CN108304281A - A kind of method that virtual machine is backed up in realtime - Google Patents
A kind of method that virtual machine is backed up in realtime Download PDFInfo
- Publication number
- CN108304281A CN108304281A CN201810166485.XA CN201810166485A CN108304281A CN 108304281 A CN108304281 A CN 108304281A CN 201810166485 A CN201810166485 A CN 201810166485A CN 108304281 A CN108304281 A CN 108304281A
- Authority
- CN
- China
- Prior art keywords
- virtual machine
- backup
- host
- machine
- sourcesink
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/14—Error detection or correction of the data by redundancy in operation
- G06F11/1479—Generic software techniques for error detection or fault masking
- G06F11/1482—Generic software techniques for error detection or fault masking by means of middleware or OS functionality
- G06F11/1484—Generic software techniques for error detection or fault masking by means of middleware or OS functionality involving virtual machines
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Hardware Redundancy (AREA)
Abstract
The invention discloses a kind of method that virtual machine is backed up in realtime, the physical machines where virtual machine operation, referred to as sourcesink host system；Physical machine where virtual machine backup, referred to as backs up host machine system；After virtual machine starts, a file identical with virtual machine size is created on backup host machine system；In virtual machine operational process, backup host machine system is copied data to by sourcesink host system, ensures that backup file and virtual machine file are completely the same.When sourcesink host system breaks down, virtual machine backup can continue to run on backup host machine system.Since the write request of virtual machine is in a manner of network message on real-time synchronization to backup host machine system, you can realize that the continuity of virtual machine operation ensure that the continuity of business for virtual server.Present invention can apply in various virtualization systems, be with a wide range of applications.
Description
Technical field
The present invention relates to the methods that virtual machine in a kind of cloud computing is backed up in realtime, when the physical machine where wherein virtual machine
When failure, virtual machine automatic running and can provide service in other physical machines.
Background technology
With the development of cloud computing, the function of internet is stronger and stronger, and user can be by cloud computing on the internet
It handles huge data and obtains required information.
In traditional server virtualization framework, the mode for storing dilatation is to increase disk, expansion capacity, but property
It can fail to be promoted, and can also reduce storage performance in Data Migration.In super fusion architecture, storage dilatation way is to add
Add physical machine node, performance and capacity dilatation together, the distribution of data automatic equalization.
Since storage is not independent distributed storage, it is therefore desirable to disaster tolerance and the backup for considering virtual machine, it can be in void
In the case of physical machine failure where quasi- machine, virtual machine can be restored in time, offer service is provided, and ensure data not
It loses.
Invention content
Goal of the invention：The technical problem to be solved by the present invention is in super fusion architecture, storage and performance are in the same of dilatation
When, a kind of method that virtual machine is backed up in realtime is provided, to ensure that service is not interrupted, data are not lost.
The English of " super fusion " is Hyper-Converged, corresponding " virtualization architecture ", earliest derived from storage establishment factory
The framework for the calculating storage fusion that the internets such as Google, Facebook manufacturer uses is used for virtualized environment by quotient, is client
A kind of fusion product accomplished storage in calculation server is provided, core change is storage.
By calculate virtualization, Storage Virtualization, network virtualization by inside every server computing resource, store
Resource, network resource integration provide required calculating, storage, net at a big resource pool for data center traffic system
The IT infrastructure such as network, safety, while management and flexible allocation these resources are facilitated by a system.But distribution is deposited
Framework is stored up in actual network environment, brings performance bottleneck.Increase with the quantity of physical machine, storage performance will be limited
In the throughput performance of network.The method that virtual machine in the present invention is backed up in realtime is built upon on the framework of non-distributed storage.
For the computing resource and storage resource of virtual machine in same physical machine, storage performance will be no longer by the throughput performance of network
Limitation.
More virtual machines share same physical machine, and for virtual machine, it operates in the simulation of virtual management software
In hardware system out.And for physical machine, it is the host machine system of more virtual machines operation.When host system
When system breaks down during use, such as：Network interruption, shutdown, restarts, virtual machine is nothing in this case
Method normal operation.For the virtual machine of server, the case where service disconnection being faced, service can not be provided.
So the method that virtual machine provided by the invention is backed up in realtime, it can be when host machine system breaks down, virtually
Machine-cut, which is changed on other host machine systems, starts and continues offer service, ensures the continuity of business.
Technical solution：A method of virtual machine is backed up in realtime in non-distributed storage architecture, where virtual machine operation
Physical machine, referred to as sourcesink host system.Physical machine where virtual machine backup, referred to as backs up host machine system.Virtual machine opens
After dynamic, a file identical with virtual machine size is created on backup host machine system；In virtual machine operational process, pass through
Sourcesink host system copies data to backup host machine system, ensures that backup file and virtual machine file are completely the same.
Wherein sourcesink host system number and backup host machine system number ratio can be 1：1, can also be 1： N(N
It is greater than 1 integer), so backup host machine system can be more.
It is assumed that sourcesink host system number and backup host machine system number ratio are 1：1, after virtual machine starts, standby
A file identical with virtual machine size is created on part host machine system.Write request in virtual machine operational process can be by
It intercepts and captures and replicates, be sent to sourcesink host system.Backup host is sent in the form of network package by sourcesink host system
Machine system.After backup host machine system receives the message of virtual machine write request, be written local disk in, ensure backup and
Virtual machine file is completely the same.Wherein procotol uses reliable network transport protocol, such as Transmission Control Protocol, to ensure backup file
Integrality.
When sourcesink host system breaks down, backup virtual machine can start on backup host machine system, and standby
Self-role is switched to new sourcesink host system by part host machine system.And old sourcesink host system is after restoring normal,
The backup host machine system of role switching Cheng Xin.Continue to repeat process above, ensure that virtual machine in the process of running, it is standby
Part always exists and real time data synchronization.
When backup ratio is 1：When N (N is greater than 1 integer), in order to reduce network load, sourcesink host system is in handle
When virtual machine write request is sent to more backup host machine systems, sent by the way of multicast.But since multicast protocol is
Non- reliable protocol, during transmission, packet loss causes backup virtual machine imperfect to data in order to prevent.So using a kind of
Improved Reliable Multicast Protocol come realize virtual machine request network transmission.
Advantageous effect：Compared with prior art, the method that a kind of virtual machine provided by the invention is backed up in realtime, and back up
Ratio is 1：M (M is greater than the integer equal to 1).When sourcesink host system breaks down, virtual machine backup can backed up
It is continued to run on host machine system.Due to virtual machine write request in a manner of network message real-time synchronization to backup host
In system, you can realize that the continuity of virtual machine operation ensure that the continuity of business for virtual server.It can be with
Referred to as, never offline server.Present invention can apply in various virtualization systems, be with a wide range of applications.
Description of the drawings
Fig. 1 is Sequence of Real-time Backups under virtual machine operating status of the present invention.
When Fig. 2 is sourcesink host system failure of the present invention, backup virtual machine starts sequence diagram.
Fig. 3 is reliability multicast protocol data transmission sequence diagram of the present invention.
Specific implementation mode
With reference to specific embodiment, the present invention is furture elucidated, it should be understood that these embodiments are merely to illustrate the present invention
Rather than limit the scope of the invention, after having read the present invention, various equivalences of the those skilled in the art to the present invention
The modification of form falls within the application range as defined in the appended claims.
A method of virtual machine is backed up in realtime in non-distributed storage architecture, the physical machine where virtual machine operation,
Referred to as sourcesink host system.Physical machine where virtual machine backup, referred to as backs up host machine system.After virtual machine starts, standby
A file identical with virtual machine size is created on part host machine system.Wherein sourcesink host system number and backup host
Machine system number ratio can be 1：1, can also be 1：N (N is greater than 1 integer), so backup host machine system can be
More.
It is assumed that sourcesink host system number and backup host machine system number ratio are 1：1, after virtual machine starts, standby
A file identical with virtual machine size is created on part host machine system.Write request in virtual machine operational process can be by
It intercepts and captures and replicates, be sent to sourcesink host system.Backup host is sent in the form of network package by sourcesink host system
Machine system.After backup host machine system receives the message of virtual machine write request, be written local disk in, ensure backup and
Virtual machine file is completely the same.Wherein procotol uses reliable network transport protocol, such as Transmission Control Protocol, to ensure backup file
Integrality.
When sourcesink host system breaks down, backup virtual machine can start on backup host machine system, and standby
Self-role is switched to new sourcesink host system by part host machine system.And old sourcesink host system is after restoring normal,
The backup host machine system of role switching Cheng Xin.Continue to repeat process above, ensure that virtual machine in the process of running, it is standby
Part always exists and real time data synchronization.
When backup ratio is 1：When N (N is greater than 1 integer), in order to reduce network load, sourcesink host system is in handle
When virtual machine write request is sent to more backup host machine systems, sent by the way of multicast.But since multicast protocol is
Non- reliable protocol, during transmission, packet loss causes backup virtual machine imperfect to data in order to prevent.So using a kind of
Improved Reliable Multicast Protocol come realize virtual machine request network transmission.
As shown in Figure 1, Sequence of Real-time Backups under virtual machine operating status.It is assumed that backup ratio is 1：1, flow description
It is as follows：
Step 101, before virtual machine starts, sourcesink host system sends virtual machine information to backup host machine system；
Step 102, after backup host machine system creates an equal amount of virtual machine file (also referred to as backup virtual machine),
Return success message sourcesink host system；
Step 103, sourcesink host system starts virtual machine, and intercepts write request when virtual machine operation；
Step 104, sourcesink host system replicates the write request of interception a, and is sent in a manner of network message
Backup host machine system；
Step 105, backup host machine system receives write request message, parses and the virtual machine being pre-created is written
In file；
Step 106, backup host machine system returns to write-in result and gives sourcesink host system；
Step 107, if write-in result success, sourcesink host system delete the message from memory；
Step 108, if write-in result is non-successful, sourcesink host system will retransmit the message.
As shown in Fig. 2, when sourcesink host system failure, backup virtual machine starts sequence diagram.It is assumed that backup ratio is 1：
1, flow is described as follows：
Step 201, backup host machine system detection source host machine system state；
Step 202, when detecting sourcesink host system failure, start backup virtual machine；
Step 203, backup host machine system starts virtual machine, and intercepts write request when virtual machine operation；
Step 204, continue the state of detection source host machine system；
Step 205, host role switches to sourcesink host system from backup host machine system, and will be in write request
Appearance is recorded in local log file；
Step 206, after old sourcesink host system failure is restored, new sourcesink host system is returned message to, and by angle
Color switches to new backup host machine system；
Step 207, new sourcesink host system is read newest according to local journal file from virtual machine file
Content；
Step 208, new sourcesink host system is sent to new backup host machine system by newest content is read；
Step 209, after new backup host machine system receives write request message, in the virtual machine file before failure
Write operation is continued to execute, is ensured and new sourcesink host system real-time synchronization.
When backup ratio is 1：When N (N is greater than 1 integer), the backup host machine system of one of operation is selected to make
For new sourcesink host system, flow is the same as step 201- steps 209.
As shown in figure 3, reliability multicast protocol data transmission sequence diagram, flow are described as follows：
Step 301, write request message is sent to multicast address by sourcesink host system；
Step 302, the content of write request issues group member's (backup host machine system) in a manner of multicast；
Step 303, it is standby to be issued each by sourcesink host system in a manner of TCP for the message information of each write request
Part host machine system.The response message of backup host machine system reply is received simultaneously.If reply is non-successful, by the message
It is retransmitted in a manner of multicast, is equivalent to repetition step 301 and 302.
In conclusion the present invention provides the implementation method that virtual machine is backed up in realtime, technique can be applied to gold
Melt, government, colleges and universities, all conglomeraties such as electric business website, for example, the cloud system for government affairs that government carries out energetically, can apply the present invention
Important virtual machine in cloud system for government affairs passes through this in this way no matter what failure the host machine system where virtual machine occurs
Virtual machine real time backup method in invention can ensure that virtual machine continues to run with, and data are not lost, it is ensured that government affairs
The high availability of cloud service.The interests of government are not only protected, more ensure that the legitimate rights and interests of the numerous common people, therefore, this technology
With very high promotional value.
Claims (7)
1. a kind of method that virtual machine is backed up in realtime, it is characterised in that：Physical machine where virtual machine operation, referred to as sourcesink host
System；Physical machine where virtual machine backup, referred to as backs up host machine system；After virtual machine starts, in backup host machine system
One file identical with virtual machine size of upper establishment；In virtual machine operational process, data are replicated by sourcesink host system
To backup host machine system, ensure that backup file and virtual machine file are completely the same.
2. the method that virtual machine as described in claim 1 is backed up in realtime, it is characterised in that：The sourcesink host system number and
Backup host machine system number ratio is 1：1 or 1：N, N are greater than 1 integer.
3. the method that virtual machine as claimed in claim 2 is backed up in realtime, it is characterised in that：If sourcesink host system number and
Backup host machine system number ratio is 1：1, after virtual machine starts, creates one on backup host machine system and virtual machine is big
Small identical file；Write request in virtual machine operational process is trapped and replicates, and is sent to sourcesink host system；By sourcesink master
Machine system is sent to backup host machine system in the form of network package；It is asked when backup host machine system receives virtual machine and writes
It after the message asked, is written in local disk, ensures that backup and virtual machine file are completely the same.
4. the method that virtual machine as claimed in claim 3 is backed up in realtime, it is characterised in that：Sourcesink host system is sealed using network
The form of packet is sent to backup host machine system, and the wherein network transmission protocol uses Transmission Control Protocol.
5. the method that virtual machine as described in claim 1 is backed up in realtime, it is characterised in that：When sourcesink host system breaks down
When, backup virtual machine starts on backup host machine system, and backs up host machine system and self-role is switched to new source
Host machine system；And old sourcesink host system is after restoring normal, the backup host machine system of role switching Cheng Xin ensure that
In the process of running, backup always exists and real time data synchronization for virtual machine.
6. the method that virtual machine as described in claim 1 is backed up in realtime, it is characterised in that：When backup ratio is 1：When N, in order to
Network load is reduced, virtual machine write request when being sent to more backup host machine systems by sourcesink host system, using multicast
Mode send.
7. the method that virtual machine as claimed in claim 6 is backed up in realtime, it is characterised in that：Multicast protocol is：Step 1, sourcesink
Write request message is sent to multicast address by host system；
Step 2, the content of write request issues group member in a manner of multicast, and the group member is N number of backup host system
System；
Step 303, the message information of each write request is issued each backup host by sourcesink host system in a manner of TCP
Machine system；The response message of backup host machine system reply is received simultaneously；If reply is non-successful, by the message with multicast
Mode is retransmitted, and repetition step 301 and 302 are equivalent to.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810166485.XA CN108304281A (en) | 2018-02-28 | 2018-02-28 | A kind of method that virtual machine is backed up in realtime |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810166485.XA CN108304281A (en) | 2018-02-28 | 2018-02-28 | A kind of method that virtual machine is backed up in realtime |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108304281A true CN108304281A (en) | 2018-07-20 |
Family
ID=62848712
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810166485.XA Pending CN108304281A (en) | 2018-02-28 | 2018-02-28 | A kind of method that virtual machine is backed up in realtime |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108304281A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108845896A (en) * | 2018-07-10 | 2018-11-20 | 中国建设银行股份有限公司 | Component-tracking system and method after disaster |
| CN111240889A (en) * | 2018-11-29 | 2020-06-05 | 阿里巴巴集团控股有限公司 | Data backup system and method and electronic device |
| WO2021008550A1 (en) * | 2019-07-16 | 2021-01-21 | 中兴通讯股份有限公司 | Method, device, and system for remote disaster tolerance |
| CN114185640A (en) * | 2021-11-04 | 2022-03-15 | 北京伟联科技有限公司 | Virtual machine management system based on industrial private cloud |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105812169A (en) * | 2014-12-31 | 2016-07-27 | 华为软件技术有限公司 | Host and standby machine switching method and device |
| CN107038096A (en) * | 2016-12-30 | 2017-08-11 | 平安科技（深圳）有限公司 | Virtual disk standby system, method and apparatus |
-
2018
- 2018-02-28 CN CN201810166485.XA patent/CN108304281A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105812169A (en) * | 2014-12-31 | 2016-07-27 | 华为软件技术有限公司 | Host and standby machine switching method and device |
| CN107038096A (en) * | 2016-12-30 | 2017-08-11 | 平安科技（深圳）有限公司 | Virtual disk standby system, method and apparatus |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108845896A (en) * | 2018-07-10 | 2018-11-20 | 中国建设银行股份有限公司 | Component-tracking system and method after disaster |
| CN111240889A (en) * | 2018-11-29 | 2020-06-05 | 阿里巴巴集团控股有限公司 | Data backup system and method and electronic device |
| WO2021008550A1 (en) * | 2019-07-16 | 2021-01-21 | 中兴通讯股份有限公司 | Method, device, and system for remote disaster tolerance |
| CN114185640A (en) * | 2021-11-04 | 2022-03-15 | 北京伟联科技有限公司 | Virtual machine management system based on industrial private cloud |
| CN114185640B (en) * | 2021-11-04 | 2023-02-17 | 北京伟联科技有限公司 | Virtual machine management system based on industrial private cloud |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7613749B2 (en) | System and method for application fault tolerance and recovery using topologically remotely located computing devices | |
| US11755232B2 (en) | Transferring of snapshot data blocks to a virtual storage volume | |
| CN108304281A (en) | A kind of method that virtual machine is backed up in realtime | |
| US10073902B2 (en) | Snapshot and replication of a multi-stream application on multiple hosts at near-sync frequency | |
| US20120271797A1 (en) | Reference volume for initial synchronization of a replicated volume group | |
| US8539087B2 (en) | System and method to define, visualize and manage a composite service group in a high-availability disaster recovery environment | |
| CN113641511B (en) | Message communication method and device | |
| US20140229763A1 (en) | Disk fault tolerance method, device and system | |
| JP2005535241A (en) | Method of moving application software in multicomputer architecture, multicomputer method and apparatus for realizing continuity of operation using the moving method | |
| US11461123B1 (en) | Dynamic pre-copy and post-copy determination for live migration between cloud regions and edge locations | |
| CN113010496A (en) | Data migration method, device, equipment and storage medium | |
| US11223678B2 (en) | Establishing paths between servers in a copy environment for different copy relationships between the servers | |
| CN101834904A (en) | Method and equipment for database backup | |
| US8683258B2 (en) | Fast I/O failure detection and cluster wide failover | |
| CN109254873B (en) | Data backup method, related device and system | |
| US11734038B1 (en) | Multiple simultaneous volume attachments for live migration between cloud regions and edge locations | |
| US11573839B1 (en) | Dynamic scheduling for live migration between cloud regions and edge locations | |
| US11947425B2 (en) | Storage volume snapshot object management | |
| CN112187523A (en) | Network high-availability implementation method and super-convergence system | |
| US9342419B2 (en) | Persistent messaging mechanism | |
| US11238010B2 (en) | Sand timer algorithm for tracking in-flight data storage requests for data replication | |
| CN113821384A (en) | Cross-data-center and same-city disaster recovery method, device and equipment based on cloud platform | |
| Georgiou et al. | Self-stabilization overhead: an experimental case study on coded atomic storage | |
| CN105512321A (en) | Method for writing data of storage area networking (SAN) shared-file system and device | |
| US10185503B1 (en) | Consistency group fault tolerance |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180720 |