CN110502364A - Across the cloud back-up restoring method of big data sandbox cluster under a kind of OpenStack platform - Google Patents
Across the cloud back-up restoring method of big data sandbox cluster under a kind of OpenStack platform Download PDFInfo
- Publication number
- CN110502364A CN110502364A CN201810472622.2A CN201810472622A CN110502364A CN 110502364 A CN110502364 A CN 110502364A CN 201810472622 A CN201810472622 A CN 201810472622A CN 110502364 A CN110502364 A CN 110502364A
- Authority
- CN
- China
- Prior art keywords
- cluster
- big data
- virtual machine
- backup
- cloud
- 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
- 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
- G06F11/1402—Saving, restoring, recovering or retrying
- G06F11/1446—Point-in-time backing up or restoration of persistent data
- G06F11/1448—Management of the data involved in backup or backup restore
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/4557—Distribution of virtual machine instances; Migration and load balancing
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)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention belongs to cloud computing and big data technical field, it is related to a kind of across the cloud back-up restoring method of big data sandbox cluster under the platform for OpenStack.The method of the present invention includes: to create multinode virtual machine snapshot in the big data sandbox cluster backup stage and be stored as mirror image, extracts cluster information in OpenStack-Sahara database and is saved；In across the cloud transmission stage, the reliable transmission for improving UDT agreement progress cluster backup image file is used；In the big data sandbox cluster recovery stage, virtual machine backup image is uploaded, restarts cluster using Sahara, the method by modifying cluster plug-in unit realizes cluster internal Hadoop, the starting of Spark big data service.The method of the present invention can solve in cloudy big data platform environment, and physical node system or hardware error occurs and the problem of the backup migration of virtual machine and big data sandbox cluster carried out and data are restored in order to prevent.
Description
Technical field
The invention belongs to cloud computings and big data technical field, and in particular to big data is husky under a kind of OpenStack platform
Across the cloud back-up restoring method of case cluster.
Background technique
Prior art discloses in cloud computing era, calculating, storage and Internet resources can be on-demand as traditional energy
Use, charging according to quantity；Virtualization technology aggregate server hardware resource provides a user computing resource in the form of virtual machine；So
And display is practiced, the unstable factor inside and outside data center affects the stability and reliability of computing resource supply, by this shadow
It rings, the virtual-machine fail of burst will lead to the problems such as Client application interrupts, virtual disk data damages；More seriously, it breaks
Recovery can not be restarted by often resulting in virtual machine when the virtual disk data damage of bad property.Cloud computing platform generally all supports single virtual
The backup and recovery of machine node.
As the development of big data technology and the continuous growth of data volume, single virtual machine have been unable to satisfy computing resource
It is required that multiple virtual machine instance layouts can be that a big data handles cluster by the Sahara component of OpenStack cloud platform,
Hadoop, the big datas computing platform such as Spark are run on this cluster.In face of multiple user's different data sizes, dissimilarity
Can and cost desired big data transaction and processing request, system need to be adapted to various using feature, the completion of dynamic high-efficiency is more
The automatic rapid configuration of kind computing cluster (Hadoop, Spark), the private space-for forming each user are referred to as at big data
Sandbox cluster is managed, secure and trusted and efficient big data analysis processing is carried out in the sandbox being relatively isolated；Therefore, big data
Handling sandbox cluster is carried at specific big data on the basis of big data platform shares physics and virtual computing/storage resource
Reason task, with other computing resources it is comparatively safe be isolated from complete independent virtual environment private space.And the prior art
The Backup and Restore function of original virtual machine single node is unable to satisfy big data sandbox cluster more piece field of points on OpenStack platform
Across cloud disaster-tolerant recovery requirement under scape.
Status based on the prior art, present inventor is quasi- to be directed to big data virtual machine sandbox cluster, designs a set of
The back-up restoring method of sandbox cluster across cloud.
Summary of the invention
It is an object of the invention to the statuses based on the prior art, propose a kind of for big data under OpenStack platform
Whole across the cloud Backup and Restore scheme of sandbox cluster.More particularly under a kind of OpenStack platform big data sandbox cluster across
Cloud back-up restoring method.
The method of the present invention includes: to create multinode virtual machine snapshot in the big data sandbox cluster backup stage and be stored as
Mirror image extracts cluster information in OpenStack-Sahara database and is saved；In across the cloud transmission stage, improvement UDT is used
The reliable transmission of agreement progress cluster backup image file；In the big data sandbox cluster cluster recovery stage, it is standby to upload virtual machine
Part mirror image restarts cluster using Sahara, passes through the method for modifying cluster plug-in unit, realizes cluster internal Hadoop, Spark
The starting of big data service.The method of the present invention can solve in OpenStack cloud platform environment, when physical node occur system or
When hardware error, the problem of migration of big data sandbox virtual machine cluster and data are restored.Using this method, can be effectively ensured
Big data calculates the stability of environment, provides help and support method for big data cluster disaster-tolerant recovery.
The present invention is applied existing by following technical proposals:
The present invention provides a kind of across the cloud back-up restoring methods of big data sandbox cluster under OpenStack platform, special
Sign is, realizes big data sandbox virtual machine cluster whole backup and recovery across cloud platform, the step by following step
Include:
1) the big data sandbox cluster backup stage
After receiving backup request, the snapshot of all virtual machines in big data sandbox cluster is created, saves image file；Number
The information for extracting cluster from database according to library backup module, is saved with dictionary format；
2) across cloud transmission big data sandbox cluster mirror image
It is transmitted using across the cloud data for improving UDT agreement progress big data sandbox cluster backup file；
3) the big data sandbox cluster recovery stage
After receiving recovery request, virtual machine backup file is uploaded as mirror image, while restoring database interior joint group
Information after starting to virtual machine, starts cluster virtual machine instance using big data sandbox cluster recovery template plug-in unit, then
Start original process of each node, service, completes cluster recovery operation.
In the present invention, the cluster backup includes following process:
1) cluster information is saved, comprising: cluster configuration, node group, virtual machine information, key pair；
2) cluster Template Information, the node group configuration including cluster, the service that each node is run are saved；
3) configuration information of registered mirror image, including mirror image basic information, plug-in version are saved；
4) virtual machine instance snapshot, downloading snapshot image and format transformation are created.
In the present invention, in the creation virtual machine snapshot, after snapshot creation, using qemu by the mirror of virtual machine
As format is converted to qcow2 by raw to promote mirror image upload/download efficiency.
In the present invention, in the improvement UDT agreement, by the ACK control packet using UDT agreement and retransmission list is lost
It optimizes, improves the reliability of UDT agreement；For UDT agreement asking using packet loss as congestion control signal in a high-speed network
Topic adjusts transmission speed by calculating network bandwidth capacity and round-trip delay.
In the present invention, the recovery process in the cluster recovery method is as follows:
1) confirm virtual machine image number to be restored, whether type of virtual machine matches with cluster information；
2) key pair, virtual machine instance type are uploaded；
3) cluster virtual machine instance mirror image is uploaded；
4) start each node virtual machine example；
5) by cluster configuration, node group information is imported in Sahara database；
6) Sahara component API is used, original process of each node of cluster is restarted, is serviced.
In the present invention, in the cluster plug-in unit, OpenStack-Sahara API is called, it is real to log in each node virtual machine
Example, starts the big data service processes of original Hadoop and Spark.
More specifically, in the present invention,
1. backup big data sandbox cluster
OpenStack itself can carry out creation snapshot this operation to virtual machine, for saving the current fortune of virtual machine
Row state and wherein data；
Sandbox cluster in the present invention is made of multiple virtual machine instances, virtual in cluster by taking Hadoop cluster as an example
Machine is generally divided into a Master node and multiple Slave nodes, runs NameNode on Master node,
The service such as ResourceManager runs the service such as DataNode on Slave node；It is same that Sahara component is responsible for scheduling
Being in communication with each other between one cluster interior nodes, it in the database, monitors the information unification storage of all nodes in cluster
The health status of each node safeguards the stable operation of cluster；Therefore, on the one hand backup of this method for cluster is deposited
The data in each virtual machine node are stored up, the integrated scheduling information of cluster in Sahara database is on the other hand saved；It is described
Include: in step
1) after receiving backup request, according to the ID of cluster, corresponding cluster information is found, in data with dictionary format
It saves, comprising: cluster configuration, node group, virtual machine information, key pair；
2) cluster Template Information, the node group configuration including cluster, the Hadoop that each node is run, Spark are saved
Service；
3) configuration information of registered mirror image, safe group information are saved；
4) all virtual machine instance snapshots of cluster are created using OpenStack Nova component, Nova is by the snapshot of virtual machine
It is stored in the form of mirror image in Glance component；Snapshot image file is downloaded from Glance component, this image file is big
The small disk size for virtual machine rather than real data size, reuse qemu-img tool and are converted to the mirror image of raw format
Qcow2, mirror image size is the size of data file in magnetic disk of virtual machine after conversion, so far completes backup operation；
2. across cloud transmission big data sandbox cluster mirror image
The Backup and Restore operation that cluster is carried out between different cloud platforms, needs to carry out a large amount of data transmission；UDT agreement
It is the agreement suitable for the transmission of high-speed wan big data, UDT agreement has transmission speed fast, stable in the environment of kilomega network
Etc. advantages, but in 10,000,000,000 net environment, that there are transmission performances is low for UDT agreement, continues the problems such as packet loss；
This method causes Retransmission timeout for UDT (UDP-based Data Transfer Protocol) agreement packet loss
Problem passes through ACK control packet in the UDT agreement based on BDP (Bandwidth-Delay Product, network bandwidth time delay product)
It is optimized with retransmission list is lost, improves the reliability of UDT agreement；Include: in the step
1) retransmission mechanism improves
This method improves the reliability control algorithm of UDT, can find the re-transmission data lost in link in time
Packet, and make full use of ACK to control package informatin, avoid Retransmission timeout and retransmit received data packet；Receiving end passes through setting meter
Number device is united the number of identical ACK control packet, is reached to retransmit after current network defined threshold and is sent NAK control packet and inform transmitting terminal
Packet drop retransmits unacknowledged data packet, as shown in Figure 2；
In the present invention,
In algorithm 1, input be receiving end send a series of ACK control packet sequence number ACK1, ACK2 ...,
ACKn }, TAsameIt indicates to send the threshold value that identical ACK control is wrapped, RAsameFor the number of the identical ACK sequence number of counters count,
Counter is increased if currently transmitted ACK control packet ACKi is identical as the ACKi-1 sent before, if not identical, in time
It resets.Work as RAsameMore than TAsameWhen, it reads receiving end and loses data packet in list RLL, retransmit NAK control packet and (preferentially select sequence
The small continuous data packet of row number)；Define network bandwidth time delay product BDP=Bwbtl×Rttlink, pass through network bandwidth BwbtlWith
Round-trip delay RttlinkIndicate link in can transmitted data amount maximum value, with this control congestion window size and limitation link in
The data volume of transmission；TAsameRelated to the BDP value of current network environment, when BDP is larger, the data packet of re-transmission may also exist
It is transmitted in network, TAsameBiggish value should be obtained；
2) file writing process optimizes
In this method: the data packet that consecutive order reaches fixed number being continually stored in circle queue, and is write together
Enter into file；The continuous maximum fixed number for reaching data packet of storage is FP every time for setting first；Then, when the receiving end UDT
The continuous data packet of FP serial number is received, is stored in circle queue Unit one；Finally, leading in the operation of write-in file
It crosses a written document system calling data information of all data packets in this unit is written in file；
In the present invention,
In the main thread of the receiving end UDT after optimization, the data received in buffer area can be sequentially written in file,
The detailed process of write-in is: obtaining the pointer that the minmal sequence number of file is not written in receiving buffer area first, and finds this
Position in the circle queue that a pointer is directed toward, marker bit 0, this indicates the starting position of Coutinuous store；Then the block is calculated
The size of this data packet is added to summation if subsequent packet marking position is 1 by the size in Coutinuous store region
In, if the marker bit of subsequent data packet is 0, indicates that Coutinuous store region is interrupted, then write the region of the block Coutinuous store
Enter into file；It has confirmed that received ordinal position updates finally, will receive in buffer area, and continuous data packet size will be counted
Summation is set to 0 byte；
In the present invention, the data packet in data buffer zone is written to the realization process in file as shown in algorithm 3:
3. restoring big data sandbox cluster
Across the cloud Backup and Restore of cluster will lead to original network environment and change, and OpenStack can be newly-established void
The new example ID of quasi- machine example allocation, IP address etc., therefore should guarantee the complete of all node datas, while also to guarantee
After recovery operation is completed, between each node of cluster can normal communication, original service can operate normally；The step
Include: in rapid
1) after receiving recovery request, confirm virtual machine image number to be restored, whether are type of virtual machine and cluster information
Matching；
2) when being related to the Backup and Restore across cloud platform, the authentication information of user can change, and it is original that cluster can be used
Key pair logs in, and can also create the original key pair of new replacement；
3) cluster virtual machine instance mirror image is uploaded；
4) start virtual machine instance, using the new example ID of node, the information such as IP address, original collection in replacement data library
Group's information；
5) by cluster configuration, node group information is imported in Sahara database；
6) Sahara component API is used, starts the service processes of corresponding Hadoop and Spark on each node, so far
Complete recovery operation.
The method of the present invention is through practical proof, the results show that can solve in OpenStack cloud platform environment, when physics section
The problem of point is when there is system or hardware error, and the migration of big data sandbox virtual machine cluster and data are restored.Utilize we
Method can be effectively ensured the stability that big data calculates environment, provide help and support method for big data cluster disaster-tolerant recovery.
Detailed description of the invention
Fig. 1 is overall workflow figure.
Fig. 2 is that receiving end retransmits NAK control packet flow chart.
Specific embodiment
Embodiment 1
Experimental section of the invention will carry out a big data sandbox cluster across cloud in two OpenStack cloud platforms
Backup and Restore operation.
Cloud platform environment is as follows:
1. cluster backup OpenStack platform: OpenStack Ocata version, a control node, five calculating sections
Point.The address ip of control node is 10.141.105.19.
2. cluster recovery OpenStack platform: OpenStack Queens version, a control node, five calculating sections
Point.The address ip of control node is 10.141.105.14.
Cluster information to be operated is as follows:
1.Master node, 1:
Virtual machine configuration: CPU:2, memory: 4GB, disk 40GB
The service processes of operation:
·NameNode
·SecondaryNameNode
·ResourceManager
·HistoryServer
·oozie
·HiveServer
·Spark History Server
2.Slave node, 3:
Virtual machine configuration: CPU:2, memory: 4GB, disk 40GB
The service processes of operation:
·DataNode
·NodeManager
First to http: // 10.141.105.19/sandbox/backup sends the request comprising cluster ID, opens
Beginning backup operation.
In this process, firstly, according to cluster ID, the information of cluster, including cluster are searched and saved from database
Template, clustered node group template, cluster key pair, the type of virtual machine of clustered node, cluster safety group information and cluster mirror
As information, these information are saved hereof with json format, are reacquired when in case restoring；
Later, the virtual machine instance for starting backup clustered node is called according to the corresponding virtual machine ID of clustered node
The API of OpenStack Nova component creates the snapshot image of virtual machine, then passes through the API of OpenStack Glance component,
The image file that format is raw is obtained, the size of raw format mirror image is equal to magnetic disk of virtual machine and presets size, in this example, empty
Quasi- machine disk size is 40GB；
Next, raw format mirror image is converted to the mirror image of qcow2 format using qemu-img tool, after conversion,
The size of image file becomes the actual size of data file in disk, later using UDT agreement by cluster information file and
Virtual machine image file is transmitted to the host that IP address is 10.141.105.14, prepares recovery operation；
After file end of transmission, recovery request is sent to http: // 10.141.105.14/sandbox/recovery,
Including cluster ID, Neutron network ID, cipher key pair information starts the cluster recovery stage,
Upload virtual machine first is mirrored to Glance, then once restores key pair, node group, cluster in the database
Information update the IP address of each node while according to the scheduling of Nova, finally call Sahara API restarting cluster former
There are service processes, completes recovery operation；
The results show that Cao Yong the method for the present invention can solve in OpenStack cloud platform environment, when physical node system
The problem of system or when hardware error, the migration of big data sandbox virtual machine cluster and data are restored.
Claims (6)
1. across the cloud back-up restoring method of big data sandbox cluster under a kind of OpenStack platform, which is characterized in that by following
Step carries out the backup and recovery of across cloud data center realization big data sandbox cluster, and steps are as follows:
1) the big data sandbox cluster backup stage
After receiving backup request, the snapshot of all virtual machines in big data sandbox cluster is created, saves image file；Database
Backup module extracts the information of cluster from database, is saved with dictionary format；
2) across cloud transmission big data sandbox cluster mirror image
It is transmitted using across the cloud data for improving UDT agreement progress big data sandbox cluster backup file；
3) the big data sandbox cluster recovery stage
After receiving recovery request, virtual machine backup file is uploaded as mirror image, while restoring the information of database interior joint group,
After to virtual machine starting, starts cluster virtual machine instance using big data sandbox cluster recovery template plug-in unit, then start
Cluster recovery operation is completed in original process of each node, service.
2. across the cloud back-up restoring method of big data sandbox cluster under OpenStack platform according to claim 1, special
Sign is that the cluster backup includes following process:
1) cluster information is saved, comprising: cluster configuration, node group, virtual machine information, key pair；
2) cluster Template Information, the node group configuration including cluster, the service that each node is run are saved；
3) configuration information of registered mirror image, including mirror image basic information, plug-in version are saved；
4) virtual machine instance snapshot, downloading snapshot image and format transformation are created.
3. across the cloud back-up restoring method of big data sandbox cluster under OpenStack platform according to claim 2, special
Sign is, in the creation virtual machine snapshot, snapshot creation after, using qemu by the mirror-image format of virtual machine by
Raw is converted to qcow2 to promote mirror image upload/download efficiency.
4. across the cloud back-up restoring method of big data sandbox cluster under OpenStack platform according to claim 1, special
Sign is, in the improvement UDT agreement, wraps and loses retransmission list by the ACK control using UDT agreement and optimize,
Improve the reliability of UDT agreement；UDT agreement in a high-speed network using packet loss as congestion control signal aiming at the problem that, pass through meter
It calculates network bandwidth capacity and round-trip delay adjusts transmission speed.
5. across the cloud back-up restoring method of big data sandbox cluster under OpenStack platform according to claim 1, special
Sign is that the recovery process in the cluster recovery method is as follows:
1) confirm virtual machine image number to be restored, whether type of virtual machine matches with cluster information；
2) key pair, virtual machine instance type are uploaded；
3) cluster virtual machine instance mirror image is uploaded；
4) start each node virtual machine example；
5) by cluster configuration, node group information is imported in Sahara database；
6) Sahara component API is used, original process of each node of cluster is restarted, is serviced.
6. across the cloud back-up restoring method of big data sandbox cluster under OpenStack platform according to claim 2, special
Sign is, in the cluster plug-in unit, calls OpenStack-Sahara API, logs in each node virtual machine example, and starting is former
The big data service processes of some Hadoop and Spark.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810472622.2A CN110502364B (en) | 2018-05-17 | 2018-05-17 | Cross-cloud backup recovery method for big data sandbox cluster under OpenStack platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810472622.2A CN110502364B (en) | 2018-05-17 | 2018-05-17 | Cross-cloud backup recovery method for big data sandbox cluster under OpenStack platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110502364A true CN110502364A (en) | 2019-11-26 |
| CN110502364B CN110502364B (en) | 2023-03-14 |
Family
ID=68583955
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810472622.2A Active CN110502364B (en) | 2018-05-17 | 2018-05-17 | Cross-cloud backup recovery method for big data sandbox cluster under OpenStack platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110502364B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111078667A (en) * | 2019-12-12 | 2020-04-28 | 腾讯科技（深圳）有限公司 | Data migration method and related device |
| CN111726403A (en) * | 2020-06-11 | 2020-09-29 | 深圳市赛宇景观设计工程有限公司 | Cross-cloud-platform big data management method and system |
| CN111797059A (en) * | 2020-07-07 | 2020-10-20 | 优刻得科技股份有限公司 | Method for online cloud host migration across clouds, migration controller and cloud server |
| CN112084491A (en) * | 2020-08-26 | 2020-12-15 | 天津七一二通信广播股份有限公司 | Sandbox mechanism-based cluster virtual user system implementation method |
| CN112929609A (en) * | 2021-02-05 | 2021-06-08 | 中国电信集团工会上海市委员会 | Video monitoring system based on cloud platform |
| CN116028277A (en) * | 2023-03-27 | 2023-04-28 | 广州智算信息技术有限公司 | Database backup method and system based on CDC mode |
| CN116049136A (en) * | 2022-12-21 | 2023-05-02 | 广东天耘科技有限公司 | Cloud computing platform-based MySQL cluster deployment method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102521071A (en) * | 2011-11-24 | 2012-06-27 | 广州杰赛科技股份有限公司 | Private cloud-based virtual machine maintaining method |
| CN102834822A (en) * | 2010-04-12 | 2012-12-19 | 微软公司 | Express-full backup of a cluster shared virtual machine |
| CN106557355A (en) * | 2016-12-01 | 2017-04-05 | 北京奇虎科技有限公司 | The generation method and generating means of virtual machine image |
| US9639428B1 (en) * | 2014-03-28 | 2017-05-02 | EMC IP Holding Company LLC | Optimized backup of clusters with multiple proxy servers |
| CN107197022A (en) * | 2017-06-02 | 2017-09-22 | 华南理工大学 | OpenStack storage optimization methods and system |
| US9842124B1 (en) * | 2013-03-13 | 2017-12-12 | EMC IP Holding Company LLC | Highly available cluster agent for backup and restore operations |
-
2018
- 2018-05-17 CN CN201810472622.2A patent/CN110502364B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102834822A (en) * | 2010-04-12 | 2012-12-19 | 微软公司 | Express-full backup of a cluster shared virtual machine |
| CN102521071A (en) * | 2011-11-24 | 2012-06-27 | 广州杰赛科技股份有限公司 | Private cloud-based virtual machine maintaining method |
| US9842124B1 (en) * | 2013-03-13 | 2017-12-12 | EMC IP Holding Company LLC | Highly available cluster agent for backup and restore operations |
| US9639428B1 (en) * | 2014-03-28 | 2017-05-02 | EMC IP Holding Company LLC | Optimized backup of clusters with multiple proxy servers |
| CN106557355A (en) * | 2016-12-01 | 2017-04-05 | 北京奇虎科技有限公司 | The generation method and generating means of virtual machine image |
| CN107197022A (en) * | 2017-06-02 | 2017-09-22 | 华南理工大学 | OpenStack storage optimization methods and system |
Non-Patent Citations (1)
| Title |
|---|
| JIAQING WANG ET AL.: "Construction and application of Red5 cluster based on OpenStack", 《AIP CONFERENCE PROCEEDINGS》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111078667A (en) * | 2019-12-12 | 2020-04-28 | 腾讯科技（深圳）有限公司 | Data migration method and related device |
| CN111726403A (en) * | 2020-06-11 | 2020-09-29 | 深圳市赛宇景观设计工程有限公司 | Cross-cloud-platform big data management method and system |
| CN111726403B (en) * | 2020-06-11 | 2021-01-29 | 和宇健康科技股份有限公司 | Cross-cloud-platform big data management method and system |
| CN111797059A (en) * | 2020-07-07 | 2020-10-20 | 优刻得科技股份有限公司 | Method for online cloud host migration across clouds, migration controller and cloud server |
| CN112084491A (en) * | 2020-08-26 | 2020-12-15 | 天津七一二通信广播股份有限公司 | Sandbox mechanism-based cluster virtual user system implementation method |
| CN112929609A (en) * | 2021-02-05 | 2021-06-08 | 中国电信集团工会上海市委员会 | Video monitoring system based on cloud platform |
| CN116049136A (en) * | 2022-12-21 | 2023-05-02 | 广东天耘科技有限公司 | Cloud computing platform-based MySQL cluster deployment method and system |
| CN116049136B (en) * | 2022-12-21 | 2023-07-28 | 广东天耘科技有限公司 | Cloud computing platform-based MySQL cluster deployment method and system |
| CN116028277A (en) * | 2023-03-27 | 2023-04-28 | 广州智算信息技术有限公司 | Database backup method and system based on CDC mode |
| CN116028277B (en) * | 2023-03-27 | 2023-07-28 | 广州智算信息技术有限公司 | Database backup method and system based on CDC mode |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110502364B (en) | 2023-03-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110502364A (en) | Across the cloud back-up restoring method of big data sandbox cluster under a kind of OpenStack platform | |
| US9888048B1 (en) | Supporting millions of parallel light weight data streams in a distributed system | |
| EP3482298B1 (en) | Multicast apparatuses and methods for distributing data to multiple receivers in high-performance computing and cloud-based networks | |
| WO2021142971A1 (en) | Transmission rate control method and apparatus, computer system, and readable storage medium | |
| US11025564B2 (en) | RDMA transport with hardware integration and out of order placement | |
| US20050141425A1 (en) | Method, system, and program for managing message transmission through a network | |
| US9667729B1 (en) | TCP offload send optimization | |
| CN106874143B (en) | Server backup method and backup system thereof | |
| US11068412B2 (en) | RDMA transport with hardware integration | |
| US20140317071A1 (en) | Method and device for transferring file | |
| US11303737B2 (en) | Method and device for data transmission | |
| JPH11143845A (en) | System and method for message transmission between network nodes | |
| CN114201421B (en) | Data stream processing method, storage control node and readable storage medium | |
| US7457845B2 (en) | Method and system for TCP/IP using generic buffers for non-posting TCP applications | |
| CN110609746B (en) | Method, apparatus and computer readable medium for managing network system | |
| US10521317B1 (en) | Compressing data to be replicated utilizing a compression method selected based on network behavior | |
| WO2016099357A1 (en) | Methods and first network node for managing a stream control transmission protocol association | |
| US20070291782A1 (en) | Acknowledgement filtering | |
| JP2019106697A (en) | Method for dynamically managing message retransmission delay in interconnection network and device | |
| CN110445580B (en) | Data transmission method and device, storage medium, and electronic device | |
| US11444882B2 (en) | Methods for dynamically controlling transmission control protocol push functionality and devices thereof | |
| US7159005B1 (en) | Methods, systems and computer program products for restartable multiplexed file transfers | |
| CN115065694A (en) | Cloud storage data transfer uploading system, method, equipment and medium | |
| CN115914380A (en) | Communication delay optimization method of cloud computing resource manager based on ZLib compression algorithm | |
| WO2022141711A1 (en) | Method for asynchronous execution of transaction in blockchain, system and related device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |