CN105607954A - Stateful container online migration method and apparatus - Google Patents
Stateful container online migration method and apparatus Download PDFInfo
- Publication number
- CN105607954A CN105607954A CN201510973448.6A CN201510973448A CN105607954A CN 105607954 A CN105607954 A CN 105607954A CN 201510973448 A CN201510973448 A CN 201510973448A CN 105607954 A CN105607954 A CN 105607954A
- Authority
- CN
- China
- Prior art keywords
- node
- physical host
- container
- host node
- status
- 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
- 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
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
- G06F9/5088—Techniques for rebalancing the load in a distributed system involving task migration
-
- 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
- G06F3/0646—Horizontal data movement in storage systems, i.e. moving data in between storage devices or systems
- G06F3/0647—Migration mechanisms
-
- 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]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention provides a stateful container online migration method and apparatus. The method comprises the steps of creating a snapshot of a data volume of a to-be-migrated stateful container; through virtual routers arranged in a first physical host node and a second physical host node, establishing a cross-physical host container network and transmitting the snapshot to the second physical host node, wherein the first physical host node is a host node where the to-be-migrated stateful container is, and the second physical host node is a target host of the to-be-migrated stateful container; determining newly added data after the creation of the snapshot is started; transmitting the newly added data to the second physical host node through the container network; and writing the snapshot and the newly added data into the data volume of the newly created container in the second physical host node to reconstruct the stateful container. With the adoption of the technical scheme provided by the invention, the cross-physical host online migration of the stateful container can be realized.
Description
Technical field
The present invention relates to computer cloud computing technology field, particularly a kind of have status container to move onlineMethod and apparatus.
Background technology
Along with Linux container technique is ripe gradually, especially yielding unusually brilliant results of Container Management instrument Docker,OpenShift, CloudFoundry and Baidu's cloud application engine etc. start to adopt Docker container as safetyCarry out the sandbox of cloud application. Also phase of AmazonAWS, GoogleAppEngine and MicrosoftAzureContainers for pushing out service continues.
For the convenient resilient expansion that realizes application, the application being deployed in cloud computing application engine is all conventionallyStateless, i.e. the internal memory of these application and file system only play a part data cached, really realize dataWhat persist is rear end stores service, comprises database server, key-value stores service and message systemDeng. Therefore these rear end stores service are called as the cloud service that has state.
Having status container is the container that has data storage function, can save data. And stateless container can notSave data. Book container migration is referred to a container transfers to another one host from a host systemSystem. The migration of stateless container service is fairly simple, even can realize online migration. But, existingCloud computing application engine is not supported the migration of status container service.
Summary of the invention
The embodiment of the present invention has proposed a kind of method and apparatus that has the online migration of status container, has in order to realizeThe online migration of status container service.
The embodiment of the present invention provides a kind of method that has the online migration of status container, comprises the steps:
Create the snapshot that has status container book to be migrated;
By be arranged on the first physical host node and the second physical host node on virtual router, buildThe vertical container network across physical host, transfers to described the second physical host node by described snapshot; DescribedOne physical host node is the host node that has status container place to be migrated; Described the second physical host nodeFor the described destination host that has status container to be migrated;
Determine that described snapshot creation starts newly-increased data afterwards;
By described container network by described newly-increased transfer of data to described the second physical host node;
Described snapshot and described newly-increased data are write to the new container that creates on described the second physical host nodeBook in, rebuild and have a status container.
The embodiment of the present invention provides a kind of device that has the online migration of status container, comprising:
Snapshot creation unit, for creating the snapshot that has status container book to be migrated;
Transmission unit, for by be arranged on the first physical host node and the second physical host node onVirtual router, sets up across physical host container network, and described snapshot is transferred to described the second physical hostNode; Described the first physical host node is the host node that has status container place to be migrated; Described secondPhysical host node is the described destination host that has status container to be migrated;
Data determining unit, for determining that described snapshot creation starts newly-increased data afterwards;
Described transmission unit is further used for by described container network described newly-increased transfer of data to instituteState the second physical host node;
Data write unit, for writing described the second physical host by described snapshot and described newly-increased dataOn node, in the new book that creates container, rebuild and have status container.
Beneficial effect of the present invention is as follows:
A kind of method and apparatus that has the online migration of status container that the embodiment of the present invention provides, by being arranged onThe virtual router having on the host node of status container place and on destination host node to be migrated, sets up containerNetwork, transmitting to be migratedly has the snapshot of status container book to destination host, after determining that snapshot creation startsAfter newly-increased data, transmit newly-increased data to destination host, snapshot and newly-increased data are write to destination hostIn the book of the upper new container creating, rebuild and have status container, can realize status container across physicsMain frame moves online.
Brief description of the drawings
Specific embodiments of the invention are described below with reference to accompanying drawings,
Fig. 1 is the method flow schematic diagram that has the online migration of status container in the embodiment of the present invention;
Fig. 2 is the schematic flow sheet that has the online migration of status container book in the embodiment of the present invention;
Fig. 3 is cluster management schematic diagram in the embodiment of the present invention;
Fig. 4 is cluster period distribution schematic diagram in the embodiment of the present invention;
Fig. 5 is host node state conversion schematic diagram in the embodiment of the present invention;
Fig. 6 is daily record reproduction schematic diagram in the embodiment of the present invention;
Fig. 7 is OpenContrail system framework figure in the embodiment of the present invention;
Fig. 8 is cluster network management schematic diagram in the embodiment of the present invention;
Fig. 9 is scheduling of resource schematic diagram in the embodiment of the present invention;
Figure 10 is the apparatus structure schematic diagram that has the online migration of status container in the embodiment of the present invention.
Detailed description of the invention
In order to make technical scheme of the present invention and advantage clearer, below in conjunction with accompanying drawing to of the present inventionExemplary embodiment is described in more detail, and obviously, described embodiment is only of the present invention onePart embodiment, instead of all embodiment's is exhaustive. And in the situation that not conflicting, in this descriptionEmbodiment and embodiment in feature can interosculate.
Fig. 1 is the method flow schematic diagram that has the online migration of status container in the embodiment of the present invention, as Fig. 1 instituteShow have the method for the online migration of status container to comprise the steps:
Step 101: create the snapshot that has status container book to be migrated;
Step 102: by be arranged on the first physical host node and the second physical host node on virtualRouter, sets up the container network across physical host, and described snapshot is transferred to described the second physical host jointPoint; Described the first physical host node is the host node that has status container place to be migrated; Described the second thingReason host node is the described destination host that has status container to be migrated;
Step 103: determine that described snapshot creation starts newly-increased data afterwards;
Step 104: by described container network by described newly-increased transfer of data to described the second physical hostNode;
Step 105: described snapshot and described newly-increased data are write on described the second physical host node newCreate in the book of container, rebuild and have status container.
Snapshot is that this copy comprises that corresponding data is at certain to specific data set complete usable copyThe reflection of time point (time point that copy starts). , snapshot can be of its represented dataCopy can be also a duplicate of data. The effect of snapshot is mainly to carry out online data backupWith recovery. In the time of memory device generation application and trouble or file corruption, can carry out data fast and recover,Data are recovered to the state of certain available time point. Another effect of snapshot is to provide for storage userAnother one data access passage, in the time that former data are carried out online application processing, user can access snapshot numberAccording to, can also utilize snapshot the work such as to test.
Determining that described snapshot creation starts newly-increased data afterwards, in concrete enforcement, can be definite described fastAccording to transmitting once completing the described status container that has to be migrated out of service, and determine from described snapshot creationAfterwards to described to be migrated have status container out of service before newly-increased data in during this period of time; Also canAfter definite described snapshot creation starts, make described to be migrated have status container continue operation a period of time afterOut of service, determine from described snapshot creation starts after extremely described to be migrated have status container out of service itBefore newly-increased data in during this period of time. Wherein, described to be migratedly have status container determining described snapshot woundBuilding the time of the follow-up reforwarding row of beginning can be determined according to actual conditions by those skilled in the art, does not do hereConcrete restriction.
In concrete enforcement, described snapshot and described newly-increased data are write on described the second physical host nodeIn the new book that creates container, can be that the snapshot being first transmitted is first write to described the second physical hostOn node, newly create in the book of container, more newly-increased data are write on described the second physical host nodeIn the new book that creates container; Also can be write together with described snapshot is merged with newly-increased data described inOn the second physical host node, newly create in the book of container; Also can be first newly-increased data to be write to instituteState in the book of new establishment container on the second physical host node, more described snapshot is write to described the second thingOn reason host node, newly create in the book of container. The data writing of concrete described snapshot and newly-increased dataAnd be to write in the book of new establishment container on described the second physical host node after merging, do not do tool hereBody limits.
A kind of method that has the online migration of status container that the embodiment of the present invention provides, to be migrated by being arranged onThere is the virtual router on the host node of status container place and on destination host node, set up across physical hostContainer network, transmitting to be migratedly has the snapshot of status container book to destination host, determines snapshot creationStart, after rear newly-increased data, to transmit newly-increased data to destination host, snapshot and newly-increased data are write to orderOn mark main frame, in the book of the new container creating, rebuild and have status container, can realize status containerMove online across main frame.
In enforcement, the snapshot that creates the book that has status container to be migrated specifically can comprise:
Based on the snapshot of the Btrfs system creation book that has status container to be migrated, the file of described containerSystem is Btrfs system.
In concrete enforcement, due to the good autgmentability of Btrfs system, B tree and dynamically inode (informationNode, information node/index node) characteristic such as establishment ensured that Btrfs system still has Zhuo on large-scale machinePerformance more, its overall performance and can not reducing along with the increase of power system capacity. And when systems face can notExpect hardware fault time, Btrfs system adopt COW (CopyOnWrite, copy-on-write) technology canTo ensure the uniformity of file system. Btrfs system is also supported check sum code, has avoided without alarm faultAppearance. In addition, the support of Btrfs system creates snapshot (snapshot) and clone (clone), can be easilyManage multiple physical equipments. Btrfs system can also significantly improve time/space performance of file system,Comprise and postpone to distribute, the storage optimization of small documents, directory index etc.
So, in the embodiment of the present invention, can select Btrfs system as there being status container place physical hostThe file system of storage data. Fig. 2 has the flow process of status container book migration to show in the embodiment of the present inventionIntention, as shown in Figure 2, is starting a PostgresSQL database service that has state from node 1Before device container, first with the storage pool "/data " being connected from node 1 create " pg " catalogue,For the data of stored data base server. Then with a PostgresSQL of the mirror image postgres of official operationContainer A 1, wherein operational factor " v " is mapped as "/the data/pg " of main frame the book of container A 1"/var/lib/postgresql/data ". If there being status container A1 to move to from node 2, need to carry outThe following step: (1) is creating snapshot "/data/pg.snapshot " the transmission of "/data/pg " from node 1At the snapshot "/data/pg.snapshot " of "/data/pg " that create from node 1 to the storage from node 2Pond. According to varying in size of the type of database, file, this step may need to continue for some time;(2), after snapshot is sent, out of service have a status container A1, finds out snapshot from node 1The difference of "/data/pg.snapshot " and "/data/pg ", i.e. newly-increased data after snapshot creation, sendTo from node 2, merge into new "/data/pg " with "/the data/pg.snapshot " of node 2. BySmaller in the data volume increasing progressively, this step can complete faster.
In enforcement, by be arranged on the first physical host node and the second physical host node on virtual roadBy device, set up across physical host container network, by described snapshot transfer to described the second physical host node itBefore, can also comprise:
On described the first physical host node, the first virtual router is set;
On described the second physical host node, the second virtual router is set; Described first and second virtual routeDevice connects, and institute's virtual router possesses kernel state forwarding capability and/or user's state control function.
In concrete enforcement, have a status container management tool due to existing, such as Docker etc. do not support acrossPhysical host have a networking between status container. Can application software define grid skill in the embodiment of the present inventionArt arranges a virtual router, the virtual route of setting on each has the main frame at status container placeDevice can be for supporting MPLS (MultiprotocolLabelSwitching, multiprotocol label switching) overGRE(GenericRoutingEncapsulation, generic route encapsulation)/UDP (UserDatagramProtocol,UDP) or VXLAN (VirtualeXtensibleLocalAreaNetwork, virtualCan expand LAN) etc. the gateway router of agreement, have the physics master at status container place for connecting differenceMachine.
Virtual router can be a kernel module, creates required virtual network and corresponding void by itIntend routing table. If virtual machine or there is status container to be connected on the virtual router of the machine, firstCreate a pair of virtual ethernet network interface, one with virtual machine/have status container to be connected, another one and virtualThe virtual routing table of network is connected. Between the virtual router of different physical host nodes, set up MPLSoverGRE/UDP or VxLAN tunnel. Like this, be positioned at different physical host nodes but belong to same virtualThe different virtual machine of network or have status container just can set up network service.
Wherein, the virtual router of setting can comprise kernel state forwarding capability and/or user's state control function,Kernel state and user's state have memory-mapped separately, i.e. address space. The application program of user's space, logicalCross system call, enter kernel spacing. Represent that by kernel this process runs on kernel spacing, user's space andKernel spacing has different address mappings, general or special purpose register group, and the process of user's space is wantedTransmit a lot of variablees, parameter to kernel spacing, kernel spacing also to preserve consumer process some registers,Variablees etc., continue to carry out so that user's space is got back in system call after finishing. Operate in the program under user's stateDirectly accessing operation system kernel data structure and program. Carry out a program in system time, large portionDivide the time to operate under user's state, it does not have power and energy to need operating system to help through some at itWhen work that power completes, will be switched to kernel state.
In concrete enforcement, the first virtual router is set on described the first physical host node and described theThe step that the second virtual router is set on two physical host nodes there is no absolute sequential relationship, as long as ensureStep 102 by be arranged on the first physical host node and the second physical host node on virtual routeDevice, sets up across physical host container network, before described snapshot is transferred to described the second physical host nodeComplete, can ensure that described the first physical host node and described the second physical host node can communicate by letterCan, this easily knows for those skilled in the art, repeats no more.
In enforcement, by be arranged on the first physical host node and the second physical host node on virtual roadBy device, described snapshot being transferred to the second physical host node specifically can comprise:
By be arranged on the first physical host node and the second physical host node on virtual router willDescribed snapshot tunnel encapsulation is sent to described the second physical host node after becoming packet.
In concrete enforcement, due to tunneling technique be a kind of infrastructure by using internet network itBetween transmit the mode of data. The data (or load) that use tunnel to transmit can be the Frames of different agreementOr bag. Then tunnel protocol sends the Frame of other agreement or bag Reseal by tunnel. New frameHead provides routing iinformation, to transmit packed load data by internet. Tunneling technique can makeThe Network of taking pride in multiple information sources is transmitted by different tunnels in same infrastructure. TunnelUtilization point to point protocol has replaced exchange to connect, and comes connection data address by route network. TunnelRoad technology allows user's access enterprise networks at any time and any place of authorizing mobile subscriber or having authorizedNetwork. So, in the embodiment of the present invention, save with the second physical host by being arranged on the first physical host nodeVirtual router on point is sent to described the second physical host after described snapshot tunnel encapsulation is become to packetNode.
In enforcement, determine being transmitted of described snapshot after newly-increased data specifically can comprise:
Determine from described snapshot creation and start to described to be transferred newly-increased during having status container out of serviceData.
In concrete enforcement, newly-increased data can be in being transmitted of described snapshot, start to described in treatTransfer have status container that time out of service during this period of time in, described to be migrated new in having status containerThe data that increase. It is described that to be transferred to have status container out of service be the time can select transferring of described snapshotAny a certain moment after becoming. In order to shorten the described time that has status container migration to be transferred, Ke YiThe transmission of described snapshot, once completing the described status container that has to be transferred out of service, reduces the number increasing newly as far as possibleAccording to data volume. Conventionally, out of service describedly to be transferredly there is state to hold in the transmission of described snapshot once completingThe processing mode of device, newly-increased data volume only has hundreds of KB, and the off-line time that has status container to serve can be veryShort.
In enforcement, described snapshot and described newly-increased data are write on described the second physical host node to newly woundBuild in the book of container, reconstruction can also comprise after having status container:
The described network configuration information that has status container to be migrated is sent to described the second physical host node,Described network configuration information comprises: affiliated virtual network information, IP (InternetProtocol, internetwork-ingAgreement) address and/or the Hostname that is positioned at;
Upgrade the route plan of the router on described the second physical host node according to described network configuration informationSlightly.
In concrete enforcement, described snapshot and described newly-increased data are write on described the second physical host nodeIn the new book that creates container, reconstruction can also check the described state that has to be migrated after having status containerThe network configuration information of container, comprise described to be migrated have virtual network, IP address that status container connects andThe information such as host name, are also sent to the described destination host that has status container to be migrated. (3) move described waitingMove on the destination host of status container, virtual router is added into the described status container place that has to be migratedVirtual network, as shown in Figure 2 with a PostgreSQL container of the mirror database postgre of official operationB2, wherein operational factor " v " is mapped as "/the data/pg " of main frame the book of container B 2"/var/lib/postgresql/data ", according to the network of the configuration setting B2 of A1. Through above-mentioned steps, fromThe status container that has on node 1 has just successfully moved to from node 2.
In enforcement, described snapshot and described newly-increased data are write on described the second physical host node to newly woundBuild in the book of container, reconstruction can also comprise after having status container:
Obtain the configuration information of described the second physical host node by proxy module, described proxy module settingOn described the second physical host node; Described configuration information comprises resource information, resource occupation status informationAnd/or the information of container that creates or move on this node, described first and second physical host node is from node;
Described configuration information is sent to corresponding with described the second physical host node by described proxy moduleHost node;
Described host node is according to described configuration information update and report global configuration information to the neck corresponding with itLead node;
By described leader node according to the resource status of described global configuration information updating container cluster.
In concrete enforcement, what the embodiment of the present invention provided has the method for status container migration can be applied in clusterIn. Fig. 3 is cluster management schematic diagram in the embodiment of the present invention, and as shown in Figure 3, cluster management parts are thingReason mainframe cluster is mapped as a resource pool. After user's submit job task, cluster management parts basisScheduling strategy is selected suitable physical host to create and is moved the status container that has of executing the task. According to role'sDifference, the physical host of composition cluster is divided into host node and from the large class of node two. Host node centralized stores is allThe configuration information of node, scheduling is from the node task that fulfils assignment. Receive the dispatch command of host node from node,Establishment and operation have status container; The configuration information of simultaneously reporting this node and having a status container to host node withAnd there is status container to take the situation of resource. Cause cluster unavailable in order to prevent host node to lose efficacy, generalCan in cluster, dispose multiple host nodes, conventionally can dispose 3,5 or 7 host nodes, comprise 1Individual leader (leader) and multiple follower (follower). From node meeting and only can with a main jointPoint is set up session (TCP connection). In the time sending request from node, the host node being attached thereto is first requestBe transmitted to leader's node. Leader's node adds the order comprising in request in daily record to, then copiesTo follower's node. The present invention adopts Raft agreement to ensure the uniformity of the store status of each host node.
Fig. 4 is cluster period distribution schematic diagram in the embodiment of the present invention, as shown in Figure 4, and in cluster, timeBetween be divided into period of random length. The sequence number of period is an integer since 1 continuous growth. EachWhen period starts, to complete leader's election, if successfully elect leader's node, be coordinated by itThe uniformity of host node, completes the function such as configuration management and scheduling of resource; If elected unsuccessfully, finish to work asThe front period, restart a new period.
Host node in cluster only can be in one of following three kinds of states: follower, candidate or leader,Between three, have specific switch condition, Fig. 5 is host node state conversion schematic diagram in the embodiment of the present invention,As described in Figure 5, in the time that cluster starts, all host nodes are all in " follower " state. Any oneHost node, as long as it receives the far call request from a leader (or candidate) node,It will keep " follower " state always. And leader's node can regularly send heartbeat message to allFollower's node. If follower's node is not received any information in heart beat cycle, it can be thoughtLeader's node lost efficacy, so open new round leader election.
First follower's node increases progressively the sequence number of current period, and the state of oneself is converted to from " follower "" candidate ". Then it can be voted to oneself, and please to all other parallel transmissions of host node in clusterAsk the information of ballot. This node can be always in " candidate " state, until one of following three conditions becomeVertical: it has won election (1), becomes leader's node; (2) there is other node to win election, become neckLead people's node; (3) the election time expires, but does not have node to be elected as leader.
The condition that candidate's node wins election is that most of host node is all thrown ballot paper to it. Fixing at onePeriod, each host node according to first first service principle, ballot to candidate's node. Once waitThe node of choosing wins election, and it just becomes leader's node, sends heartbeat message at once to other host node,Thereby prevent new election.
In election process, there are two kinds of need special circumstances to be processed. The first situation is candidate's node AIn the time waiting for and collect the ballot paper of other host node, likely can receive the number that another one Node B sendsAccording to, and B node claims it oneself is leader's node. Now, A node can compare the period of two nodesThe size of sequence number. If the period sequence number of B is greater than the period sequence number of A, A admits that B is one and closes soThe leader of method, and the state of oneself is revised as to " follower " from " candidate "; Otherwise, A nodeCan refuse the request of B node, continue to keep candidate's state.
The second special circumstances are in this period, do not have node to win election. For example,, if there are a lot of jointsPoint almost transfers candidate to simultaneously, may cause can obtaining most choosings without any candidate's nodeTicket. Now, all candidate's nodes can finish the current period after election is overtime, open a new choosingLift the period. In order to prevent from repeatedly occurring electing failed situation, for each candidate's node is set different choosingsLift overtime, for example, in a regular time interval (, 150-300 millisecond), choose at random one superDuration. This method can reduce effectively because ballot paper disperses the election frequency of failure causing, and ensures fastSelect leader's node of cluster.
Described snapshot and described newly-increased data are write to the new container that creates on described the second physical host nodeBook in, rebuild have status container after, complete the migration of status container, operate in the second physicsOn host node new create have status container after, actuator the occupation condition of container report to fromThe proxy module of node 1, by being reported to leader's node 2 from the proxy module of node 1, amendment clusterAvailable resources state. The new running status that has status container creating of actuator monitoring, once this has state to holdDevice operation makes mistakes, and actuator is reported error message to scheduler, and can also process and set according to mistake in advance,Autoboot container.
In enforcement, described host node reports extremely leader corresponding to described host node of global configuration information after renewalBefore node, can also comprise:
Described host node adopts internal memory key-Value Data storehouse to preserve described configuration information to described cluster global configurationIn information.
In concrete enforcement, because key assignments stored data base is a kind of non-relational database, its data are with keyThe form that value is right is organized, index and storage, is applicable to not relating to the business of too much data relationship business relationsData can effectively reduce the number of times of read-write disk simultaneously, than SQL (StructuredQueryLanguage,SQL) database storage has better readwrite performance, so, report at described host nodeGlobal configuration information after renewal is to before leader node corresponding to described host node, can adopt internal memory key-Described configuration information is preserved to described cluster global configuration information, to ensure the described cluster overall situation in Value Data storehouseConfiguration information has good readwrite performance.
In theory, as long as preserving configuration information to its in described cluster global configuration informationIts technology also can be implemented this programme, and internal memory key-Value Data storehouse system is only for instructing this area skillHow art personnel specifically implement the present invention, but do not mean and only can use system one side, internal memory key-Value Data storehouseFormula, can need to carry out to determine corresponding mode in conjunction with practice in implementation process.
Send request to the host node being attached thereto from node, each request comprises a needs host node and holdsThe order of row. All requests are forwarded to leader's node there. First leader's node adds one newlyDaily record, the sequence number of the period that comprise order, receives orders and the call number of daily record. Journal entry nowIn submit state not, also there is no fill order. Next, leader's node copies to institute journal entrySome follower's nodes. If most follower's nodes have successfully write the journal entry receiving,Leader's node can be submitted this journal entry to, and the order comprising in execution journal item, then notifies all followingAlso submit this journal entry to person's node. Like this, in cluster, the state of most host nodes is consistent.
Fig. 6 is daily record reproduction schematic diagram in the embodiment of the present invention, as shown in Figure 6, will write one from node 1Item configuration " X=5 ". What be attached thereto is host node 1, and it is transmitted to leader's node this write request2. Host node 2 generates a new journal entry " (4,2, X=5) ", and wherein " 4 " are the call numbers of journal entry," 2 " are period sequence numbers now, and " X=5 " is the order that will carry out. Next, leader's node 2Journal entry " (4,2, X=5) " is sent to all follower's nodes, i.e. host node 1 and 3. If node1 and the daily record of node 3 oneself in there is not the daily record conflicting with this journal entry, they can return successfully and writeThe message of journal entry. Leader's node 2 is received after message, submits this journal entry to, i.e. getting the X of storageValue is revised as " 5 ", finally notifies follower's node also to submit this daily record to. Node 1 and node 3 also divideThe other value the X of storage is revised as " 5 ". Now, in cluster, the state of 3 host nodes is consistent,Leader's node 1 is notified from node 1 data and is successfully write. In said process, if do not have manyNumber follower node is confirmed to write journal entry, and leader's node 1 will can not be submitted journal entry to, and notice is from jointPoint 1 write request failure.
Above-mentioned leader's election and daily record reproduction process can ensure the configuration information of submitting to from node to writeIn host node cluster, and effectively solve host node Problem of Failure.
In enforcement, join according to the overall situation of described global configuration information updating container cluster by described leader nodePut information.
Described sets of containers is mined massively with OpenContrail system made across physical host container network, described OpenNode in Contrail system comprises analysis node, controls node and configuration node; Described control node rootAccording to the analysis result of described analysis node, upgrade the global configuration information of described container network; Described configuration sectionPoint is preserved the global configuration information of described container network.
Fig. 7 is OpenContrail system framework figure in the embodiment of the present invention, as shown in Figure 7, and first at collectionIn group, dispose an OpenContail system, it is made up of a series of node. Each node is onePlatform physics or virtual server, bear different roles. Configuration node, analysis node and control node are altogetherComposition OpenContrail controller together, have status container or the virtual machine managed on multiple nodes by it are distinguishedThe virtual network of composition. Wherein, configuration node has been preserved the configuration information of all nodes and network; Analyze jointThe information of some collection, storage and analyzing virtual network; Control node and realize the control plane function of network, andThe data retransmission aspect function of network, is realized by the virtual router of installing on each computing node.
Fig. 8 is cluster network management schematic diagram in the embodiment of the present invention, as shown in Figure 8, from node 1 and fromNode 2 all belongs to the node in OpenContrail system, therefore, on these two nodes, install and joinPut virtual router. If moved A1, A2 and tri-of A3 have status container from node 1, and from jointPoint has moved B1 on 2 and two of B2 have status container. On each virtual router, create 2 voidIntend network net1 and net2, each virtual router has virtual routing table separately. From node 1,Having status container A1 and A2 to be assigned to virtual network net1, there being status container A3 to be assigned to virtual netNetwork net2; From node 2, there being status container B1 to be assigned to virtual network net1, there being state to holdDevice B2 is assigned to network net2. For having status container, each creates a pair of virtual ethernet network interface, oneThe individual status container that is connected with, another one connects virtual routing forwarding corresponding to virtual network. Virtual routeThe built-in DHCP of device (DynamicHostConfigurationProtocol, DHCP) serviceDevice, the status container that has being connected with each virtual network can Dynamic Acquisition to network address. ConfigurationNode is collected the virtual network configuration of all virtual routers, and notifies virtual router switching and routing list item.Therefore, from the routing table of the virtual network net1 of node 1, finally also can include status container B1 roadBy, and also can include status container A1 and A2 from the routing table of the virtual network net1 of node 2Route. When having status container A1 to there being status container B1 to send when packet, first virtual router existsOn the basis of raw data packets, add one deck tunnel encapsulation, then send to from node 2, the latter is according to tunnelThe information of encapsulation, then initial data has been transmitted to status container B1.
Under implied terms, the node in different virtual network cannot be communicated by letter. For example, A1 and A3 are with regard to nothingMethod is set up network and is connected. If the communication between support virtual network must manually be added toward virtual routerAdd corresponding routing policy.
In theory, be also to implement this as long as realizing other technology of network virtualizationScheme, OpenContrail system is only for instructing those skilled in the art specifically how to implement the present invention,But do not mean and only can use a kind of mode of OpenContrail system, in implementation process, can need in conjunction with practiceDetermine corresponding mode.
In enforcement, on the first physical host node and on the second physical host node, arrange virtual router itBefore, can also comprise:
Determine described the second physical host node.
In concrete enforcement, on the first physical host node and on the second physical host node, virtual route is setBefore device, can pre-determine the resources such as certain CPU, internal memory and be applicable to the described status container migration that has to be migratedPhysical host as described the second physical host node.
In enforcement, determine that described the second physical host node specifically can comprise:
Send and have status container migration request to the main joint corresponding with it by described the first physical host nodePoint;
Described in forwarding based on described host node, there is status container migration request to the leader node corresponding with it;
By described leader node according to described in have described in having the resource status of status container cluster to determine to meetThe resource requirement of status container migration request be described the second physical host node from node.
In concrete enforcement, physical host cluster adopts two-level scheduler mechanism, and Fig. 9 provides in the embodiment of the present inventionSource scheduling schematic diagram, as shown in Figure 9, all configuration letters that reports this node from node to leader's nodeBreath, the resource distribution module on leader's node provides the resource operation API of standard. Developer calls moneySource distribution module, writes different operation frameworks. It is (for example, long that each framework is applicable to operation particular typeStay service, batch processing or stream processing operation) application program, mainly formed by scheduler and actuator.Scheduler is arranged on host node conventionally, and it turns according to wheel or has scheduling strategies such as priority, operationTask is assigned to available from node; Actuator is arranged on from node, receives and execution scheduler transmissionTask, the simultaneously implementation status of monitor task. Actuator calls Docker establishment and operation is executed the taskContainer, realize isolation and the resource limitation of different task.
4 core CPU and 4GB internal memory from the configuration of node 1. Leader's node 2 is received from node 1After configuration, can notify the scheduler of each framework. The scheduler of framework 1 is checked the operation that user submits to, sends outExisting operation 1 comprises two tasks, needs respectively 2 core CPU/2GB internal memories and monokaryon CPU/1GB internal memory.Satisfy the demands from the configuration of node 1. So the scheduler of framework 1 distributes to these two tasks from node1. The mission bit stream sending comprises framework title (being framework 1) here, (is from joint here from node IDPoint 1), resource limits (CPU, internal memory etc.) and the Docker mirror image of executing the task required or for structureBuild the Dockerfile of mirror image. Actuator and the Docker of framework 1 have been installed from node 1. If hadNecessity, first actuator builds the Docker mirror image of executing the task required. Actuator taking Docker mirror image asTemplate, taking the resource that limits as start-up parameter, the container that operation is executed the task.
In the embodiment of the present invention, in the time determining described the second physical host node, can pass through described the first physicsMain frame sends and has status container migration request to the host node corresponding with it from node, then is turned by described host nodeDescribed in sending out, have status container migration request to the leader node corresponding with it, by described leader node according to instituteState the resource status of status container cluster and determine described in can meeting have status container migration request resource needWhat ask is described the second physical host node from node, determines that resource distribution has state described in being more than or equal toThe resource requirement of container migration request from node as the described target physical main frame that has status container to be migratedNode.
Based on same inventive concept, in the embodiment of the present invention, also providing a kind of has status container to move onlineDevice, the principle of dealing with problems due to this device has the method for the online migration of status container similar to a kind of, because ofThe enforcement of this this device can be referring to the enforcement of method, repeats part and repeat no more.
Figure 10 is the apparatus structure schematic diagram that has status container migration in the embodiment of the present invention, for as Figure 10 instituteShow, the device that this has status container migration, can comprise:
Snapshot creation unit, for creating the snapshot that has status container book to be migrated;
Transmission unit, for by be arranged on the first physical host node and the second physical host node onVirtual router, sets up across physical host container network, and described snapshot is transferred to described the second physical hostNode; Described the first physical host node is the host node that has status container place to be migrated; Described secondPhysical host node is the described destination host that has status container to be migrated;
Data determining unit, for determining that described snapshot creation starts newly-increased data afterwards;
Described transmission unit is further used for by described container network described newly-increased transfer of data to instituteState the second physical host node;
Data write unit, for writing described the second physical host by described snapshot and described newly-increased dataOn node, in the new book that creates container, rebuild and have status container.
In enforcement, described snapshot creation unit is specifically for treating based on B tree file system Btrfs system creationMigration has the snapshot of the book of status container, and the file system of described container is Btrfs system.
In enforcement, this has the device of status container migration to comprise:
Setting unit, for by be arranged on the first physical host node and the second physical host node onVirtual router, sets up across physical host container network, and described snapshot is transferred to described the second physical hostBefore node, on described the first physical host node, the first virtual router is set; At described the second physicsThe second virtual router is set on host node; Described first and second virtual router connects, and institute is virtualRouter possesses kernel state forwarding capability and/or user's state control function.
In enforcement, described transmission unit specifically can be for by being arranged on the first physical host node and theVirtual router on two physical host nodes carries out described snapshot to be sent to described second after tunnel encapsulationPhysical host node
In enforcement, described data determining unit specifically can start to described from described snapshot creation for determiningTo be transferred have status container out of service during newly-increased data.
In enforcement, described transmission unit can be further used for described snapshot and described newly-increased data to writeOn described the second physical host node, newly create in the book of container, after reconstruction has status container, by instituteState the network configuration information that has status container to be migrated and be sent to described the second physical host node, described networkConfiguration information comprises: affiliated virtual network information, IP address and/or the Hostname being positioned at;
The described device that has status container migration, can also comprise:
Routing policy updating block, for upgrading described the second physical host joint according to described network configuration informationThe routing policy of the router on point.
In enforcement, this has the device of status container migration to comprise:
Acquiring unit, for writing described the second physical host node by described snapshot and described newly-increased dataIn the upper new book that creates container, reconstruction obtains described second by proxy module after having status containerThe configuration information of physical host node, described proxy module is arranged on described the second physical host node; InstituteState configuration information and comprise resource information, resource occupation status information and/or the appearance that creates or move on this nodeDevice information, described first and second physical host node is from node;
Described transmission unit is further used for by described proxy module, described configuration information being sent to and instituteState host node corresponding to the second physical host node;
Updating block, for by described host node according to described configuration information update and report global configuration letterBreath is to the leader node corresponding with it; By described leader node according to described global configuration information updating containerThe resource status of cluster.
In enforcement, this has the device of status container migration to comprise:
Storage unit, reports the extremely described host node correspondence of global configuration information after renewal for described host nodeLeader node before, described host node adopt internal memory key-Value Data storehouse preserve described configuration information to described inIn cluster global configuration information.
In enforcement, described updating block is specifically for being mined massively and used OpenContrail system by described sets of containersSet up across physical host container network, the node in described OpenContrail system comprises analysis node, controlNode processed and configuration node; Described control node, according to the analysis result of described analysis node, upgrades described appearanceThe global configuration information of device network; Described configuration node is preserved the global configuration information of described container network.
In enforcement, this has the device of status container migration to comprise:
Destination host determining unit, for establishing on the first physical host node and on the second physical host nodeBefore putting virtual router, determine described the second physical host node.
In enforcement, described destination host determining unit is specifically for sending by described the first physical host nodeThere is status container migration request to the host node corresponding with it; Described in forwarding based on described host node, there is state to holdDevice migration request is to the leader node corresponding with it; Described in described leader node basis, there is status container collectionGroup's resource status determines that what have the resource requirement of status container migration request described in meeting is described the from nodeTwo physical host nodes.
For convenience of description, the each several part of the above device is divided into various modules or unit difference with functionDescribe. Certainly, can be the function of each module or unit at same or multiple software in the time that enforcement is of the present inventionOr realize in hardware.
Those skilled in the art should understand, embodiments of the invention can be provided as method, system or meterCalculation machine program product. Therefore, the present invention can adopt complete hardware implementation example, completely implement software example or knotClose the form of the embodiment of software and hardware aspect. And the present invention can adopt at one or more wherein bagsThe computer-usable storage medium that contains computer usable program code (include but not limited to magnetic disc store,CD-ROM, optical memory etc.) form of the upper computer program of implementing.
The present invention is that reference is according to the method for the embodiment of the present invention, equipment (system) and computer program productThe flow chart of product and/or block diagram are described. Should understand can be by computer program instructions realization flow figure and/ or block diagram in each flow process and/or flow process in square frame and flow chart and/or block diagram and/Or the combination of square frame. Can provide these computer program instructions to all-purpose computer, special-purpose computer, embeddingThe processor of formula processor or other programmable data processing device, to produce a machine, makes by calculatingThe instruction that the processor of machine or other programmable data processing device is carried out produces for realizing at flow chart oneThe device of the function of specifying in square frame of individual flow process or multiple flow process and/or block diagram or multiple square frame.
These computer program instructions also can be stored in energy vectoring computer or other programmable data processing are establishedIn the standby computer-readable memory with ad hoc fashion work, make to be stored in this computer-readable memoryInstruction produce and comprise the manufacture of command device, this command device is realized in flow process or multiple of flow chartThe function of specifying in square frame of flow process and/or block diagram or multiple square frame.
These computer program instructions also can be loaded in computer or other programmable data processing device, makeMust on computer or other programmable devices, carry out sequence of operations step to produce computer implemented placeReason, thus the instruction of carrying out on computer or other programmable devices is provided for realizing one of flow chartThe step of the function of specifying in square frame of flow process or multiple flow process and/or block diagram or multiple square frame.
Although described the preferred embodiments of the present invention, once those skilled in the art obtain cicada baseThis creative concept, can make other change and amendment to these embodiment. So appended right is wantedAsk and be intended to be interpreted as comprising preferred embodiment and fall into all changes and the amendment of the scope of the invention.
Obviously, those skilled in the art can carry out various changes and modification and not depart from this present inventionBright spirit and scope. Like this, if of the present invention these amendment and modification belong to the claims in the present invention andWithin the scope of its equivalent technologies, the present invention be also intended to comprise these change and modification interior.
Claims (22)
1. a method that has the online migration of status container, is characterized in that, comprises the steps:
Create the snapshot that has status container book to be migrated;
By be arranged on the first physical host node and the second physical host node on virtual router, buildThe vertical container network across physical host, transfers to described the second physical host node by described snapshot; DescribedOne physical host node is the host node that has status container place to be migrated; Described the second physical host nodeFor the described destination host that has status container to be migrated;
Determine that described snapshot creation starts newly-increased data afterwards;
By described container network by described newly-increased transfer of data to described the second physical host node;
Described snapshot and described newly-increased data are write to the new container that creates on described the second physical host nodeBook in, rebuild and have a status container.
2. the method for claim 1, is characterized in that, creates the data that have status container to be migratedThe snapshot of volume specifically comprises:
Create the snapshot of the book that has status container to be migrated based on B tree file system Btrfs system, described inThe file system of container is Btrfs system.
3. the method for claim 1, is characterized in that, by being arranged on the first physical host nodeVirtual router above and on the second physical host node, sets up across physical host container network, by described fastBefore transferring to described the second physical host node, also comprise:
On described the first physical host node, the first virtual router is set;
On described the second physical host node, the second virtual router is set; Described first and second virtual routeDevice connects, and institute's virtual router possesses kernel state forwarding capability and/or user's state control function.
4. the method for claim 1, is characterized in that, by being arranged on the first physical host nodeVirtual router above and on the second physical host node, sets up across physical host container network, by described fastSpecifically comprise according to transferring to described the second physical host node:
By be arranged on the first physical host node and the second physical host node on virtual router willDescribed snapshot carries out being sent to described the second physical host node after tunnel encapsulation.
5. the method for claim 1, is characterized in that, after determining being transmitted of described snapshotNewly-increased data specifically comprise:
Determine from described snapshot creation and start to described to be transferred newly-increased during having status container out of serviceData.
6. the method for claim 1, is characterized in that, by described snapshot and described newly-increased dataWrite in the book of new establishment container on described the second physical host node, after reconstruction has status container,Also comprise:
The described network configuration information that has status container to be migrated is sent to described the second physical host node,Described network configuration information comprises: affiliated virtual network information, IP address and/or the Hostname being positioned at;
Upgrade the route plan of the router on described the second physical host node according to described network configuration informationSlightly.
7. the method for claim 1, is characterized in that, by described snapshot and described newly-increased dataWrite in the book of new establishment container on described the second physical host node, after reconstruction has status container,Also comprise:
Obtain the configuration information of described the second physical host node by proxy module, described proxy module settingOn described the second physical host node; Described configuration information comprises resource information, resource occupation status informationAnd/or the information of container that creates or move on this node, described first and second physical host node is from node:
Described configuration information is sent to corresponding with described the second physical host node by described proxy moduleHost node;
Described host node is according to described configuration information update and report global configuration information to the neck corresponding with itLead node;
By described leader node according to the resource status of described global configuration information updating container cluster.
8. method as claimed in claim 7, is characterized in that, described host node reports the overall situation after renewalBefore the configuration information leader node that extremely described host node is corresponding, also comprise:
Described host node adopts internal memory key-Value Data storehouse to preserve described configuration information to described cluster global configurationIn information.
9. method as claimed in claim 7, is characterized in that, by described leader node according to described completeThe global configuration information of office's configuration information update container cluster specifically comprises:
Described sets of containers is mined massively with OpenContrail system made across physical host container network, described OpenNode in Contrail system comprises analysis node, controls node and configuration node; Described control node rootAccording to the analysis result of described analysis node, upgrade the global configuration information of container network; Described configuration node is protectedDeposit the global configuration information of described container network.
10. method as claimed in claim 7, is characterized in that, on the first physical host node andBefore virtual router being set on two physical host nodes, also comprise:
Determine described the second physical host node.
11. methods as claimed in claim 10, is characterized in that, determine described the second physical host nodeSpecifically comprise:
Send and have status container migration request to the main joint corresponding with it by described the first physical host nodePoint;
Described in forwarding based on described host node, there is status container migration request to the leader node corresponding with it;
By described leader node according to described in have described in having the resource status of status container cluster to determine to meetThe resource requirement of status container migration request be described the second physical host node from node.
12. 1 kinds have the device of the online migration of status container, it is characterized in that, comprising:
Snapshot creation unit, for creating the snapshot that has status container book to be migrated;
Transmission unit, for by be arranged on the first physical host node and the second physical host node onVirtual router, sets up across physical host container network, and described snapshot is transferred to described the second physical hostNode; Described the first physical host node is the host node that has status container place to be migrated; Described secondPhysical host node is the described destination host that has status container to be migrated;
Data determining unit, for determining that described snapshot creation starts newly-increased data afterwards;
Described transmission unit is further used for by described container network described newly-increased transfer of data to instituteState the second physical host node;
Data write unit, for writing described the second physical host by described snapshot and described newly-increased dataOn node, in the new book that creates container, rebuild and have status container.
13. devices as claimed in claim 12, is characterized in that, described snapshot creation unit specifically forBased on the snapshot of the B tree file system Btrfs system creation book that has status container to be migrated, described appearanceThe file system of device is Btrfs system.
14. devices as claimed in claim 12, is characterized in that, also comprise:
Setting unit, for by be arranged on the first physical host node and the second physical host node onVirtual router, sets up across physical host container network, and described snapshot is transferred to described the second physical hostBefore node, on described the first physical host node, the first virtual router is set; At described the second physicsThe second virtual router is set on host node; Described first and second virtual router connects, and institute is virtualRouter possesses kernel state forwarding capability and/or user's state control function.
15. devices as claimed in claim 12, is characterized in that, described transmission unit is specifically for passing throughBe arranged on the first physical host node and the second physical host node on virtual router by described snapshotCarry out being sent to described the second physical host node after tunnel encapsulation.
16. devices as claimed in claim 12, is characterized in that, described data determining unit specifically forDetermine from described snapshot creation start to described to be transferred have status container out of service during newly-increased numberAccording to.
17. devices as claimed in claim 12, is characterized in that, described transmission unit be further used for byDescribed snapshot and described newly-increased data write the new data that create container on described the second physical host nodeIn volume, reconstruction after having status container is sent to the described network configuration information that has status container to be migratedDescribed the second physical host node, described network configuration information comprises: affiliated virtual network information, IP addressAnd/or the Hostname being positioned at;
The described device that has status container migration, also comprises:
Routing policy updating block, for upgrading described the second physical host joint according to described network configuration informationThe routing policy of the router on point.
18. devices as claimed in claim 12, is characterized in that, also comprise:
Acquiring unit, for writing described the second physical host node by described snapshot and described newly-increased dataIn the upper new book that creates container, reconstruction obtains described second by proxy module after having status containerThe configuration information of physical host node, described proxy module is arranged on described the second physical host node; InstituteState configuration information and comprise resource information, resource occupation status information and/or the appearance that creates or move on this nodeDevice information, described first and second physical host node is from node;
Described transmission unit is further used for by described proxy module, described configuration information being sent to and instituteState host node corresponding to the second physical host node;
Updating block, for by described host node according to described configuration information update and report global configuration letterBreath is to the leader node corresponding with it; By described leader node according to described global configuration information updating containerThe resource status of cluster.
19. devices as claimed in claim 18, is characterized in that, also comprise:
Storage unit, reports the extremely described host node correspondence of global configuration information after renewal for described host nodeLeader node before, described host node adopt internal memory key-Value Data storehouse preserve described configuration information to described inIn cluster global configuration information.
20. devices as claimed in claim 18, is characterized in that, described updating block is specifically for passing throughDescribed sets of containers is mined massively with OpenContrail system made across physical host container network, described OpenNode in Contrail system comprises analysis node, controls node and configuration node; Described control node rootAccording to the analysis result of described analysis node, upgrade the global configuration information of described container network; Described configuration sectionPoint is preserved the global configuration information of described container network.
21. devices as claimed in claim 18, is characterized in that, also comprise:
Destination host determining unit, for establishing on the first physical host node and on the second physical host nodeBefore putting virtual router, determine described the second physical host node.
22. devices as claimed in claim 21, is characterized in that, described destination host determining unit is concreteFor sending and have status container migration request to the main joint corresponding with it by described the first physical host nodePoint; Described in forwarding based on described host node, there is status container migration request to the leader node corresponding with it; LogicalCross described leader node according to described in have state to hold described in having the resource status of status container cluster to determine to meetThe resource requirement of device migration request be described the second physical host node from node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510973448.6A CN105607954B (en) | 2015-12-21 | 2015-12-21 | A kind of method and apparatus that stateful container migrates online |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510973448.6A CN105607954B (en) | 2015-12-21 | 2015-12-21 | A kind of method and apparatus that stateful container migrates online |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105607954A true CN105607954A (en) | 2016-05-25 |
| CN105607954B CN105607954B (en) | 2019-05-14 |
Family
ID=55987908
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510973448.6A Active CN105607954B (en) | 2015-12-21 | 2015-12-21 | A kind of method and apparatus that stateful container migrates online |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105607954B (en) |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106502830A (en) * | 2016-10-27 | 2017-03-15 | 铭软件股份有限公司 | A kind of method for restoring system backup based on Btrfs file system |
| CN106506255A (en) * | 2016-09-21 | 2017-03-15 | 微梦创科网络科技（中国）有限公司 | A kind of method of pressure test, apparatus and system |
| CN107070691A (en) * | 2017-01-12 | 2017-08-18 | 阿里巴巴集团控股有限公司 | Docker containers across host communication method and system |
| CN107193489A (en) * | 2017-05-15 | 2017-09-22 | 郑州云海信息技术有限公司 | A kind of storage level data migration method and device based on storage virtual gateway |
| CN107193499A (en) * | 2017-05-26 | 2017-09-22 | 郑州云海信息技术有限公司 | A kind of moving method and device of container data volume |
| CN107391318A (en) * | 2017-09-18 | 2017-11-24 | 郑州云海信息技术有限公司 | A kind of method, system and device of container data persistence |
| CN107515782A (en) * | 2017-07-26 | 2017-12-26 | 北京天云融创软件技术有限公司 | Implementation method of the container across host migration under a kind of Docker environment |
| CN107728947A (en) * | 2017-10-13 | 2018-02-23 | 郑州云海信息技术有限公司 | A kind of cold moving method of container and system |
| CN107729020A (en) * | 2017-10-11 | 2018-02-23 | 北京航空航天大学 | A kind of method for realizing extensive container rapid deployment |
| CN108170514A (en) * | 2017-12-19 | 2018-06-15 | 中国科学院声学研究所 | A kind of container operation method and device |
| CN108268304A (en) * | 2017-01-03 | 2018-07-10 | 中国科学院声学研究所 | A kind of Web app moving methods based on container |
| CN108279969A (en) * | 2018-02-26 | 2018-07-13 | 中科边缘智慧信息科技（苏州）有限公司 | Stateful service container thermomigration process based on memory compression transmission |
| CN108628874A (en) * | 2017-03-17 | 2018-10-09 | 北京京东尚科信息技术有限公司 | Method, apparatus, electronic equipment and the readable storage medium storing program for executing of migrating data |
| CN108897557A (en) * | 2018-06-20 | 2018-11-27 | 中国联合网络通信集团有限公司 | The update method and device of micro services framework |
| CN109150941A (en) * | 2017-06-27 | 2019-01-04 | 中国科学院信息工程研究所 | A kind of data center's physical resource flooding method |
| CN109246167A (en) * | 2017-07-11 | 2019-01-18 | 阿里巴巴集团控股有限公司 | A kind of container dispatching method and device |
| CN109857516A (en) * | 2018-12-29 | 2019-06-07 | 航天信息股份有限公司 | Cluster moving method and device based on container |
| CN111092770A (en) * | 2019-12-23 | 2020-05-01 | 联想(北京)有限公司 | Virtual network management method and electronic equipment |
| CN111190702A (en) * | 2020-01-14 | 2020-05-22 | 山东超越数控电子股份有限公司 | Cross-node hot migration method for Docker |
| US10691504B2 (en) | 2017-08-14 | 2020-06-23 | International Business Machines Corporation | Container based service management |
| CN111343219A (en) * | 2018-12-18 | 2020-06-26 | 同方威视技术股份有限公司 | Computing service cloud platform |
| CN111339061A (en) * | 2020-02-12 | 2020-06-26 | 杭州涂鸦信息技术有限公司 | Data migration method and system of distributed database |
| CN111352689A (en) * | 2018-12-21 | 2020-06-30 | 中国电信股份有限公司 | Method and device for realizing seamless migration of application containerized deployment |
| CN111684419A (en) * | 2018-02-01 | 2020-09-18 | 西门子股份公司 | Method and system for migrating containers in a container orchestration platform between computing nodes |
| CN111694789A (en) * | 2020-04-22 | 2020-09-22 | 西安电子科技大学 | Embedded reconfigurable heterogeneous determination method, system, storage medium and processor |
| CN116204305A (en) * | 2022-12-21 | 2023-06-02 | 山东未来网络研究院(紫金山实验室工业互联网创新应用基地) | Method for limiting number of dock container inodes |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060029054A1 (en) * | 2004-04-29 | 2006-02-09 | International Business Machines Corporation | System and method for modeling and dynamically deploying services into a distributed networking architecture |
| CN101169740A (en) * | 2007-08-07 | 2008-04-30 | 北京大学 | EJB component dynamic migration method |
| EP1636704A4 (en) * | 2003-06-09 | 2008-06-11 | Ind Defender Inc | Event monitoring and management |
| CN103024053A (en) * | 2012-12-18 | 2013-04-03 | 华为技术有限公司 | Cloud storage method, resource scheduling system and cloud storage node and system |
| WO2013112023A1 (en) * | 2012-01-26 | 2013-08-01 | Samsung Electronics Co., Ltd. | Processing state information |
-
2015
- 2015-12-21 CN CN201510973448.6A patent/CN105607954B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1636704A4 (en) * | 2003-06-09 | 2008-06-11 | Ind Defender Inc | Event monitoring and management |
| US20060029054A1 (en) * | 2004-04-29 | 2006-02-09 | International Business Machines Corporation | System and method for modeling and dynamically deploying services into a distributed networking architecture |
| CN101169740A (en) * | 2007-08-07 | 2008-04-30 | 北京大学 | EJB component dynamic migration method |
| WO2013112023A1 (en) * | 2012-01-26 | 2013-08-01 | Samsung Electronics Co., Ltd. | Processing state information |
| CN103024053A (en) * | 2012-12-18 | 2013-04-03 | 华为技术有限公司 | Cloud storage method, resource scheduling system and cloud storage node and system |
Non-Patent Citations (2)
| Title |
|---|
| 王晓鹏: "《一种面向构件化软件的在线演化方法》", 《计算机学报》 * |
| 范国闯: "《Web应用服务器关键技术研究》", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (42)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106506255A (en) * | 2016-09-21 | 2017-03-15 | 微梦创科网络科技（中国）有限公司 | A kind of method of pressure test, apparatus and system |
| CN106506255B (en) * | 2016-09-21 | 2019-11-05 | 微梦创科网络科技（中国）有限公司 | A kind of method, apparatus and system of pressure test |
| CN106502830A (en) * | 2016-10-27 | 2017-03-15 | 铭软件股份有限公司 | A kind of method for restoring system backup based on Btrfs file system |
| CN106502830B (en) * | 2016-10-27 | 2019-01-22 | 一铭软件股份有限公司 | A kind of method for restoring system backup based on Btrfs file system |
| CN108268304B (en) * | 2017-01-03 | 2021-06-08 | 中国科学院声学研究所 | Web app migration method based on container |
| CN108268304A (en) * | 2017-01-03 | 2018-07-10 | 中国科学院声学研究所 | A kind of Web app moving methods based on container |
| CN107070691A (en) * | 2017-01-12 | 2017-08-18 | 阿里巴巴集团控股有限公司 | Docker containers across host communication method and system |
| CN107070691B (en) * | 2017-01-12 | 2020-01-21 | 阿里巴巴集团控股有限公司 | Cross-host communication method and system of Docker container |
| CN108628874A (en) * | 2017-03-17 | 2018-10-09 | 北京京东尚科信息技术有限公司 | Method, apparatus, electronic equipment and the readable storage medium storing program for executing of migrating data |
| CN108628874B (en) * | 2017-03-17 | 2020-12-22 | 北京京东尚科信息技术有限公司 | Method and device for migrating data, electronic equipment and readable storage medium |
| CN107193489A (en) * | 2017-05-15 | 2017-09-22 | 郑州云海信息技术有限公司 | A kind of storage level data migration method and device based on storage virtual gateway |
| CN107193499A (en) * | 2017-05-26 | 2017-09-22 | 郑州云海信息技术有限公司 | A kind of moving method and device of container data volume |
| CN109150941B (en) * | 2017-06-27 | 2021-04-27 | 中国科学院信息工程研究所 | Data center physical resource floating method |
| CN109150941A (en) * | 2017-06-27 | 2019-01-04 | 中国科学院信息工程研究所 | A kind of data center's physical resource flooding method |
| CN109246167A (en) * | 2017-07-11 | 2019-01-18 | 阿里巴巴集团控股有限公司 | A kind of container dispatching method and device |
| CN109246167B (en) * | 2017-07-11 | 2022-04-01 | 阿里巴巴集团控股有限公司 | Container scheduling method and device |
| CN107515782A (en) * | 2017-07-26 | 2017-12-26 | 北京天云融创软件技术有限公司 | Implementation method of the container across host migration under a kind of Docker environment |
| US11023286B2 (en) | 2017-08-14 | 2021-06-01 | International Business Machines Corporation | Container based service management |
| US10691504B2 (en) | 2017-08-14 | 2020-06-23 | International Business Machines Corporation | Container based service management |
| CN107391318A (en) * | 2017-09-18 | 2017-11-24 | 郑州云海信息技术有限公司 | A kind of method, system and device of container data persistence |
| CN107729020B (en) * | 2017-10-11 | 2020-08-28 | 北京航空航天大学 | Method for realizing rapid deployment of large-scale container |
| CN107729020A (en) * | 2017-10-11 | 2018-02-23 | 北京航空航天大学 | A kind of method for realizing extensive container rapid deployment |
| CN107728947A (en) * | 2017-10-13 | 2018-02-23 | 郑州云海信息技术有限公司 | A kind of cold moving method of container and system |
| CN108170514A (en) * | 2017-12-19 | 2018-06-15 | 中国科学院声学研究所 | A kind of container operation method and device |
| CN111684419A (en) * | 2018-02-01 | 2020-09-18 | 西门子股份公司 | Method and system for migrating containers in a container orchestration platform between computing nodes |
| CN111684419B (en) * | 2018-02-01 | 2024-06-04 | 西门子股份公司 | Method and system for migrating containers in a container orchestration platform between computing nodes |
| US11809900B2 (en) | 2018-02-01 | 2023-11-07 | Siemens Aktiengesellschaft | Method and system for migration of containers in a container orchestration platform between compute nodes |
| CN108279969A (en) * | 2018-02-26 | 2018-07-13 | 中科边缘智慧信息科技（苏州）有限公司 | Stateful service container thermomigration process based on memory compression transmission |
| CN108897557A (en) * | 2018-06-20 | 2018-11-27 | 中国联合网络通信集团有限公司 | The update method and device of micro services framework |
| CN108897557B (en) * | 2018-06-20 | 2022-06-10 | 中国联合网络通信集团有限公司 | Updating method and device of microservice architecture |
| CN111343219A (en) * | 2018-12-18 | 2020-06-26 | 同方威视技术股份有限公司 | Computing service cloud platform |
| CN111352689A (en) * | 2018-12-21 | 2020-06-30 | 中国电信股份有限公司 | Method and device for realizing seamless migration of application containerized deployment |
| CN109857516B (en) * | 2018-12-29 | 2023-12-12 | 航天信息股份有限公司 | Cluster migration method and device based on container |
| CN109857516A (en) * | 2018-12-29 | 2019-06-07 | 航天信息股份有限公司 | Cluster moving method and device based on container |
| CN111092770B (en) * | 2019-12-23 | 2021-09-14 | 联想(北京)有限公司 | Virtual network management method and electronic equipment |
| CN111092770A (en) * | 2019-12-23 | 2020-05-01 | 联想(北京)有限公司 | Virtual network management method and electronic equipment |
| CN111190702A (en) * | 2020-01-14 | 2020-05-22 | 山东超越数控电子股份有限公司 | Cross-node hot migration method for Docker |
| CN111339061B (en) * | 2020-02-12 | 2023-09-26 | 杭州涂鸦信息技术有限公司 | Data migration method and system for distributed database |
| CN111339061A (en) * | 2020-02-12 | 2020-06-26 | 杭州涂鸦信息技术有限公司 | Data migration method and system of distributed database |
| CN111694789A (en) * | 2020-04-22 | 2020-09-22 | 西安电子科技大学 | Embedded reconfigurable heterogeneous determination method, system, storage medium and processor |
| CN116204305A (en) * | 2022-12-21 | 2023-06-02 | 山东未来网络研究院(紫金山实验室工业互联网创新应用基地) | Method for limiting number of dock container inodes |
| CN116204305B (en) * | 2022-12-21 | 2023-11-03 | 山东未来网络研究院(紫金山实验室工业互联网创新应用基地) | Method for limiting number of dock container inodes |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105607954B (en) | 2019-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105607954A (en) | Stateful container online migration method and apparatus | |
| US11429369B2 (en) | Distributed upgrade in virtualized computing environments | |
| US20170063598A1 (en) | Network functions virtualization network system and data processing method, and apparatus | |
| CN104601680B (en) | A kind of method for managing resource and device | |
| US20180024866A1 (en) | System, virtualization control apparatus, method for controlling a virtualization control apparatus, and program | |
| CN107003983A (en) | Configuration network | |
| CN102346460A (en) | Transaction-based service control system and method | |
| CN103761309A (en) | Operation data processing method and system | |
| CN102012944B (en) | Distributed NOSQL (not only structured query language) database capable of providing replication property | |
| CN109547512A (en) | A kind of method and device of the distributed Session management based on NoSQL | |
| CN106911648B (en) | Environment isolation method and equipment | |
| CN112463366A (en) | Cloud-native-oriented micro-service automatic expansion and contraction capacity and automatic fusing method and system | |
| JP2023500669A (en) | Cloud services for cross-cloud operations | |
| US20190370376A1 (en) | High-availability network device database synchronization | |
| CN104793981B (en) | A kind of online snapshot management method and device of cluster virtual machine | |
| CN112351106B (en) | Service grid platform containing event grid and communication method thereof | |
| CN110213359A (en) | A kind of car networking networking data delivery system and method based on D2D | |
| CN103139243A (en) | File synchronization method based on star distributed system | |
| WO2016198004A1 (en) | Network gambit configuration method and apparatus under virtualization technology | |
| CN112540827A (en) | Load balancing system based on k8s platform and implementation method | |
| CN116723077A (en) | Distributed IT automatic operation and maintenance system | |
| CN116346834A (en) | Session synchronization method, device, computing equipment and computer storage medium | |
| EP3306471B1 (en) | Automatic server cluster discovery | |
| CN113824801B (en) | Intelligent integration terminal unified access management component system | |
| US11290318B2 (en) | Disaster recovery of cloud resources |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |