CN107147511A - Data center's control method and device - Google Patents
Data center's control method and device Download PDFInfo
- Publication number
- CN107147511A CN107147511A CN201610117910.7A CN201610117910A CN107147511A CN 107147511 A CN107147511 A CN 107147511A CN 201610117910 A CN201610117910 A CN 201610117910A CN 107147511 A CN107147511 A CN 107147511A
- Authority
- CN
- China
- Prior art keywords
- data center
- data
- center
- virtual machine
- run
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1095—Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0654—Management of faults, events, alarms or notifications using network fault recovery
- H04L41/0668—Management of faults, events, alarms or notifications using network fault recovery by dynamic selection of recovery network elements, e.g. replacement by the most appropriate element after failure
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a kind of data center's control method, including step:First data center sets up with the second data center and communicated to connect, wherein, first data center surpasses including many respectively with second data center merges all-in-one, all super fusion all-in-ones constitute a clusters, and first data center virtual machine different from each self-operating of the second data center is to run corresponding service;When the virtual machine of first data center or second data center operation business writes the first data, first data are stored in the corresponding second default stored copies of the corresponding first default stored copies of first data center and second data center simultaneously, with the data of synchronous first data center and second data center.The invention also discloses a kind of data center's control device.The present invention is realized while data center systems reliability is ensured, improves the resource utilization of data center systems.
Description
Technical field
The present invention relates to network communication technology field, more particularly to a kind of data center's control method and device.
Background technology
For calamity standby (Disaster Recovery) purpose, data center systems typically can all build two
(or multiple) data center.One is primary data center, the business for undertaking user, and one is standby
Data center, the data for backing up primary data center, prevents loss of data, it is ensured that data center systems
Reliability.But it is standby because generally only primary data center undertakes the business of user in data center systems
Data center only backs up to the data of primary data center, or only in primary data center failure,
The business of adapter primary data center.Namely in most cases, for data center all in idle state,
The wasting of resources is caused, the resource utilization of data center systems is relatively low.
The content of the invention
It is a primary object of the present invention to propose a kind of data center's control method and device, it is intended to solve existing
There is the low technical problem of the resource utilization of data center systems in technology.
To achieve the above object, the present invention provides a kind of data center's control method, data center's control
Method processed comprises the following steps:
First data center sets up with the second data center and communicated to connect, wherein, first data center
Surpass respectively including many with second data center and merge all-in-one, all super fusion all-in-ones compositions one
Individual cluster, first data center virtual machine different from each self-operating of the second data center is to transport
Row corresponding service;
In the virtual machine write-in first of first data center or second data center operation business
During data, first data are stored in the corresponding first default storage of first data center simultaneously
In the corresponding second default stored copies of copy and second data center, with synchronous first number
According to center and the data of second data center.
Preferably, the void that business is run in first data center or second data center
When plan machine writes the first data, first data are stored in first data center simultaneously corresponding
In the corresponding second default stored copies of first default stored copies and second data center, with same
After the step of walking the data of first data center and second data center, in addition to:
When first data center breaks down, according to synchronous data, in second data
The heart starts virtual machine, by the service switching run in first data center to second data center
Operation;Or
When second data center breaks down, according to synchronous data, in first data
The heart starts virtual machine, by the service switching run in second data center to first data center
Operation.
Preferably, it is described when second data center breaks down, according to synchronous data, in institute
State the first data center and start virtual machine, by the service switching run in second data center to described
Before the step of first data center runs, in addition to:
Fault detection information is sent to second data center, wherein, if second data center is just
Often, corresponding response message is fed back when receiving the fault detection information by second data center;
When not receiving the response message of the second data center feedback in preset duration, judge described
Second data center's failure.
Preferably, it is described when second data center breaks down, according to synchronous data, in institute
State the first data center and start virtual machine, by the service switching run in second data center to described
After the step of first data center runs, in addition to:
When detecting the second data center fault recovery, second data center failure is extracted in
The second data stored in first default stored copies described in period;
Second data are sent to second data center, for by second data storage in
In described second default stored copies.
Preferably, set up logical with arbitration device in first data center and second data center
It is described to run the virtual of business in first data center or second data center during letter connection
When machine writes the first data, first data are stored in first data center corresponding the simultaneously
Step in the corresponding second default stored copies of one default stored copies and second data center
Afterwards, in addition to:
First data center and the arbitration device disconnecting or second data center with
During the disconnecting of the arbitration device, first data center with second data center each
Continue to run with corresponding service;
Connection and first data center and institute in first data center and the arbitration device
When the connection for stating the second data center is interrupted, virtual machine is started in second data center, will be described
The service switching run in first data center to second data center runs.
In addition, to achieve the above object, the present invention also proposes a kind of data center's control device, the number
Include according to central controller:
Communication module, sets up for the first data center and the second data center and communicates to connect, wherein, institute
State the first data center and second data center include respectively many it is super merge all-in-one, it is all surpass melt
Unify body machine and constitute a cluster, first data center and each self-operating of the second data center are not
With virtual machine to run corresponding service;
Memory module, for running business in first data center or second data center
When virtual machine writes the first data, first data are stored in the first data center correspondence simultaneously
The corresponding second default stored copies of the first default stored copies and second data center in, with
Synchronous first data center and the data of second data center.
Preferably, data center's control device also includes:
Processing module, for when first data center breaks down, according to synchronous data,
Second data center starts virtual machine, by the service switching run in first data center to institute
State the center operation of the second data center;Or when second data center breaks down, according to synchronization
Data, first data center start virtual machine, the industry that will be run in second data center
Business switches to the first data center operation.
Preferably, the communication module, is additionally operable to send fault detection information to second data center,
Wherein, if second data center is normal, second data center is receiving the fault detect
During information, corresponding response message is fed back;
Data center's control device also includes judge module, for not receiving institute in preset duration
When stating the response message of the second data center feedback, second data center failure is judged.
Preferably, the processing module, is additionally operable to when detecting the second data center fault recovery,
The second number stored during being extracted in second data center failure in the described first default stored copies
According to;
The communication module, is additionally operable to send second data to second data center, for
By second data storage in the described second default stored copies.
Preferably, set up logical with arbitration device in first data center and second data center
During letter connection, the processing module is additionally operable to:
First data center and the arbitration device disconnecting or second data center with
During the disconnecting of the arbitration device, first data center with second data center each
Continue to run with corresponding service;
Connection and first data center and institute in first data center and the arbitration device
When the connection for stating the second data center is interrupted, virtual machine is started in second data center, will be described
The service switching run in first data center to second data center runs.
Data center's control method and device proposed by the present invention, the first data center and have communication link with it
Each self-operating different business of the second data center connect, when in the first data center or the second data center
During the virtual machine write-in data of operation business, the data are stored in the first data center correspondence simultaneously
The corresponding second default stored copies of the first default stored copies and second data center in, together
The data of first data center and second data center are walked, so as to realize data backup, it is to avoid
Loss of data, it is ensured that the reliabilities of data center systems;And each data center is equal in the present invention
Operation business, it is to avoid only primary data center operation business in data center systems, standby data center leaves unused
The problem of, improve the resource utilization of data center systems.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of data center's control method first embodiment of the present invention;
Fig. 2 is the schematic flow sheet of data center's control method second embodiment of the present invention;
Fig. 3 is a kind of running status schematic diagram of data center systems in the embodiment of the present invention;
Fig. 4 is another running status schematic diagram of data center systems in the embodiment of the present invention;
Fig. 5 is the high-level schematic functional block diagram of data center's control device first embodiment of the present invention;
Fig. 6 is the high-level schematic functional block diagram of data center's control device second embodiment of the present invention.
The realization, functional characteristics and advantage of the object of the invention will be done further referring to the drawings in conjunction with the embodiments
Explanation.
Embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, limit is not used to
The fixed present invention.
The present invention provides a kind of data center's control method, and reference picture 1, Fig. 1 is controlled for data center of the present invention
The schematic flow sheet of method first embodiment processed.
In the present embodiment, data center's control method comprises the following steps:
Step S10, the first data center sets up with the second data center and communicated to connect, wherein, described first
Data center surpasses including many respectively with second data center merges all-in-one, all super fusion one
Machine constitutes a cluster, first data center void different from each self-operating of the second data center
Plan machine is to run corresponding service;
In the present embodiment, in order to ensure that business is uninterrupted and data are not lost, set up in a data
System is felt concerned about, the data center systems include the first data center and the second data center.In first data
The heart and the second data center include some super fusion all-in-ones, and all of Liang Ge data centers super are melted
Unification body unit builds up a super fusion cluster, and all application servers of client are all the shapes with virtual machine
Formula is operated in inside super fusion cluster, the physical disk quilt of all super fusion all-in-ones in the super fusion cluster
Inventing memory resource pool, there is provided storage service.Pass through between first data center and the second data center
Express network sets up communication connection.And Liang Ge data centers are separately operable difference all in active state
Virtual machine perform user business.
Step S20, the virtual machine of business is run in first data center or second data center
When writing the first data, first data are stored in first data center corresponding first simultaneously
In the corresponding second default stored copies of default stored copies and second data center, with synchronous institute
State the data of the first data center and second data center.
In the present embodiment, first data center data storage corresponding first is pre-set default to deposit
Store up copy, and the corresponding second default stored copies of second data center data storage.First
During data center and second data center's operation business, when in first data center or institute
Some virtual machine of the second data center is stated when writing the first data, first data are deposited simultaneously
It is stored in the corresponding first default stored copies of first data center and second data center correspondence
The second default stored copies in, therefore, realize synchronous first data center and second data
The data stored in center.
Further, the corresponding first default stored copies of first data center and second data
The quantity of the corresponding second default stored copies in center can be one, or multiple.For example, working as
First data center corresponding first, which presets stored copies, includes stored copies A and stored copies B, institute
When stating the corresponding second default stored copies of the second data center for stored copies C, then data to be written is same
When be stored in first data center corresponding stored copies A and stored copies B, and second number
According in the corresponding stored copies C in center.
Specifically, when in first data center or second data center operation system it is (empty
Plan machine) when writing the first data, the operation system sends write request and communicated to having with the operation system
The client component of connection, when the client component receives write request, the write request is sent simultaneously
To the service end component and the service end component of the second data center of the first data center.First data center
Service end component when receiving the write request, write first data into the first data center corresponding
In one default stored copies, when a write is completed, feedback data writes successful information to the client component.
When the service end component of second data center receives the write request, write first data into the second data
In the corresponding second default stored copies in center, when a write is completed, feedback data write-in successful information is extremely
The client component.When the client component receives the number of the service end component feedback of the first data center
When writing successful information according to the data of the service end component feedback of write-in successful information and the second data center,
Data write-in successful information is sent to operation system, when operation system receives data write-in successful information,
Data write operation is completed.
Preferably, pre-set and each virtual of business is run in the first data center and the second data center
The corresponding stored copies of machine write-in data.For example, pre-setting the number of virtual machine VM1 operation business write-ins
According to being stored in the stored copies 11 of first data center, and second data center storage
In copy 12;The data storage of virtual machine VM2 operation business write-ins is in the storage of first data center
In copy 21, and in the stored copies 22 of second data center;Virtual machine VM3 operations business is write
The data storage entered is in the stored copies 31 of first data center, and second data center
Stored copies 32 in;The data storage of virtual machine VM4 operation business write-ins is in first data center
Stored copies 41 in, and in the stored copies 42 of second data center.As the virtual machine VM1
In the business of operation write-in data, the data are stored in the storage pair of first data center simultaneously
In the stored copies 12 of this 11 neutralization, second data center.
And for example, the data storage of virtual machine VM1 operation business write-ins is pre-set in first data
In the stored copies 11 and stored copies 12 of the heart, and in the stored copies 13 of second data center;
Virtual machine VM2 runs the data storage of business write-in in the stored copies 21 of first data center,
And in the stored copies 22 and stored copies 23 of second data center.When the virtual machine VM1 exists
When operation business writes data, the data are stored in the stored copies of first data center simultaneously
11 and stored copies 12, and the stored copies 13 of second data center in.
The scheme that the present embodiment is provided, the first data center and the second data center for having communication connection with it
Each self-operating different business, when the virtual machine that business is run in the first data center or the second data center
When writing data, the data are stored in the corresponding first default storage of first data center simultaneously secondary
This and second data center corresponding second are preset in stored copies, in synchronous first data
The heart and the data of second data center, so as to realize data backup, it is to avoid loss of data, it is ensured that
The reliabilities of data center systems;And each data center runs business in the present invention, it is to avoid
Only primary data center runs business in data center systems, and the problem of standby data center leaves unused improves number
According to the resource utilization of centring system.
Further, data center's control method second embodiment of the present invention is proposed based on first embodiment.
In the present embodiment, after the step S20, in addition to:
Step a, when first data center breaks down, according to synchronous data, described second
Data center starts virtual machine, by the service switching run in first data center to the described second number
Run according to center;Or
Step b, when second data center breaks down, according to synchronous data, described first
Data center starts virtual machine, by the service switching run in second data center to the described first number
Run according to center.
During first data center and the second data center operation business, it may go out
Existing failure.Do not lost with data to ensure that the business of user is not interrupted, when detecting first data
Center break down when, first data center out of service, and according to first data center with
The synchrodata of second data center, virtual machine is started in second data center, by described the
The service switching run in one data center to second data center runs.When detecting described second
When data center breaks down, second data center out of service, and according in first data
The heart and the synchrodata of second data center, start virtual machine, by institute in first data center
Service switching to first data center run in the second data center is stated to run.So as to ensure that use
The business at family is not interrupted does not lose with data.
Further, in the present embodiment, before the step b, in addition to step:
Step c, sends fault detection information to second data center, wherein, if second data
Center is normal, and second data center is when receiving the fault detection information, and feedback is corresponding to ring
Answer information;
Step d, when not receiving the response message of the second data center feedback in preset duration, sentences
Disconnected second data center failure.
During first data center and the second data center operation business, described first
Data center is in real time or timing sends fault detection information to second data center, to detect described the
Whether two data centers are normal.If second data center is normal, the fault detect letter is being received
During breath, corresponding response message is fed back to first data center.First data center is receiving
During to the response message, then judge that second data center is normal.If the second data center event
Barrier, then can not feedback response information to first data center.Therefore, first data center exists
When not receiving the response message of the second data center feedback in preset duration, then described second is judged
Data center's failure.
Similarly, second data center is in real time or timing sends fault detection information to first data
Center, to detect whether first data center is normal.If first data center is normal, connecing
When receiving the fault detection information, corresponding response message is fed back to second data center.It is described
Second data center judges that first data center is normal when receiving the response message, then.If
First data center failure, then can not feedback response information to second data center.Therefore,
Second data center does not receive the response message of the first data center feedback in preset duration
When, then judge first data center failure.
Further, as shown in Fig. 2 in the present embodiment, after the step b, in addition to step:
Step S30, when detecting the second data center fault recovery, is extracted in second data
The second data stored during central fault in the described first default stored copies;
Step S40, second data are sent to second data center, for described second is counted
Preset according to being stored in described second in stored copies.
In the present embodiment, when detecting the second data center failure, described second is counted
Run according to service switching is run in center into first data center.Now, in first data
The heart starts asynchronous with the data that second data center stores.When first data center detects
During the second data center fault recovery, it is extracted in described first during second data center failure
The second data stored in the corresponding first default stored copies of data center, then by second data
Send to second data center.Second data center, will when receiving second data
Second data storage is in the corresponding second default stored copies of second data center so that institute
It is synchronous with the data holding that second data center stores to state the first data center.
Similarly, when detecting the first data center failure, by first data center
Operation service switching is run into second data center.Now, first data center with it is described
The data of second data center storage start asynchronous.When second data center detects described first
During data center's fault recovery, second data center during being extracted in first data center failure
The 3rd data stored in corresponding second default stored copies, then send the 3rd data to institute
State the first data center.First data center is when receiving three data, by the described 3rd
Data storage is in the corresponding first default stored copies of first data center so that first number
It is synchronous with the data holding that second data center stores according to center.
The scheme that the present embodiment is provided, when detecting the failure of the second data center, by the second data
The service switching run in center to the first data center runs, or goes out when detecting the first data center
During existing failure, the service switching run in the first data center to the second data center is run so that industry
Business is not interrupted, so as to while the resource utilization of data center systems is improved, further ensure number
According to the reliability of centring system.
Further, data center's control method of the present invention is proposed based on first embodiment or second embodiment
3rd embodiment.In the present embodiment, first data center and second data center with it is secondary
Device is cut out to set up after communication connection, the step S20, in addition to step:
Step e, in first data center and the disconnecting or second data of the arbitration device
When center and the disconnecting of the arbitration device, in first data center and second data
The heart each continues to run with corresponding service;
Step f, in the connection of first data center and the arbitration device and first data
When the heart and the connection of second data center are interrupted, virtual machine is started in second data center,
The service switching run in first data center to second data center is run.
In the present embodiment, data center systems also include arbitration device.The arbitration device and described the
One data center and second data center are set up by network to be communicated to connect.The network is not required to using height
Fast network, only need to enable the arbitration device enter with first data center and second data center
Row communication.During data center systems are run, it is possible that first data center
Failure, second data center failure, the arbitration device failure and network connection interruption etc. are various
Situation, for different situations, data center systems use different processing modes.Specifically include following
Several situations:
1), first data center and second data center are normal
Because first data center and second data center are normal, Liang Ge data centers Jun Chu
In running status, in this case, no matter whether the arbitration device normally runs, even if the arbitration
During plant failure, first data center continues normal operation with second data center.
2), first data center failure, second data center and the arbitration device are normal
When the first data center failure, first data center is out of service.Due to described
Two data centers are normal with the arbitration device, and the virtual machine-cut of business is run in first data center
Shift in second data center and run.
3), second data center failure, first data center and the arbitration device are normal
When the second data center failure, second data center is out of service.Due to described
One data center is normal with the arbitration device, and the virtual machine-cut of business is run in second data center
Shift in first data center and run.
4), the broken communication connection of first data center and second data center
When first data center and second data center are normal, but first data center
During with the broken communication connection of second data center, the void of business is run in first data center
The virtual machine that plan machine continues to run with operation business in business, second data center continues to run with business.
But the data of the virtual machine write-in of business are run in first data center can not send to the described second number
According to center, it is impossible in the stored copies for being stored in second data center;In second data center
The data of the virtual machine write-in of operation business can not be sent to first data center, it is impossible to be stored in institute
In the stored copies for stating the first data center.When first data center and second data center
When communication connection recovers, the data of synchronous first data center and second data center.
5), first data center and the broken communication connection of the arbitration device, or described second
Data center and the broken communication connection of the arbitration device
When the broken communication connection of first data center and the arbitration device, or described second
During the broken communication connection of data center and the arbitration device, the not operation shape of change data centring system
The virtual machine that business is run in state, first data center is continued to run with business, second data
The virtual machine of operation business continues to run with business, also, first data center and described second in the heart
The data of data center are also synchronous.
6), the broken communication connection of first data center and the arbitration device, and with described second
The broken communication connection of data center
As shown in figure 3, when the broken communication connection of first data center and the arbitration device, and institute
When the communication connection for stating the first data center and second data center is also interrupted, in order to avoid described the
One data center has respectively run an example of virtual machine, two example difference with second data center
A copy is all write, causes to produce fissure, first data center is out of service, and described first is counted
Switch in second data center and run according to the virtual machine that business is run in center.
7), the broken communication connection of second data center and the arbitration device, and with described first
The broken communication connection of data center
When the broken communication connection of second data center and the arbitration device, and second data
When center and the communication connection of first data center are also interrupted, second data center is out of service,
The virtual machine that business is run in second data center is switched in first data center and run.
8), first data center is normal, second data center and the arbitration device failure
As shown in figure 4, when second data center and arbitration device all failures, although described
One data center is normal, but first data center is out of service, and second data center is due to event
Barrier is also out of service.In this case, customer service is interrupted.
9), second data center is normal, first data center and the arbitration device failure
When first data center and arbitration device all failures, although second data center
Normally, but second data center is out of service, first data center also stops fortune due to failure
OK.In this case, customer service is interrupted.
10), first data center and the equal failure of the second data center
When the equal failure of first data center and second data center, no matter the arbitration device
Whether normal, first data center and second data center are out of service, in customer service
It is disconnected.
The scheme that the present embodiment is provided, the first data center and the second data center are connected with arbitration device,
Connection and first data center and described the when first data center and the arbitration device
When the connection of two data centers is interrupted, start virtual machine in second data center, by described first
The service switching run in data center to second data center runs, so as to avoid two data
An example of virtual machine has respectively been run at center, and two examples write a copy respectively, causes to produce fissure
The problem of, therefore further increase the reliability of data center systems.
The present invention further provides a kind of data center's control device, as shown in figure 5, Fig. 5 is the present invention
The high-level schematic functional block diagram of data center's control device first embodiment.
In the present embodiment, data center's control device includes:
Communication module 10, sets up for the first data center and the second data center and communicates to connect, wherein,
First data center surpasses including many respectively with second data center merges all-in-one, Suo Youchao
Merge all-in-one and constitute a cluster, first data center and each self-operating of the second data center
Different virtual machines is to run corresponding service;
In the present embodiment, in order to ensure that business is uninterrupted and data are not lost, set up in a data
System is felt concerned about, the data center systems include the first data center and the second data center.In first data
The heart and the second data center include some super fusion all-in-ones, and all of Liang Ge data centers super are melted
Unification body unit builds up a super fusion cluster, and all application servers of client are all the shapes with virtual machine
Formula is operated in inside super fusion cluster, the physical disk quilt of all super fusion all-in-ones in the super fusion cluster
Inventing memory resource pool, there is provided storage service.Pass through between first data center and the second data center
Communication module 10 sets up communication connection using express network.And Liang Ge data centers are all in active state,
It is separately operable the business that different virtual machines perform user.
Memory module 20, in first data center or second data center operation business
Virtual machine write the first data when, by first data simultaneously be stored in first data center pair
In the corresponding second default stored copies of the first default stored copies answered and second data center,
With the data of synchronous first data center and second data center.
In the present embodiment, first data center data storage corresponding first is pre-set default to deposit
Store up copy, and the corresponding second default stored copies of second data center data storage.First
During data center and second data center's operation business, when in first data center or institute
Some virtual machine of the second data center is stated when writing the first data, memory module 20 is by described first
Data are stored in the corresponding first default stored copies of first data center and second number simultaneously
According in the corresponding second default stored copies in center, therefore, synchronous first data center and institute are realized
State the data stored in the second data center.
Further, the corresponding first default stored copies of first data center and second data
The quantity of the corresponding second default stored copies in center can be one, or multiple.For example, working as
First data center corresponding first, which presets stored copies, includes stored copies A and stored copies B, institute
When stating the corresponding second default stored copies of the second data center for stored copies C, then memory module 20 will be write
The data that enter while be stored in first data center corresponding stored copies A and stored copies B, and
In the corresponding stored copies C of second data center.
Preferably, pre-set and each virtual of business is run in the first data center and the second data center
The corresponding stored copies of machine write-in data.For example, pre-setting the number of virtual machine VM1 operation business write-ins
According to being stored in the stored copies 11 of first data center, and second data center storage
In copy 12;The data storage of virtual machine VM2 operation business write-ins is in the storage of first data center
In copy 21, and in the stored copies 22 of second data center;Virtual machine VM3 operations business is write
The data storage entered is in the stored copies 31 of first data center, and second data center
Stored copies 32 in;The data storage of virtual machine VM4 operation business write-ins is in first data center
Stored copies 41 in, and in the stored copies 42 of second data center.As the virtual machine VM1
In the business of operation write-in data, the data are stored in first data by memory module 20 simultaneously
The stored copies 11 of the heart are neutralized in the stored copies 12 of second data center.
And for example, the data storage of virtual machine VM1 operation business write-ins is pre-set in first data
In the stored copies 11 and stored copies 12 of the heart, and in the stored copies 13 of second data center;
Virtual machine VM2 runs the data storage of business write-in in the stored copies 21 of first data center,
And in the stored copies 22 and stored copies 23 of second data center.When the virtual machine VM1 exists
When operation business writes data, the data are stored in first data center by memory module 20 simultaneously
Stored copies 11 and stored copies 12, and in the stored copies 13 of second data center.
The scheme that the present embodiment is provided, the first data center and the second data center for having communication connection with it
Each self-operating different business, when the virtual machine that business is run in the first data center or the second data center
When writing data, the data are stored in first data center corresponding first by memory module 20 simultaneously
It is synchronous described in the corresponding second default stored copies of default stored copies and second data center
First data center and the data of second data center, so as to realize data backup, it is to avoid data
Lose, it is ensured that the reliability of data center systems;And each data center runs industry in the present invention
Business, it is to avoid only primary data center operation business in data center systems, the problem of standby data center is idle,
Improve the resource utilization of data center systems.
Further, as shown in fig. 6, proposing data center's control device of the present invention based on first embodiment
Second embodiment.In the present embodiment, data center's control device also includes:
Processing module 30, for when first data center breaks down, according to synchronous data,
Start virtual machine in second data center, by the service switching run in first data center extremely
Second data center center operation;Or when second data center breaks down, according to same
The data of step, start virtual machine, by what is run in second data center in first data center
Service switching to first data center runs.
During first data center and the second data center operation business, it may go out
Existing failure.Do not lost with data to ensure that the business of user is not interrupted, when detecting first data
When center is broken down, first data center out of service of processing module 30, and according to described first
The synchrodata of data center and second data center, virtual machine is started in second data center,
The service switching run in first data center to second data center is run.When detecting
When second data center breaks down, second data center out of service of processing module 30, and
According to the synchrodata of first data center and second data center, in first data
The heart starts virtual machine, by the service switching run in second data center to first data center
Operation.Do not lost with data so as to ensure that the business of user is not interrupted.
Further, in the present embodiment, the communication module 10, is additionally operable to send fault detection information
To second data center, wherein, if second data center is normal, second data center
When receiving the fault detection information, corresponding response message is fed back;
Data center's control device also includes judge module, for not receiving institute in preset duration
When stating the response message of the second data center feedback, second data center failure is judged.
During first data center and the second data center operation business, communication module
10 also send fault detection information to second data center in real time or regularly, to detect second number
It is whether normal according to center.If second data center is normal, when receiving the fault detection information,
Corresponding response message is fed back to first data center.Received in first data center described
During response message, then judge module judges that second data center is normal.If second data center
Failure, then can not feedback response information to first data center.Therefore, when in first data
When the heart does not receive the response message of second data center feedback in preset duration, then judge module
Judge second data center failure.
Similarly, communication module 10 also sends fault detection information to first data center in real time or regularly,
To detect whether first data center is normal.If first data center is normal, receiving
When stating fault detection information, corresponding response message is fed back to second data center.Described second
When data center receives the response message, then judge module judges that first data center is normal.
If first data center failure, can not feedback response information to second data center.Therefore,
Second data center does not receive the response message of the first data center feedback in preset duration
When, then judge module judges first data center failure.
Further, in the present embodiment, the processing module 30, is additionally operable to detecting described second
During data center's fault recovery, the described first default storage during being extracted in second data center failure
The second data stored in copy;
The communication module 10, is additionally operable to send second data to second data center, with
For second data storage is preset in stored copies in described second.
In the present embodiment, when detecting the second data center failure, processing module 30 will
Service switching is run in second data center to run into first data center.Now, it is described
First data center starts asynchronous with the data that second data center stores.When detecting described
During the fault recovery of two data centers, processing module 30 is described during being extracted in second data center failure
First data center corresponding first presets the second data stored in stored copies, then communication module 10
Second data are sent to second data center.Received in second data center described
During the second data, memory module 20 is by second data storage in second data center corresponding
In two default stored copies so that the data that first data center stores with second data center
Keep synchronous.
Similarly, when detecting first data center and breaking down, processing module 30 is by described first
Service switching is run in data center to run into second data center.Now, first data
Center starts asynchronous with the data that second data center stores.When detecting in first data
During heart fault recovery, second data during being extracted in first data center failure of processing module 30
The 3rd data stored in the corresponding second default stored copies in center, then communication module 10 is by described the
Three data are sent to first data center.The 3rd data are received in first data center
When, the 3rd data storage is preset and deposited by memory module 20 in first data center corresponding first
Store up in copy so that first data center is synchronous with the data holding that second data center stores.
The scheme that the present embodiment is provided, when detecting the failure of the second data center, processing module 30
The service switching run in second data center to the first data center is run, or first ought be detected
When data center breaks down, processing module 30 is by the service switching run in the first data center to second
Data center is run so that business is not interrupted, so as to improve the resource utilization of data center systems
Meanwhile, further ensure the reliability of data center systems.
Further, data center's control device of the present invention is proposed based on first embodiment or second embodiment
3rd embodiment.In the present embodiment, first data center and second data center with it is secondary
Cut out device and set up communication connection, the processing module 30 is additionally operable to:
First data center and the arbitration device disconnecting or second data center with
During the disconnecting of the arbitration device, first data center with second data center each
Continue to run with corresponding service;
Connection and first data center and institute in first data center and the arbitration device
When the connection for stating the second data center is interrupted, virtual machine is started in second data center, will be described
The service switching run in first data center to second data center runs.
In the present embodiment, data center systems also include arbitration device.The arbitration device and described the
One data center and second data center are set up by network to be communicated to connect.The network is not required to using height
Fast network, only need to enable the arbitration device enter with first data center and second data center
Row communication.During data center systems are run, it is possible that first data center
Failure, second data center failure, the arbitration device failure and network connection interruption etc. are various
Situation, for different situations, processing module 30 uses different processing modes.Specifically include following several
The situation of kind:
1), first data center and second data center are normal
Because first data center and second data center are normal, Liang Ge data centers Jun Chu
In running status, in this case, no matter whether the arbitration device normally runs, even if the arbitration
During plant failure, processing module 30 maintains first data center to continue normal operation;Second number
Continue normal operation according to center.
2), first data center failure, second data center and the arbitration device are normal
When the first data center failure, processing module 30 is out of service by first data center.
Because second data center and the arbitration device are normal, business is run in first data center
Virtual machine switch in second data center run.
3), second data center failure, first data center and the arbitration device are normal
When the second data center failure, second data center is out of service.Due to described
One data center is normal with the arbitration device, and processing module 30 will run industry in second data center
The virtual machine of business is switched in first data center and run.
4), the broken communication connection of first data center and second data center
When first data center and second data center are normal, but first data center
During with the broken communication connection of second data center, the void of business is run in first data center
The virtual machine that plan machine continues to run with operation business in business, second data center continues to run with business.
But the data of the virtual machine write-in of business are run in first data center can not send to the described second number
According to center, it is impossible in the stored copies for being stored in second data center;In second data center
The data of the virtual machine write-in of operation business can not be sent to first data center, it is impossible to be stored in institute
In the stored copies for stating the first data center.When first data center and second data center
When communication connection recovers, the data of synchronous first data center and second data center.
5), first data center and the broken communication connection of the arbitration device, or described second
Data center and the broken communication connection of the arbitration device
When the broken communication connection of first data center and the arbitration device, or described second
During the broken communication connection of data center and the arbitration device, the not operation shape of change data centring system
The virtual machine that business is run in state, first data center is continued to run with business, second data
The virtual machine of operation business continues to run with business, also, first data center and described second in the heart
The data of data center are also synchronous.
6), the broken communication connection of first data center and the arbitration device, and with described second
The broken communication connection of data center
As shown in figure 3, when the broken communication connection of first data center and the arbitration device, and institute
When the communication connection for stating the first data center and second data center is also interrupted, in order to avoid described the
One data center has respectively run an example of virtual machine, two example difference with second data center
A copy is all write, causes to produce fissure, processing module 30 is out of service by first data center,
The virtual machine that business is run in first data center switches to operation in second data center.
7), the broken communication connection of second data center and the arbitration device, and with described first
The broken communication connection of data center
When the broken communication connection of second data center and the arbitration device, and second data
When center and the communication connection of first data center are also interrupted, second data center is out of service,
Processing module 30 switches to the virtual machine that business is run in second data center in first data
Run in the heart.
8), first data center is normal, second data center and the arbitration device failure
As shown in figure 4, when second data center and arbitration device all failures, although described
One data center is normal, and processing module 30 is out of service by first data center, second data
Center is also out of service due to failure.In this case, customer service is interrupted.
9), second data center is normal, first data center and the arbitration device failure
When first data center and arbitration device all failures, although second data center
Normally, second data center is out of service, and first data center is also out of service due to failure.
In this case, customer service is interrupted.
10), first data center and the equal failure of the second data center
When the equal failure of first data center and second data center, no matter the arbitration device
Whether normal, first data center and second data center are out of service, in customer service
It is disconnected.
The scheme that the present embodiment is provided, the first data center and the second data center are connected with arbitration device,
Connection and first data center and described the when first data center and the arbitration device
When the connection of two data centers is interrupted, processing module 30 starts virtual machine in second data center,
The service switching run in first data center to second data center is run, so as to avoid
Liao Liangge data centers have respectively run an example of virtual machine, and two examples are write a copy, led respectively
The problem of causing to produce fissure, therefore further increase the reliability of data center systems.
The preferred embodiments of the present invention are these are only, are not intended to limit the scope of the invention, it is every
The equivalent structure or equivalent flow conversion made using description of the invention and accompanying drawing content, or directly or
Connect and be used in other related technical fields, be included within the scope of the present invention.
Claims (10)
1. a kind of data center's control method, it is characterised in that data center's control method include with
Lower step:
First data center sets up with the second data center and communicated to connect, wherein, first data center
Surpass respectively including many with second data center and merge all-in-one, all super fusion all-in-ones compositions one
Individual cluster, first data center virtual machine different from each self-operating of the second data center is to transport
Row corresponding service;
In the virtual machine write-in first of first data center or second data center operation business
During data, first data are stored in the corresponding first default storage of first data center simultaneously
In the corresponding second default stored copies of copy and second data center, with synchronous first number
According to center and the data of second data center.
2. data center's control method as claimed in claim 1, it is characterised in that described described
When the virtual machine of one data center or second data center operation business writes the first data, by institute
The first data are stated while being stored in the corresponding first default stored copies of first data center and described
In the corresponding second default stored copies of second data center, with synchronous first data center and described
After the step of data of second data center, in addition to:
When first data center breaks down, according to synchronous data, in second data
The heart starts virtual machine, by the service switching run in first data center to second data center
Operation;Or
When second data center breaks down, according to synchronous data, in first data
The heart starts virtual machine, by the service switching run in second data center to first data center
Operation.
3. data center's control method as claimed in claim 2, it is characterised in that described described
When two data centers break down, according to synchronous data, virtual machine is started in first data center,
The step of service switching run in second data center to first data center is run it
Before, in addition to:
Fault detection information is sent to second data center, wherein, if second data center is just
Often, corresponding response message is fed back when receiving the fault detection information by second data center;
When not receiving the response message of the second data center feedback in preset duration, judge described
Second data center's failure.
4. data center's control method as claimed in claim 2, it is characterised in that described described
When two data centers break down, according to synchronous data, virtual machine is started in first data center,
The step of service switching run in second data center to first data center is run it
Afterwards, in addition to:
When detecting the second data center fault recovery, second data center failure is extracted in
The second data stored in first default stored copies described in period;
Second data are sent to second data center, for by second data storage in
In described second default stored copies.
5. data center's control method as described in claim any one of 1-4, it is characterised in that in institute
State the first data center and second data center with arbitration device set up communicate to connect when, it is described
The virtual machine of first data center or second data center operation business writes the first data
When, first data are stored in the corresponding first default stored copies of first data center simultaneously
And after the step in the corresponding second default stored copies of second data center, in addition to:
First data center and the arbitration device disconnecting or second data center with
During the disconnecting of the arbitration device, first data center with second data center each
Continue to run with corresponding service;
Connection and first data center and institute in first data center and the arbitration device
When the connection for stating the second data center is interrupted, virtual machine is started in second data center, will be described
The service switching run in first data center to second data center runs.
6. a kind of data center's control device, it is characterised in that data center's control device includes:
Communication module, sets up for the first data center and the second data center and communicates to connect, wherein, institute
State the first data center and second data center include respectively many it is super merge all-in-one, it is all surpass melt
Unify body machine and constitute a cluster, first data center and each self-operating of the second data center are not
With virtual machine to run corresponding service;
Memory module, for running business in first data center or second data center
When virtual machine writes the first data, first data are stored in the first data center correspondence simultaneously
The corresponding second default stored copies of the first default stored copies and second data center in, with
Synchronous first data center and the data of second data center.
7. data center's control device as claimed in claim 6, it is characterised in that the data center
Control device also includes:
Processing module, for when first data center breaks down, according to synchronous data,
Second data center starts virtual machine, by the service switching run in first data center to institute
State the operation of the second data center;Or when second data center breaks down, according to synchronous number
According to, first data center start virtual machine, the business run in second data center is cut
Shift to the first data center operation.
8. data center's control device as claimed in claim 7, it is characterised in that the communication module,
It is additionally operable to send fault detection information to second data center, wherein, if second data center
Normally, second data center is when receiving the fault detection information, the corresponding response letter of feedback
Breath;
Data center's control device also includes judge module, for not receiving institute in preset duration
When stating the response message of the second data center feedback, second data center failure is judged.
9. data center's control device as claimed in claim 7, it is characterised in that the processing module,
It is additionally operable to when detecting the second data center fault recovery, is extracted in the second data center event
The second data stored during barrier in the described first default stored copies;
The communication module, is additionally operable to send second data to second data center, for
By second data storage in the described second default stored copies.
10. data center's control device as described in claim any one of 7-9, it is characterised in that in institute
When stating the first data center and second data center with arbitration device foundation communication connection, the place
Reason module is additionally operable to:
First data center and the arbitration device disconnecting or second data center with
During the disconnecting of the arbitration device, first data center with second data center each
Continue to run with corresponding service;
Connection and first data center and institute in first data center and the arbitration device
When the connection for stating the second data center is interrupted, virtual machine is started in second data center, will be described
The service switching run in first data center to second data center runs.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610117910.7A CN107147511A (en) | 2016-03-01 | 2016-03-01 | Data center's control method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610117910.7A CN107147511A (en) | 2016-03-01 | 2016-03-01 | Data center's control method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107147511A true CN107147511A (en) | 2017-09-08 |
Family
ID=59783389
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610117910.7A Pending CN107147511A (en) | 2016-03-01 | 2016-03-01 | Data center's control method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107147511A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109947591A (en) * | 2017-12-20 | 2019-06-28 | 腾讯科技(深圳)有限公司 | Database strange land disaster recovery and backup systems and its dispositions method, deployment device |
CN110297730A (en) * | 2019-07-04 | 2019-10-01 | 北京谷数科技有限公司 | A kind of storage dual-active solution and system for removing arbitration node |
Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101188787A (en) * | 2006-11-17 | 2008-05-28 | 华为技术有限公司 | Disaster tolerance method and system for home register |
CN101394260A (en) * | 2007-09-17 | 2009-03-25 | 华为技术有限公司 | Method and apparatus for realizing main standby inverting and load sharing |
CN101547155A (en) * | 2009-04-30 | 2009-09-30 | 杭州华三通信技术有限公司 | Master and standby tunnel load sharing method and flow forwarding equipment |
CN101789854A (en) * | 2009-01-24 | 2010-07-28 | 华为技术有限公司 | Main/standby load sharing method, device and system |
CN102388570A (en) * | 2011-09-30 | 2012-03-21 | 华为技术有限公司 | Single board running method and system under active-standby mode |
CN103647849A (en) * | 2013-12-24 | 2014-03-19 | 华为技术有限公司 | Method and device for migrating businesses and disaster recovery system |
CN103812929A (en) * | 2014-01-11 | 2014-05-21 | 浪潮电子信息产业股份有限公司 | Active-active method for cloud data center management platforms |
CN104243527A (en) * | 2013-06-20 | 2014-12-24 | 华为技术有限公司 | Data synchronization method and device and distributed system |
CN104331254A (en) * | 2014-11-05 | 2015-02-04 | 浪潮电子信息产业股份有限公司 | Dual-active storage system design method based on dual-active logical volumes |
CN104469699A (en) * | 2014-11-27 | 2015-03-25 | 华为技术有限公司 | Cluster quorum method and multi-cluster cooperation system |
CN204836244U (en) * | 2015-08-25 | 2015-12-02 | 杭州九州方园科技有限公司 | Use two data center of living systems |
CN204859222U (en) * | 2015-06-02 | 2015-12-09 | 郑州银行股份有限公司 | With two high available systems that live of city data center |
CN105224385A (en) * | 2015-09-03 | 2016-01-06 | 成都中机盈科科技有限公司 | A kind of virtualization system based on cloud computing and method |
CN204993445U (en) * | 2015-08-25 | 2016-01-20 | 杭州九州方园科技有限公司 | Two data center systems that live of storage |
-
2016
- 2016-03-01 CN CN201610117910.7A patent/CN107147511A/en active Pending
Patent Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101188787A (en) * | 2006-11-17 | 2008-05-28 | 华为技术有限公司 | Disaster tolerance method and system for home register |
CN101394260A (en) * | 2007-09-17 | 2009-03-25 | 华为技术有限公司 | Method and apparatus for realizing main standby inverting and load sharing |
CN101789854A (en) * | 2009-01-24 | 2010-07-28 | 华为技术有限公司 | Main/standby load sharing method, device and system |
CN101547155A (en) * | 2009-04-30 | 2009-09-30 | 杭州华三通信技术有限公司 | Master and standby tunnel load sharing method and flow forwarding equipment |
CN102388570A (en) * | 2011-09-30 | 2012-03-21 | 华为技术有限公司 | Single board running method and system under active-standby mode |
CN104243527A (en) * | 2013-06-20 | 2014-12-24 | 华为技术有限公司 | Data synchronization method and device and distributed system |
CN103647849A (en) * | 2013-12-24 | 2014-03-19 | 华为技术有限公司 | Method and device for migrating businesses and disaster recovery system |
CN103812929A (en) * | 2014-01-11 | 2014-05-21 | 浪潮电子信息产业股份有限公司 | Active-active method for cloud data center management platforms |
CN104331254A (en) * | 2014-11-05 | 2015-02-04 | 浪潮电子信息产业股份有限公司 | Dual-active storage system design method based on dual-active logical volumes |
CN104469699A (en) * | 2014-11-27 | 2015-03-25 | 华为技术有限公司 | Cluster quorum method and multi-cluster cooperation system |
CN204859222U (en) * | 2015-06-02 | 2015-12-09 | 郑州银行股份有限公司 | With two high available systems that live of city data center |
CN204836244U (en) * | 2015-08-25 | 2015-12-02 | 杭州九州方园科技有限公司 | Use two data center of living systems |
CN204993445U (en) * | 2015-08-25 | 2016-01-20 | 杭州九州方园科技有限公司 | Two data center systems that live of storage |
CN105224385A (en) * | 2015-09-03 | 2016-01-06 | 成都中机盈科科技有限公司 | A kind of virtualization system based on cloud computing and method |
Non-Patent Citations (1)
Title |
---|
郭涛: "超融合初探", 《计算机学报》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109947591A (en) * | 2017-12-20 | 2019-06-28 | 腾讯科技(深圳)有限公司 | Database strange land disaster recovery and backup systems and its dispositions method, deployment device |
CN110297730A (en) * | 2019-07-04 | 2019-10-01 | 北京谷数科技有限公司 | A kind of storage dual-active solution and system for removing arbitration node |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101582787B (en) | Double-computer backup system and backup method | |
JP4611922B2 (en) | Control program, control method, and control apparatus | |
CN101136900B (en) | Fast transparent fault shift device and implementing method facing to service | |
US11307943B2 (en) | Disaster recovery deployment method, apparatus, and system | |
CN102355369B (en) | Virtual clustered system as well as processing method and processing device thereof | |
CN202798798U (en) | High availability system based on cloud computing technology | |
CN106330475B (en) | Method and device for managing main and standby nodes in communication system and high-availability cluster | |
CN107733726A (en) | A kind of processing method and processing device of service request | |
CN109474465A (en) | A kind of method and system of the high availability that can dynamically circulate based on server cluster | |
CN104469699B (en) | Cluster quorum method and more cluster coupled systems | |
CN102394914A (en) | Cluster brain-split processing method and device | |
US20080288812A1 (en) | Cluster system and an error recovery method thereof | |
JP2009266015A (en) | Storage system, and method of changing monitoring condition thereof | |
CN104427002A (en) | Cluster system and method for providing service availability in cluster system | |
CN106330523A (en) | Cluster server disaster recovery system and method, and server node | |
US20080270823A1 (en) | Fast node failure detection via disk based last gasp mechanism | |
CN104317803A (en) | Data access structure and method of database cluster | |
CN106533736A (en) | Network device reboot method and apparatus | |
CN100520724C (en) | Failure switch recovery realization network system and method | |
CN103560907A (en) | Management method and device for double machine load balancing devices | |
CN104753987B (en) | A kind of distributed conversation management method and system | |
CN107368485A (en) | The management method and Database Systems of a kind of database | |
CN107357800A (en) | A kind of database High Availabitity zero loses solution method | |
CN107147511A (en) | Data center's control method and device | |
CN102457400B (en) | Method for preventing split brain phenomenon from occurring on distributed replicated block device (DRBD) resource |
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 |
Application publication date: 20170908 |
|
RJ01 | Rejection of invention patent application after publication |