with first metadata；If be stored in the distributed memory system this first
Mirror image data then obtains first mirror image data by the mirror image management module from the distributed memory system；Warehouse clothes
Business device, is also used to: receiving the downloading request of first node transmission；According to the mark of the first metadata in downloading request
Search first metadata；And first metadata is sent to the first node.
So, since the mirror image data of mirror image is shared in the distributed storage system in N number of node in docker system
In system, therefore, during image download, only need to transmit the metadata of mirror image between first node and warehouse server, and
The mirror image data for not needing transmission mirror image, can reduce the number of transmissions of mirror image data, and improve the utilization rate of memory space, thus
Reduce the expense of docker system.
In a kind of possible design, which is also used to: will be in first meta-data preservation to the first node
Ware-house here；And add 1 operation to executing for the preset reference counter of the first mirror image.
In a kind of possible design, which is also used to: deleting first metadata in the first node；
To the reference counter execution operation that subtracts 1.
In a kind of possible design, which is also used to send removal request, the deletion to the warehouse server
Request is used to indicate the first mirror image deleted in the warehouse server；The warehouse server, is also used to: receiving first node hair
The removal request sent；According to the removal request, first metadata in the warehouse server is deleted；If each node is equal
First mirror image is not used, then deletes first mirror image data being stored in the distributed memory system.
It so far, can mirror image data to mirror image and metadata difference in docker system provided in an embodiment of the present invention
It is managed, the mirror image data of mirror image is uniformly stored in distributed memory system in each node, that is to say, that each node can
It, so, not only can be to avoid being protected in great deal of nodes in the prior art to share the mirror image data in distributed memory system
The problem of depositing identical mirror image and leading to the waste of memory space, also makes each node and warehouse server only need to manage
The metadata of mirror image in itself reduces the expense of docker system so as to reduce the number of transmissions of mirror image data.
Second aspect, the embodiment of the present invention provide a kind of mirror image management method, and this method is applied in first aspect
Any one docker system, this method comprises: first node generates the first mirror image, which includes the first metadata and the
One mirror image data, the first node are any of N number of node in the docker system；The first node is to warehouse server
Upload request is sent, first metadata is carried in the upload request, so that the warehouse server saves first yuan of number
According to；The first node saves first mirror image data into the distributed memory system.
In a kind of possible design, which saves first mirror image data to the distributed memory system
In, comprising: the first node sends first mirror image data to the warehouse server so that the warehouse server by this first
Mirror image data is saved into the distributed memory system.
In a kind of possible design, this method further include: the first node sends downloading request to the warehouse server,
The mark of the first metadata is carried in downloading request；The first node receive the warehouse server transmission with this first yuan
Corresponding first metadata of the mark of data；The first node is searched in the distributed memory system according to first metadata
Whether with first metadata corresponding first mirror image data is stored with；If being stored with first mirror in the distributed memory system
As data, then the first node obtains first mirror image data from the distributed memory system.
In a kind of possible design, in the first metadata that the first node is sent according to the warehouse server, search
Whether it is stored in the distributed memory system after the first mirror image data corresponding with first metadata, this method is also wrapped
Include: if not stored in the distributed memory system have first mirror image data, which obtains from the warehouse server
Take first mirror image data.
In a kind of possible design, the first node receive the warehouse server transmission the first metadata it
Afterwards, this method further include: the first node is by the ware-house here in first meta-data preservation to the first node.
In a kind of possible design, in the first node by the local in first meta-data preservation to the first node
After warehouse, this method further include: the first node is to the reference counter execution plus 1 operation preset for the first mirror image.
In a kind of possible design, if the first node determines the first mirror image deleted in the first node, the party
Method further include: the first node deletes first metadata in the first node；The first node holds the reference counter
Operation that row subtracts 1.
The third aspect, the embodiment of the present invention provide a kind of mirror image management method, and this method is applied in first aspect
Any one docker system, this method comprises: warehouse server receives the removal request that first node is sent, which is used
The first mirror image in the warehouse server is deleted in instruction, which includes the first metadata and the first mirror image data, is somebody's turn to do
First node is any of N number of node in the docker system；The warehouse server deletes the storehouse according to the removal request
First metadata in the server of library；If first mirror image is not used in each node, warehouse server deletion is deposited
Store up first mirror image data in the distributed memory system.
In a kind of possible design, if first mirror image is not used in each node, which is deleted
The first mirror image data being stored in the distributed memory system, comprising: the warehouse server judges the reference of first mirror image
Whether counter is 0；If the reference counter of first mirror image is 0, warehouse server deletion is stored in the distribution and deposits
First mirror image data in storage system.
Fourth aspect, the embodiment of the present invention provide a kind of node, which is applied to any one of first aspect
Docker system, the node include: mirror image generation unit, and for generating the first mirror image, which includes the first metadata
With the first mirror image data；Transmission unit, for sending upload request to warehouse server, carry in the upload request this first
Metadata, so that the warehouse server saves first metadata；Storage unit, for by first mirror image data save to
In the distributed memory system.
In a kind of possible design, which is also used to send first mirror image data to the warehouse server,
So that the warehouse server saves first mirror image data into the distributed memory system.
In a kind of possible design, which further includes acquiring unit and searching unit, wherein the transmission unit, also
For sending downloading request to the warehouse server, the mark of the first metadata is carried in downloading request；The acquiring unit,
For receiving the first metadata corresponding with the mark of first metadata of warehouse server transmission；And if the distribution
It is stored with first mirror image data in storage system, then obtains first mirror image data from the distributed memory system；This is looked into
Unit is looked for, for searching and whether being stored in the distributed memory system and first metadata pair according to first metadata
The first mirror image data answered.
In a kind of possible design, the acquiring unit, if be also used in the distributed memory system it is not stored have this
One mirror image data then obtains first mirror image data from the warehouse server.
In a kind of possible design, which is also used in first meta-data preservation to the first node
Ware-house here.
In a kind of possible design, the node further include: counting unit, by based on for the preset reference of the first mirror image
Number device executes plus 1 operation.
In a kind of possible design, which further includes deleting unit, wherein the deletion unit, for delete this
First metadata in one node；The counting unit is also used to subtract 1 to the reference counter execution operation.
5th aspect, the embodiment of the present invention provide a kind of warehouse server, which is applied to first aspect
Any one of docker system, the warehouse server include: receiving unit, for receive first node transmission deletion ask
Ask, the removal request be used to indicate delete the warehouse server in the first mirror image, first mirror image include the first metadata and
First mirror image data, the first node are any of N number of node in the docker system；Unit is deleted, it should for basis
Removal request deletes first metadata in the warehouse server；If first mirror image is not used in each node, delete
Except first mirror image data being stored in the distributed memory system.
In a kind of possible design, which further includes judging unit, wherein the judging unit, for sentencing
Whether the reference counter of first mirror image of breaking is 0；The deletion unit, if being specifically for the reference counter of first mirror image
0, then delete first mirror image data being stored in the distributed memory system.
6th aspect, the embodiment of the present invention provide a kind of node, comprising: processor, memory, bus and communication connect
Mouthful；For storing computer executed instructions, which is connect with the memory by the bus memory, when the node is transported
When row, which executes the computer executed instructions of memory storage, so that the node is executed as appointed in second aspect
The mirror image management method of meaning one.
7th aspect, the embodiment of the present invention provides a kind of warehouse server, comprising: processor, memory, bus and logical
Believe interface；The memory is for storing computer executed instructions, which is connect with the memory by the bus, when the storehouse
When library server is run, which executes the computer executed instructions of memory storage, so that the warehouse server is held
The mirror image management method of row any one of such as third aspect.
Eighth aspect, the embodiment of the invention provides a kind of computer storage mediums, for being stored as used in above-mentioned node
Computer software instructions, it includes for execute above-mentioned aspect be node designed by program.
9th aspect, the embodiment of the invention provides a kind of computer storage mediums, for being stored as above-mentioned repository service
Computer software instructions used in device, it includes be program designed by warehouse server for executing above-mentioned aspect.
In the present invention, the name of node and warehouse server does not constitute restriction to equipment itself, in practical implementations, this
A little equipment can occur with other titles.As long as the function of each equipment is similar with the present invention, belong to the claims in the present invention and
Within the scope of its equivalent technologies.
In addition, second aspect technical effect brought by any possible design method into the 9th aspect can be found in the
Technical effect brought by different designs mode in one side, details are not described herein again.
The aspects of the invention or other aspects can more straightforwards in the following description.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described.
Fig. 1 is the schematic diagram of mirror image in the prior art；
Fig. 2 is a kind of configuration diagram one of docker system provided in an embodiment of the present invention；
Fig. 3 is a kind of configuration diagram two of docker system provided in an embodiment of the present invention；
Fig. 4 is a kind of configuration diagram three of docker system provided in an embodiment of the present invention；
Fig. 5 is the architecture diagram of docker system interior joint in the prior art；
Fig. 6 is the architecture diagram of docker system interior joint provided in an embodiment of the present invention；
Fig. 7 is a kind of flow diagram of image download method provided in an embodiment of the present invention；
Fig. 8 is a kind of flow diagram of mirror image method for uploading provided in an embodiment of the present invention；
Fig. 9 is a kind of flow diagram of mirror image delet method provided in an embodiment of the present invention；
Figure 10 is a kind of structural schematic diagram one of node provided in an embodiment of the present invention；
Figure 11 is a kind of structural schematic diagram two of node provided in an embodiment of the present invention；
Figure 12 is a kind of structural schematic diagram three of node provided in an embodiment of the present invention；
Figure 13 is a kind of structural schematic diagram four of node provided in an embodiment of the present invention；
Figure 14 is a kind of structural schematic diagram one of warehouse server provided in an embodiment of the present invention；
Figure 15 is a kind of structural schematic diagram two of warehouse server provided in an embodiment of the present invention；
Figure 16 is a kind of structural schematic diagram of computer equipment provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.
In addition, term " first ", " second " are used for descriptive purposes only and cannot be understood as indicating or suggesting relative importance
Or implicitly indicate the quantity of indicated technical characteristic.Define " first " as a result, the feature of " second " can be expressed or
Implicitly include one or more of the features.In the description of the present invention, unless otherwise indicated, the meaning of " plurality " is two
It is a or more than two.
Mirror image management method provided in an embodiment of the present invention, applied to docker system as shown in Figure 2, the docker system
It include warehouse server 01 in system, and N (N is more than or equal to 2 integer) a node being connected with warehouse server 01
02, wherein warehouse server 01 is the warehouse of a memory image, and node 02 can obtain required from warehouse server 01
Mirror image, wherein each mirror image specifically includes mirror image data and metadata.The mirror image data of mirror image can be unified to store in node 02
In distributed memory system 03, i.e., each node 02 shares the mirror image data stored in distributed memory system 03, and each section
The metadata of mirror image is saved and is safeguarded by each node 02 oneself in point 02.
Wherein, mirror image data contains the data of a complete (SuSE) Linux OS, and metadata usually passes through one
A kind of JSON (JavaScript Object Notation, data interchange format of lightweight) file describes, for characterizing
The attributive character of the mirror image, such as the information such as ID of the title of mirror image, leaf node.
It should be noted that in present example, the arbitrary node 02 in above-mentioned distributed memory system may be by one
A entity device is realized, such as computer, portable mobile termianl or server, it is also possible to jointly real by multiple entity devices
It is existing.It is understood that any functional node 02 in distributed memory system all may be a logic in entity device
Functional module a, it is also possible to logic functional block being made of multiple entity devices.Therefore, implement in following present invention
In example, each step in method provided in an embodiment of the present invention can be executed by an entity device, it can be by multiple entities
Equipment cooperation executes each step in method provided in an embodiment of the present invention, is not construed as limiting to this present invention.
In addition, the specific structure of docker system interior nodes 02 will elaborate in subsequent embodiment, therefore herein no longer
It repeats.
Illustratively, as shown in figure 3, being stored with the metadata 1 of mirror image 1 in first node, mirror is stored in second node
As 1 metadata 1 and the metadata 2 of mirror image 2, and the mirror image data 2 of the mirror image data 1 of mirror image 1 and mirror image 2 is uniformly stored in
In distributed memory system 03, then, when first node 1 needs to download the mirror image data of (pull) mirror image 2, first node is only
The metadata 2 of mirror image 2 need to be obtained from warehouse server 01 and is saved to first node local, and then according to the metadata of mirror image 22
The mirror image data 2 of mirror image 2 is read from distributed memory system 03.
As can be seen that being based on above-mentioned docker system, node 02 is in the required mirror image of downloading, it is only necessary to from repository service
Device 01 obtain the mirror image metadata, without as the prior art simultaneously from warehouse server 01 obtain mirror image metadata with
Mirror image data, so as to reduce the expense of docker system.
Simultaneously as the mirror image data of mirror image is uniformly stored in distributed memory system 03 in each node 02, then,
The mirror image data of each mirror image be only used in distributed memory system 03 save it is a, so as to avoid in the prior art, largely
The problem of saving identical mirror image in node and leading to the waste of memory space.
Certainly, similarly to the prior art, the part of nodes 02 in docker system shown in Fig. 2, can also be by mirror
The mirror image data of picture is stored in warehouse server 01, at this point, this part of nodes 02 still can be according to the method for the prior art, Xiang Cang
Library server 01 uploads mirror image data and metadata, and downloads mirror image data and metadata, the present invention from warehouse server 01
This is not intended to be limited in any.
Further, the storage unit in warehouse server 01 for memory image data also can be set deposits in distribution
In storage system 03, that is to say, that as shown in figure 4,04 can be driven by setting warehouse, the mirror image that will be stored in warehouse server 01
Data are saved into distributed memory system 03, so, when node 02 generates some mirror image, due to the mirror image number of the mirror image
According to can directly be generated in distributed memory system 03, therefore, when node 02 uploads (push) mirror image to warehouse server 01, only
The metadata of mirror image need to be sent to warehouse server 01, so that node 02 is avoided to upload mirror image data to warehouse server 01,
It can be further reduced the number of transmissions of mirror image data, to reduce the expense of docker system.
Further, as shown in figure 5, being the architecture diagram of docker system interior joint 02.Wherein, docker system interior joint
02 specifically includes: client 11 (i.e. docker client), server 12, driving 13 (i.e. docker driver), container tube
Manage library 14 (i.e. libcontainer), container 15 (i.e. Docker container) and the bottom module 16 for supporting above-mentioned framework
(i.e. rootfs).
Wherein, client 11 (i.e. docker client) is that user is used in docker framework and server-side 12 is established and led to
The client of letter.After user sends Container Management request using docker order line by client 11, received by server 12
And handle Container Management request.
Server-side 12 (i.e. docker daemon) specifically may include the runtime engine in docker framework, i.e. docker
Engine and special service are in the docker server of client 11.
Wherein, docker server can receive and dispatch the various requests of the transmission of distribution customer terminal 11, specifically include
HTTP server, router (Router) and processing module (Handler).
And docker engine is responsible for the life cycle management of mirror image and container, such as mirror image downloading (pull) and
Upload (push) operation and instantiation operation of container etc..In addition, docker engine can be with the storehouse in Fig. 2 or Fig. 4
Library server 01 is communicated.
13 (i.e. docker driver) are driven, the setting to docker container performing environment may be implemented.Due to
In the life cycle of docker operation, and all operation of non-user is both for the management of docker container, in addition also related
In the acquisition etc. of docker operation information, therefore, in order to by the management of Docker container from the business inside docker engine
It is distinguished in logic, the management request of the non-docker container in this part is managed provided with driving 13.Wherein, in docker
In the realization of driver, can be divided into following three classes driving: graphdriver (figure driving), (network drives networkdriver
It is dynamic) and execdriver (external drive).
Container Management library 14 (i.e. libcontainer) is a library realized using Go language design in docker framework,
By calling libcontainer, namespace, cgroups, apparmor, the network equipment and the fire prevention of container can be manipulated
Wall rule etc..
Container 15 (i.e. docker container) is the final avatar of service offering in docker framework, can be according to
The demand and instruction of user, customized corresponding Docker container.
Bottom module 16, i.e. rootfs (root file system), comprising in typical linux system standard directories and
File, also referred to as root file system.
Based on above-mentioned docker architecture diagram shown in fig. 5, as shown in fig. 6, node provided in an embodiment of the present invention 02 also increases
If mirror image management module 100 (i.e. Images Manager), mirror image management module 100 can take with server-side 12 and warehouse respectively
Business device 01 is communicated, and the agency of warehouse server 01 is accessed as client 11.
In addition, mirror image management module 100 is also used to safeguard that the metadata of the mirror image on node 02 and maintenance image are being divided
Whether the reference number in cloth storage system 03, the mirror image data for judging that node 02 needs are deposited in distributed memory system 03
Wait.
Hereinafter, the docker system that will be provided in conjunction with Fig. 2-Fig. 6, elaborates a kind of mirror image provided in an embodiment of the present invention
Management method, wherein mirror image management method provided in an embodiment of the present invention specifically can be related to downloading (pull) method of mirror image, mirror
Upload (push) method of picture and deletion (delete) method of mirror image will be introduced one by one in following examples.
As shown in fig. 7, the embodiment of the present invention provides a kind of image download method, comprising:
101, first node sends downloading request to warehouse server, and the mark of the first metadata is carried in downloading request
Know.
102, warehouse server searches the first metadata corresponding with the mark of the first metadata according to downloading request.
103, first node receives the first metadata that warehouse server is sent,
104, first node is by the ware-house here in first meta-data preservation to first node.
105, first node will execute for the preset reference counter of the first mirror image plus 1 operates.
106, whether first node is searched and is stored with and this first yuan in distributed memory system according to first metadata
Corresponding first mirror image data of data.
If 107, being stored with the first mirror image data in distributed memory system, first node is from distributed memory system
It obtains this and takes the first mirror image data.
In a step 101, when first node (first node is any node in above-mentioned docker system) needs to make
With the first mirror image (first mirror image include the first mirror image data and the first metadata), and be not stored in first node this
When the first metadata of one mirror image, first node sends downloading request to warehouse server, carries first in downloading request
The mark of metadata, for example, the ID etc. of the first metadata.
For example, as shown in Fig. 2, be stored with the metadata of mirror image 1 in node 1, mirror image 1 and mirror image 2 are stored in node 2
Metadata, and the mirror image data of mirror image 1 and mirror image 2 is uniformly stored in distributed memory system 03, then, when node 1 needs
When wanting the mirror image data of mirror image 2, however, it is determined that the metadata of itself not memory image 2 then sends downloading to warehouse server and asks
It asks, the mark of the metadata of mirror image 2 is carried in downloading request.
Specifically, as shown in fig. 6, client 11 in first node generates downloading request, and it is sent to server-side 12, mirror
Picture 100 agency service end 12 of management module is communicated with warehouse server, sends downloading request to warehouse server.
In a step 102, it after the downloading that warehouse server receives first node transmission is requested, is taken according in downloading request
The mark of first metadata of band searches the first metadata corresponding with the mark of the first metadata.Wherein, in warehouse server
It is stored with the metadata of all mirror images in docker system.
Further, the first metadata found is sent to the mirror image management module of first node by warehouse server
100。
In step 103, mirror image management module 100 receives first metadata of warehouse server transmission.
In turn, at step 104, first node is by the ware-house here in first meta-data preservation to first node, into
And the first mirror image data is obtained by executing following step 105-106, the first mirror image is needed again in order to which first node is subsequent
It, can be by searching for the first metadata being locally stored, directly according to the title or leaf of the first mirror image in the first metadata when data
The attributes such as sub- ID obtain this from distributed memory system and take the first mirror image data.
In step 105, if mirror image management module 100 receive warehouse server transmission first metadata, first
Node can also execute the reference counter of the first mirror image plus 1 operation.
Specifically, can be the preset reference counter of each mirror image, for recording the mirror in docker system
The access times of picture, for the reference counter of some mirror image, which be may be provided in distributed memory system,
N number of node in docker system can share the reference counter, and when node executes image download operation, mirror image manages mould
Block 100 just carries out the reference counter of corresponding mirror image plus 1 operates, since the same mirror image can be made by multiple nodes simultaneously
With therefore, the reference counter of the mirror image may be used to indicate the access times of the mirror image, and when counter is 0, then explanation does not have
Node uses the mirror image.
It should be noted that step 104-105 is optional step, also, the execution sequence of step 104 and step 105 is not
It limits.
In step 106, first metadata that mirror image management module 100 is sent according to warehouse server, can be with service
Communication is established at end 12, and then searches in distributed memory system whether be stored with the first mirror image number corresponding with first metadata
According to.
Due to giving two kinds of docker systems as shown in Figure 2 and Figure 4 respectively in the embodiment of the present invention, it is based on
Docker system shown in Fig. 2, the storage unit due to being used for memory image data in warehouse server are not disposed on distribution
In storage system, therefore, the first mirror image data of the first mirror image needed for first node is possible to be stored in the distributed storage
In system, it is also possible to it is stored in warehouse server, at this point, after first node receives the first metadata, due to first yuan
The attributive character such as title, the label of the first mirror image data are carried in data, therefore, first node can be according to the first mirror image number
According to attributive character search distributed memory system in whether be stored with the first mirror image data corresponding with first metadata.
Further, if being stored with the first mirror image data, in step 107, first node in distributed memory system
This is obtained from the distributed memory system and takes the first mirror image data, if not stored in distributed memory system have the first mirror image number
According to then first node can obtain the first mirror image data corresponding with first metadata out of warehouse server.
In addition, docker system based on shown in Fig. 4, due to being used for the storage list of memory image data in warehouse server
Member also can be set in distributed memory system, that is to say, that the mirror image data and each node stored in warehouse server
Interior mirror image data, is respectively positioned in distributed memory system, therefore, can direct root after first node receives the first metadata
This is obtained from the distributed memory system according to the attributive character of the first mirror image data carried in the first metadata takes the first mirror image
Data.
So far, it is based on above-mentioned docker system, the embodiment of the present invention provides a kind of image download method, wherein first
Node sends downloading request to warehouse server, the mark of the first metadata is carried in downloading request, so as to warehouse
Server searches the first metadata corresponding with the mark of the first metadata according to downloading request；In turn, if receiving warehouse clothes
The first metadata that business device is sent, then first node is obtained from distributed memory system and first yuan of number according to the first metadata
According to corresponding first mirror image data, so, during image download, due to mirror image in each node in docker system
Mirror image data is stored in distributed memory system, therefore, only needs to transmit mirror image between first node and warehouse server
Metadata, without transmit mirror image mirror image data, the number of transmissions of mirror image data can be reduced, and improve memory space
Utilization rate, to reduce the expense of docker system.
In addition, the step 101-107 in based on the above embodiment, the embodiment of the present invention also provides a kind of mirror image deletion side
Method, this method are used to delete the mirror image in docker system in arbitrary node.
Illustratively, for deleting the first mirror image in above-described embodiment in first node, which includes the
One metadata and the first mirror image data.
Specifically, the client 11 in first node sends the removal request for deleting the first mirror image, service to server-side 12
The removal request is forwarded to mirror image management module 100 by end 12, and mirror image management module 100 deletes first segment according to the removal request
The first metadata stored in point, in turn, reference meter of the mirror image management module 100 to the first mirror image in distributed memory system
Operation that number device execution subtracts 1, to indicate that the access times of first mirror image subtract 1.
As can be seen that since the first mirror image can be used by multiple nodes simultaneously in distributed memory system, the
One node is not necessarily to delete the first mirror image data of the first mirror image, need to only delete the first metadata stored in first node, and will
The reference counter of first mirror image subtracts 1.
Further, as shown in figure 8, by taking above-mentioned first node uploads the first mirror image to warehouse server as an example, the present invention
Embodiment provides a kind of mirror image method for uploading, comprising:
201, first node generates the first mirror image, which includes the first mirror image data and the first metadata.
202, first node sends upload request to warehouse server, carries the first metadata in the upload request.
203, warehouse server saves first metadata.
204, (optional) warehouse server sends first to first node and uploads success message.
205, (optional) first node sends the first mirror image data to warehouse server.
206, (optional) warehouse server saves first mirror image data into distributed memory system.
207, (optional) warehouse server sends second to first node and uploads success message.
In step 201, first node (first node is any node in above-mentioned docker system) was being run
New mirror image is produced in journey, which is the first mirror image described in step 201, which specifically includes the first mirror
As data and the first metadata.
In step 202, after first node generates the first mirror image, first node sends to upload to warehouse server and ask
Ask, request by first mirrored storage into warehouse server, can be with when needing in order to other nodes using first mirror image
By sending downloading request first mirror image to warehouse server, wherein carried in the upload request that first node is sent
There is the first metadata of first mirror image.
What needs to be explained here is that the step of generating first mirror image can be executed by the server-side 12 in first node,
So, when the server-side of first node 12 is arranged in the distributed memory system shown in Fig. 2 or Fig. 4, server-side 12 is raw
At the first mirror image in include the first mirror image data be just stored directly in the distributed memory system, at this point, in step
In 203, after warehouse server receives first metadata that first node is sent, first metadata is directly saved.
Optionally, after warehouse server saves first metadata, in step 204, warehouse server can be with
First is sent to first node and uploads success message, to inform that first node first metadata uploads successfully.
As can be seen that in this case, first node need to only upload first number of the new mirror image generated to warehouse server
According to, without uploading the mirror image data of new mirror image, so as to reduce the number of transmissions of mirror image data in docker system, reduction
The expense of entire docker system.
Or when server-side 12 is not disposed in distributed memory system shown in Fig. 4, first node is for general
First mirror image data is stored into the distributed memory system, it is also necessary to execute step 205-207.
Wherein, in step 205, first node sends first mirror image data to warehouse server.In turn, in step
In 206, warehouse server can call warehouse as shown in Figure 4 to drive 04, which is saved to distributed storage
In system.Same optional, warehouse server can also send second to first node and upload success message, to inform first segment
Point first mirror image data uploads successfully.
In addition, the step 101-107 and step 201-207, the embodiment of the present invention in based on the above embodiment are also mentioned
For another mirror image delet method, this method is used to delete the mirror image in the warehouse server in docker system.
Illustratively, as shown in figure 9, (first mirror image includes the first mirror to delete the first mirror image in warehouse server
As data and the first metadata) for, the embodiment of the present invention provides a kind of mirror image delet method, comprising:
301, warehouse server receives the removal request that first node is sent, which, which is used to indicate, deletes warehouse clothes
The first mirror image being engaged in device.
302, warehouse server deletes the first metadata in warehouse server according to the removal request.
303, warehouse server judges whether the reference counter of the first mirror image is 0.
If 304, the reference counter of the first mirror image is 0, warehouse server deletion is stored in distributed memory system
The first mirror image data.
In step 301, the service in first node (first node is any node in above-mentioned docker system)
End 12, such as HTTP server can send removal request to warehouse server, which, which is used to indicate, deletes warehouse clothes
The first mirror image being engaged in device, for example, carrying the mark of the first metadata of the first mirror image in the removal request.
In turn, in step 302, warehouse server is according to the removal request received, first deletion warehouse server
The first interior metadata.
Further, since the mirror image data of mirror image is uniformly stored in distributed storage in each node in docker system
In system, even, the mirror image data in warehouse server also be can store in the distributed memory system, therefore, warehouse clothes
Device be engaged in when deleting the first mirror image data of the first mirror image, it is thus necessary to determine that after first mirror image is not used in each node, ability
Delete first mirror image data.
Certainly, since above-mentioned removal request is that first node is sent to warehouse server, warehouse server can write from memory
Recognize first node and first mirror image is not used, and then need to only determine that first mirror is not used in other nodes in addition to first node
As.
Illustratively, as previously mentioned, in docker system, it can be the preset reference counter of each mirror image, be used to
The access times of the mirror image are recorded, therefore, in step 303, warehouse server can be by judging the reference meter of the first mirror image
Whether number devices are 0, to determine whether first mirror image is used by one or more nodes.
Specifically, warehouse server can drive 04 to search first in distributed memory system with the warehouse in calling figure 4
The reference counter of mirror image illustrates do not have node in current docker system if the reference counter of the first mirror image is 0
Using first mirror image, at this point, in step 304, warehouse server deletes the first mirror being stored in distributed memory system
As data, to complete the delete operation of first mirror image.
So far, a kind of mirror image management method provided in an embodiment of the present invention specifically may include method for down loading, the mirror image of mirror image
Method for uploading and mirror image delet method, wherein, can be to mirror image in docker system provided in an embodiment of the present invention
Mirror image data and metadata are managed respectively, and the mirror image data of mirror image is uniformly stored in distributed memory system in each node
In, that is to say, that each node can share the mirror image data in distributed memory system, so, not only can be to avoid existing
Have in technology the problem of saving identical mirror image in great deal of nodes and lead to the waste of memory space, also make each node and
Warehouse server only needs to manage the metadata of mirror image in itself, so as to reduce the number of transmissions of mirror image data, reduces
The expense of docker system.
Figure 10 is a kind of structural schematic diagram of docker system interior nodes 02 provided in an embodiment of the present invention, and the present invention is implemented
The node 02 that example provides can be used for implementing the method that above-mentioned Fig. 2-various embodiments of the present invention shown in Fig. 9 are realized, for the ease of
Illustrate, only parts related to embodiments of the present invention are shown, and it is disclosed by specific technical details, referring to figure 2. shown in-Fig. 9
Various embodiments of the present invention.
Specifically, as shown in Figure 10, node 02 includes: mirror image generation unit 21, transmission unit 22 and storage unit 23.
Wherein, mirror image generation unit 21, for generating the first mirror image, first mirror image includes the first metadata and first
Mirror image data；
Transmission unit 22 carries described first in the upload request for sending upload request to warehouse server
Metadata, so that the warehouse server saves first metadata；
Storage unit 23, for saving first mirror image data into the distributed memory system.
Further, the transmission unit 22 is also used to send first mirror image data to the warehouse server, with
So that the warehouse server saves first mirror image data into the distributed memory system.
Further, as shown in figure 11, the node 02 further includes acquiring unit 24 and searching unit 25, wherein
The transmission unit 22 is also used to send downloading request to the warehouse server, carries in the downloading request
There is the mark of the first metadata；
The acquiring unit 24, for receiving the corresponding with the mark of first metadata of the warehouse server transmission
The first metadata；And if first mirror image data is stored in the distributed memory system, from the distribution
First mirror image data is obtained in storage system；
The searching unit 25, for searching in the distributed memory system and whether depositing according to first metadata
Contain the first mirror image data corresponding with first metadata.
Further, the acquiring unit 24, if being also used to not stored in the distributed memory system have described first
Mirror image data then obtains first mirror image data from the warehouse server.
Further, the preservation node 23, the sheet being also used in first meta-data preservation to the first node
Ground warehouse.
Further, as shown in figure 12, the node 02 further include:
Counting unit 26, for the reference counter execution plus 1 operation preset for the first mirror image.
Further, as shown in figure 13, the node 02 further includes deleting unit 27, wherein
The deletion unit 27, for deleting first metadata in the first node；
The counting unit 26 is also used to subtract 1 to the reference counter execution operation.
Wherein, above-mentioned image generation unit 21, transmission unit 22, storage unit 23, transmission unit 24, acquiring unit 24,
The concrete function of searching unit 25 and counting unit 26 can mirror image management module 100 as shown in Figure 6 with it is shown in fig. 6
Client 11, server 12, driving 13, Container Management library 14 and container 15 interact realization.
In addition, Figure 14 is the structural representation of warehouse server 01 in a kind of docker system provided in an embodiment of the present invention
Figure, warehouse server 01 provided in an embodiment of the present invention can be used for implementing above-mentioned Fig. 2-various embodiments of the present invention shown in Fig. 9
The method of realization, for ease of description, only parts related to embodiments of the present invention are shown, and particular technique details does not disclose
, referring to figure 2.-various embodiments of the present invention shown in Fig. 9.
Specifically, as shown in figure 14, warehouse server 01 includes: receiving unit 31 and deletion unit 32.
Wherein, receiving unit 31, for receiving the removal request of first node transmission, the removal request, which is used to indicate, is deleted
Except the first mirror image in the warehouse server, first mirror image includes the first metadata and the first mirror image data, and described the
One node is any of N number of node in the docker system；
Unit 32 is deleted, for deleting first metadata in the warehouse server according to the removal request；
If first mirror image is not used in each node, described first be stored in the distributed memory system is deleted
Mirror image data.
Further, as shown in figure 15, the warehouse server 01 further includes judging unit 33, wherein the judgement is single
Member 33, for judging whether the reference counter of first mirror image is 0；The deletion unit 32, if being specifically used for described the
The reference counter of one mirror image is 0, then deletes first mirror image data being stored in the distributed memory system.
As shown in figure 16, warehouse server 01 shown in node 02 or Figure 14-15 shown in Figure 10-13 can be to scheme
The mode of computer equipment (or system) in 16 is realized.
Figure 16 show computer equipment schematic diagram provided in an embodiment of the present invention.Computer equipment 100 includes at least one
A processor 41, communication bus 42, memory 43 and at least one communication interface 44.
For example, the concrete function of above-mentioned receiving unit 31, deletion unit 32 and judging unit 33 can be taken by warehouse
The computer instruction in the calling memory 43 of processor 41 being engaged in device 01 is realized.
Specifically, processor 41 can be a general central processor (CPU), microprocessor, the integrated electricity of specific application
Road (application-specific integrated circuit, ASIC), or it is one or more for controlling present invention side
The integrated circuit that case program executes.
Communication bus 42 may include an access, and information is transmitted between said modules.The communication interface 44, use is any
The device of transceiver one kind is used for and other equipment or communication, such as Ethernet, wireless access network (RAN), wireless office
Domain net (Wireless Local Area Networks, WLAN) etc..
Memory 43 can be read-only memory (read-only memory, ROM) or can store static information and instruction
Other kinds of static storage device, random access memory (random access memory, RAM) or letter can be stored
The other kinds of dynamic memory of breath and instruction, is also possible to Electrically Erasable Programmable Read-Only Memory (Electrically
Erasable Programmable Read-Only Memory, EEPROM), CD-ROM (Compact Disc Read-
Only Memory, CD-ROM) or other optical disc storages, optical disc storage (including compression optical disc, laser disc, optical disc, digital universal
Optical disc, Blu-ray Disc etc.), magnetic disk storage medium or other magnetic storage apparatus or can be used in carrying or store to have referring to
Enable or data structure form desired program code and can by any other medium of computer access, but not limited to this.
Memory 43, which can be, to be individually present, and is connected by bus with processor.Memory 43 can also be integrated in one with processor
It rises.
Wherein, the memory 43 be used for store execution the present invention program application code, and by processor 41
Control executes.The processor 41 is for executing the application code stored in the memory 43.
In the concrete realization, as one embodiment, processor 41 may include one or more CPU, such as in Figure 16
CPU0 and CPU1.
In the concrete realization, as one embodiment, computer equipment 100 may include multiple processors, such as Figure 16
In processor 41 and processor 48.Each of these processors can be monokaryon (single-CPU) processor,
It is also possible to multicore (multi-CPU) processor.Here processor can refer to one or more equipment, circuit, and/or
For handling the processing core of data (such as computer program instructions).
In the concrete realization, as one embodiment, computer equipment 100 can also include that output equipment 45 and input are set
Standby 46.Output equipment 45 and processor 41 communicate, and can show information in many ways.For example, output equipment 45 can be
Liquid crystal display (liquid crystal display, LCD), Light-Emitting Diode (light emitting diode, LED) are aobvious
Show equipment, cathode-ray tube (cathode ray tube, CRT) shows equipment or projector (projector) etc..Input is set
It is communicated for 46 and processor 41, the input of user can be received in many ways.For example, input equipment 46 can be mouse, key
Disk, touch panel device or sensing equipment etc..
Above-mentioned computer equipment 100 can be a general purpose computing device either dedicated computing machine equipment.
In the concrete realization, computer equipment 100 can be desktop computer, portable computer, network server, palm PC
(Personal Digital Assistant, PDA), cell phone, tablet computer, wireless terminal device, communication equipment, insertion
Formula equipment or the equipment for having similar structures in Figure 16.The unlimited type for determining computer equipment 100 of the embodiment of the present invention.
It should be noted that in present example, any one functional node in above-mentioned docker system, for example,
Warehouse server 01, node 02 may be realized by an entity device, it is also possible to it is realized jointly by multiple entity devices,
Each functional node in docker system may be realized by different entity devices respectively, it is also possible to all be set by the same entity
It is standby to realize.It is understood that any one functional node in docker system all may be that one in entity device patrols
Collect functional module, it is also possible to a logic functional block being made of multiple entity devices.
So far, a kind of mirror image managing device provided in an embodiment of the present invention, specifically includes warehouse server and each node,
Wherein, in docker system provided in an embodiment of the present invention, can mirror image data to mirror image and metadata be managed respectively,
The mirror image data of mirror image is uniformly stored in distributed memory system in each node, that is to say, that each node can be shared point
Mirror image data in cloth storage system so not only can be identical to avoid saving in great deal of nodes in the prior art
Mirror image and the problem of lead to the waste of memory space, also make each node and warehouse server only need to manage itself scope
The metadata of picture reduces the expense of docker system so as to reduce the number of transmissions of mirror image data.
It is apparent to those skilled in the art that for convenience and simplicity of description, only with above-mentioned each function
The division progress of module can according to need and for example, in practical application by above-mentioned function distribution by different function moulds
Block is completed, i.e., the internal structure of device is divided into different functional modules, to complete all or part of function described above
Energy.The specific work process of the system, apparatus, and unit of foregoing description, can be with reference to corresponding in preceding method embodiment
Journey, details are not described herein.
In several embodiments provided herein, it should be understood that disclosed system, device and method can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the module or
The division of unit, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units
Or component can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, institute
Display or the mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, device or unit
Indirect coupling or communication connection can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
It is each that equipment (can be personal computer, server or the network equipment etc.) or processor (processor) execute the present invention
The all or part of the steps of embodiment the method.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory
(ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic or disk
Etc. the various media that can store program code.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, protection scope of the present invention should be based on the protection scope of the described claims.
Claims (19)
1. a kind of mirror image management method, which is characterized in that the method is applied to containment system, and the containment system includes warehouse
Server and the N number of node being connected with the warehouse server, each node include memory module, and N is more than or equal to 2
Integer；The described method includes:
First node sends downloading request to the warehouse server, and first yuan of the first mirror image is carried in the downloading request
The mark of data, the first node are any of described N number of node；
The first node receives the warehouse server and sends first metadata；
The first node obtains first mirror image according to first metadata from the memory module of N number of node
First mirror image data.
2. method according to claim 1, which is characterized in that the first node is according to first metadata from institute
It states before obtaining first mirror image data in the memory module of N number of node, the method also includes:
The first node is determined and is stored in the memory module of N number of node and described according to first metadata
Corresponding first mirror image data of one metadata.
3. any method in -2 according to claim 1, which is characterized in that receive the warehouse in the first node
After the first metadata that server is sent, the method also includes:
The first node is by the ware-house here in first meta-data preservation to the first node.
4. method according to claim 3, which is characterized in that the first node by first meta-data preservation extremely
After ware-house here in the first node, the method also includes:
The first node is executed to the reference counter of first mirror image plus 1 operation.
5. method according to any one of claims 1-4, which is characterized in that the method also includes:
The first node generates the second mirror image, and second mirror image includes the second metadata and the second mirror image data；
The first node sends upload request to the warehouse server, carries the binary number in the upload request
According to so that the warehouse server saves first metadata；
The first node saves second mirror image data to the memory module of N number of node.
6. method according to claim 5, which is characterized in that the first node saves second mirror image data
To the memory module of N number of node, comprising:
The first node sends second mirror image data to the warehouse server, so that the warehouse server is by institute
The second mirror image data is stated to save to the memory module of N number of node.
7. according to the method described in claim 6, it is characterized in that, the method also includes:
The first node determines the third mirror image deleted in the first node, and the third mirror image includes third metadata；
The first node deletes the third metadata in the first node；
The first node is to the reference counter of first mirror image execution operation that subtracts 1.
8. a kind of mirror image management method, which is characterized in that the method is applied to containment system, and the containment system includes warehouse
Server and the N number of node being connected with the warehouse server, each node include memory module, and N is more than or equal to 2
Integer；The described method includes:
The warehouse server receives the removal request that first node is sent, and the removal request, which is used to indicate, deletes the warehouse
The first mirror image in server, first mirror image include the first metadata and the first mirror image data, and the first node is institute
State any of N number of node in containment system；
The warehouse server deletes first metadata in the warehouse server according to the removal request；
If first mirror image is not used in N number of node, the warehouse server deletion is stored in N number of node
First mirror image data in memory module.
9. according to the method described in claim 8, it is characterized in that, if first mirror image is not used in N number of node,
The warehouse server deletes the first mirror image data being stored in the memory module of N number of node, comprising:
The warehouse server judges whether the reference counter of first mirror image is 0；
If the reference counter of first mirror image is 0, the warehouse server deletes the storage for being stored in N number of node
First mirror image data in module.
10. a kind of containment system, which is characterized in that the system comprises: warehouse server and with the warehouse server phase
N number of node even, each node include memory module, and N is the integer more than or equal to 2；Wherein,
First node is used for: Xiang Suoshu warehouse server sends downloading request, carries the first mirror image in the downloading request
The mark of first metadata, the first node are any of described N number of node；It receives the warehouse server and sends institute
State the first metadata；The of first mirror image is obtained from the memory module of N number of node according to first metadata
One mirror image data；
The warehouse server, is used for: receiving the downloading request that the first node is sent；According in downloading request
The first metadata identifier lookup described in the first metadata；And first metadata is sent to the first node.
11. system according to claim 10, which is characterized in that according to first metadata from N number of node
Memory module in obtain first mirror image data before, the first node is also used to:
According to first metadata, determine be stored in the memory module of N number of node it is corresponding with first metadata
The first mirror image data.
12. any system in 0-12 according to claim 1, which is characterized in that receiving the warehouse server hair
After the first metadata sent, the first node is also used to:
By the ware-house here in first meta-data preservation to the first node.
13. wanting system described in 12 according to right, which is characterized in that by first meta-data preservation to the first node
After interior ware-house here, the first node is also used to:
The reference counter of first mirror image is executed plus 1 operates.
14. any system in 0-13 according to claim 1, which is characterized in that
The first node is also used to: generating the second mirror image, second mirror image includes the second metadata and the second mirror image data；
Second mirror image data is saved to the memory module of N number of node；Upload request, institute are sent to the warehouse server
It states and carries second metadata in upload request；
The warehouse server, is used for: receiving the upload request that the first node is sent；And save the upload request
In second metadata.
15. system according to claim 14, which is characterized in that the warehouse server includes warehouse driving；Wherein,
Second mirror image data is saved to the memory module of N number of node, is specifically included: the first node is used for:
Second mirror image data is sent to the warehouse server；
The warehouse server, is also used to: receiving second mirror image data that the first node is sent；And call the storehouse
Library driving saves second mirror image data to the memory module of N number of node.
16. system according to claim 15, which is characterized in that
The first node, is also used to: determining the third mirror image deleted in the first node, the third mirror image includes third
Metadata；Delete the third metadata in the first node；Delete first metadata in the first node；
To the reference counter of first mirror image execution operation that subtracts 1.
17. system described in any one of 0-16 according to claim 1, which is characterized in that
The first node is also used to send removal request to the warehouse server, and the removal request is used to indicate deletion
The third mirror image in the warehouse server；
The warehouse server, is also used to: receiving the removal request that the first node is sent；It is asked according to the deletion
It asks, deletes the third metadata in the warehouse server；If the third mirror image is not used in N number of node,
Delete the mirror image data for the third mirror image being stored in the memory module of N number of node.
18. a kind of node, which is characterized in that the node is suitable for containment system, and the containment system includes warehouse server
With N number of node；The node includes interface & processor, and the interface is used for and other nodes in N number of node
And the warehouse server communication, the processor require method described in 1-7 for perform claim, N is more than or equal to 2
Integer.
19. a kind of warehouse server, which is characterized in that the warehouse server belongs to containment system, and the containment system also wraps
N number of node is included, the warehouse server includes interface & processor, and the interface with N number of node for communicating, the place
It manages device and requires method described in 8-9 for perform claim, N is for the integer more than or equal to 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910145316.2A CN109976667B (en) | 2016-01-26 | 2016-01-26 | Mirror image management method, device and system |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910145316.2A CN109976667B (en) | 2016-01-26 | 2016-01-26 | Mirror image management method, device and system |
| CN201610051837.8A CN105740048B (en) | 2016-01-26 | 2016-01-26 | A kind of mirror image management method, apparatus and system |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610051837.8A Division CN105740048B (en) | 2016-01-26 | 2016-01-26 | A kind of mirror image management method, apparatus and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109976667A true CN109976667A (en) | 2019-07-05 |
| CN109976667B CN109976667B (en) | 2021-09-14 |
Family
ID=56247362
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610051837.8A Active CN105740048B (en) | 2016-01-26 | 2016-01-26 | A kind of mirror image management method, apparatus and system |
| CN201910145316.2A Active CN109976667B (en) | 2016-01-26 | 2016-01-26 | Mirror image management method, device and system |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610051837.8A Active CN105740048B (en) | 2016-01-26 | 2016-01-26 | A kind of mirror image management method, apparatus and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN105740048B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110347535A (en) * | 2019-07-16 | 2019-10-18 | 北京谷数科技有限公司 | A method of promoting dual controller write performance |
| CN111221623A (en) * | 2019-12-26 | 2020-06-02 | 中科曙光国际信息产业有限公司 | Mirror image library management system |
Families Citing this family (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107797806A (en) * | 2016-08-29 | 2018-03-13 | 北京雪球信息科技有限公司 | A kind of dispositions method of program |
| CN106506587B (en) * | 2016-09-23 | 2021-08-06 | 中国人民解放军国防科学技术大学 | Docker mirror image downloading method based on distributed storage |
| CN106708599B (en) * | 2016-12-01 | 2020-08-25 | 携程旅游信息技术（上海）有限公司 | Virtual machine mirror image issuing system and method |
| CN108205481B (en) * | 2016-12-19 | 2022-03-08 | 华为技术有限公司 | Application container instance creation method and device |
| CN106997392B (en) * | 2017-04-10 | 2020-03-03 | 北京搜狐新媒体信息技术有限公司 | Processing method and device and reading method and device of Docker mirror image cache information |
| CN107105054B (en) * | 2017-05-17 | 2019-12-24 | 苏州浪潮智能科技有限公司 | Mirror image garbage cleaning system and method for docker mirror image warehouse |
| CN107515776B (en) * | 2017-07-18 | 2021-04-09 | 深信服科技股份有限公司 | Method for upgrading service continuously, node to be upgraded and readable storage medium |
| CN107450961B (en) * | 2017-09-22 | 2020-10-16 | 济南浚达信息技术有限公司 | Distributed deep learning system based on Docker container and construction method and working method thereof |
| CN109669808A (en) * | 2017-10-17 | 2019-04-23 | 中移(苏州)软件技术有限公司 | A kind of method and apparatus of system backup |
| CN109962843B (en) * | 2017-12-22 | 2021-04-27 | 上海威派格智慧水务股份有限公司 | Method for realizing edge information processing of gateway of Internet of things |
| CN108156246B (en) * | 2017-12-27 | 2020-12-04 | 北京微网通联股份有限公司 | File distribution method, system, server and client |
| CN110099076A (en) * | 2018-01-29 | 2019-08-06 | 中兴通讯股份有限公司 | A kind of method and its system that mirror image pulls |
| CN110198330B (en) * | 2018-03-27 | 2022-03-22 | 腾讯科技（深圳）有限公司 | Mirror image downloading method and device |
| US11132293B2 (en) | 2018-04-10 | 2021-09-28 | Red Hat Israel, Ltd. | Intelligent garbage collector for containers |
| CN108647248B (en) * | 2018-04-16 | 2021-03-09 | 新华三技术有限公司成都分公司 | WORM state monitoring transfer method and device |
| CN110908671A (en) * | 2018-09-18 | 2020-03-24 | 北京京东尚科信息技术有限公司 | Method and device for constructing docker mirror image and computer readable storage medium |
| CN109936571B (en) * | 2019-02-22 | 2020-05-29 | 全球能源互联网研究院有限公司 | Mass data sharing method, open sharing platform and electronic equipment |
| CN109951551B (en) * | 2019-03-21 | 2021-03-19 | 联想(北京)有限公司 | Container mirror image management system and method |
| CN110119251B (en) * | 2019-05-14 | 2023-01-03 | 北京杰思安全科技有限公司 | Buffer control method and system based on disk large flow |
| CN110704162B (en) * | 2019-09-27 | 2022-09-20 | 北京百度网讯科技有限公司 | Method, device and equipment for sharing container mirror image by physical machine and storage medium |
| CN113872999A (en) * | 2020-06-30 | 2021-12-31 | 中兴通讯股份有限公司 | Mirror image distribution method, electronic device and storage medium |
| CN112748879B (en) * | 2020-12-30 | 2023-03-10 | 中科曙光国际信息产业有限公司 | Data acquisition method, system, device, computer equipment and storage medium |
| CN117667298A (en) * | 2022-08-31 | 2024-03-08 | 成都华为技术有限公司 | Method and device for starting container, computing node and shared storage equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100036879A1 (en) * | 2008-05-08 | 2010-02-11 | Thomas Friese | Method, computer-readable medium, and system for storing, allocating and retrieving medical image data in a distributed computerized system of a clinical facility. |
| CN103761275A (en) * | 2014-01-09 | 2014-04-30 | 浪潮电子信息产业股份有限公司 | Management method for metadata in distributed file system |
| CN104391802A (en) * | 2014-11-24 | 2015-03-04 | 浪潮电子信息产业股份有限公司 | Streamline pool metadata node refreshing consistency protecting method |
| CN104516967A (en) * | 2014-12-25 | 2015-04-15 | 国家电网公司 | Electric power system mass data management system and use method thereof |
| US20150149410A1 (en) * | 2013-11-27 | 2015-05-28 | Microsoft Corporation | Publishing to a content delivery system |
| WO2015127083A2 (en) * | 2014-02-21 | 2015-08-27 | Solidfire, Inc. | Data syncing in a distributed system |
| CN104885076A (en) * | 2012-05-10 | 2015-09-02 | 以太存储股份有限公司 | Systems and methods for distributed storage |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8495356B2 (en) * | 2010-12-31 | 2013-07-23 | International Business Machines Corporation | System for securing virtual machine disks on a remote shared storage subsystem |
| CN102932415B (en) * | 2012-09-26 | 2015-04-15 | 东软集团股份有限公司 | Method and device for storing mirror image document |
| CN103544045A (en) * | 2013-10-16 | 2014-01-29 | 南京大学镇江高新技术研究院 | HDFS-based virtual machine image storage system and construction method thereof |
| CN104731516B (en) * | 2013-12-18 | 2019-03-01 | 腾讯科技（深圳）有限公司 | A kind of method, apparatus and distributed memory system of accessing file |
| CN105045656B (en) * | 2015-06-30 | 2018-11-30 | 深圳清华大学研究院 | Big data storage and management method based on virtual container |
| CN104951308B (en) * | 2015-06-30 | 2019-05-21 | 北京奇虎科技有限公司 | The management optimization mode and device of Docker Registry |
| CN105187500A (en) * | 2015-08-07 | 2015-12-23 | 浪潮(北京)电子信息产业有限公司 | Container-based distributed storage system deployment method |
| CN105094924B (en) * | 2015-08-21 | 2018-06-26 | 华为技术有限公司 | The introduction method and device of image file |
-
2016
- 2016-01-26 CN CN201610051837.8A patent/CN105740048B/en active Active
- 2016-01-26 CN CN201910145316.2A patent/CN109976667B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100036879A1 (en) * | 2008-05-08 | 2010-02-11 | Thomas Friese | Method, computer-readable medium, and system for storing, allocating and retrieving medical image data in a distributed computerized system of a clinical facility. |
| CN104885076A (en) * | 2012-05-10 | 2015-09-02 | 以太存储股份有限公司 | Systems and methods for distributed storage |
| US20150149410A1 (en) * | 2013-11-27 | 2015-05-28 | Microsoft Corporation | Publishing to a content delivery system |
| CN103761275A (en) * | 2014-01-09 | 2014-04-30 | 浪潮电子信息产业股份有限公司 | Management method for metadata in distributed file system |
| WO2015127083A2 (en) * | 2014-02-21 | 2015-08-27 | Solidfire, Inc. | Data syncing in a distributed system |
| CN104391802A (en) * | 2014-11-24 | 2015-03-04 | 浪潮电子信息产业股份有限公司 | Streamline pool metadata node refreshing consistency protecting method |
| CN104516967A (en) * | 2014-12-25 | 2015-04-15 | 国家电网公司 | Electric power system mass data management system and use method thereof |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110347535A (en) * | 2019-07-16 | 2019-10-18 | 北京谷数科技有限公司 | A method of promoting dual controller write performance |
| CN110347535B (en) * | 2019-07-16 | 2020-05-26 | 北京谷数科技股份有限公司 | Method for improving writing performance of double controllers |
| CN111221623A (en) * | 2019-12-26 | 2020-06-02 | 中科曙光国际信息产业有限公司 | Mirror image library management system |
| CN111221623B (en) * | 2019-12-26 | 2024-01-19 | 中科曙光国际信息产业有限公司 | Mirror image library management system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105740048A (en) | 2016-07-06 |
| CN105740048B (en) | 2019-03-08 |
| CN109976667B (en) | 2021-09-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105740048B (en) | A kind of mirror image management method, apparatus and system | |
| US10291704B2 (en) | Networked solutions integration using a cloud business object broker | |
| CN109582441A (en) | For providing system, the method and apparatus of container service | |
| Teng et al. | A medical image archive solution in the cloud | |
| CN103620549B (en) | The storage medium stored for uniform data is abstract | |
| US8788760B2 (en) | Adaptive caching of data | |
| CN110447017A (en) | Rule-based modification in Data-Storage Equipment monitor | |
| CN108885554A (en) | Specific dispersion computer system | |
| CN111901294A (en) | Method for constructing online machine learning project and machine learning system | |
| JP2021515294A (en) | Transaction processing in a multi-master distributed data management system | |
| US20190005067A1 (en) | Multi-tenant data service in distributed file systems for big data analysis | |
| CN105515872B (en) | The update method of configuration information, apparatus and system | |
| US11630695B1 (en) | Dynamic reassignment in a search and indexing system | |
| CN104160381A (en) | Managing tenant-specific data sets in a multi-tenant environment | |
| CN108027828A (en) | It is synchronous with the trust instrument of stateless synchronization node | |
| WO2020215752A1 (en) | Graph computing method and device | |
| CN101388891A (en) | Apparatus and method for managing presence information of a presentity | |
| CN108196787A (en) | The quota management method and cluster storage system of cluster storage system | |
| CN109992956A (en) | The processing method and relevant apparatus of the security strategy of container | |
| JP6760904B2 (en) | Information processing equipment, information processing methods, and information processing programs | |
| US11693710B1 (en) | Workload pool hierarchy for a search and indexing system | |
| CN107408239A (en) | Carry out the framework of Mass Data Management in communication applications by multiple mailboxes | |
| CN110457257A (en) | Archives knowledge managing device and mobile terminal for mobile terminal | |
| CN109445966A (en) | Event-handling method, device, medium and calculating equipment | |
| US10423785B2 (en) | Data scanning within distributed computing components |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220218 Address after: 550025 Huawei cloud data center, jiaoxinggong Road, Qianzhong Avenue, Gui'an New District, Guiyang City, Guizhou Province Patentee after: Huawei Cloud Computing Technology Co.,Ltd. Address before: 518129 Bantian HUAWEI headquarters office building, Longgang District, Guangdong, Shenzhen Patentee before: HUAWEI TECHNOLOGIES Co.,Ltd. |
|
| TR01 | Transfer of patent right |