CN109151028A - A kind of distributed memory system disaster recovery method and device - Google Patents
A kind of distributed memory system disaster recovery method and device Download PDFInfo
- Publication number
- CN109151028A CN109151028A CN201810968427.9A CN201810968427A CN109151028A CN 109151028 A CN109151028 A CN 109151028A CN 201810968427 A CN201810968427 A CN 201810968427A CN 109151028 A CN109151028 A CN 109151028A
- Authority
- CN
- China
- Prior art keywords
- memory node
- client
- address
- virtual
- memory
- 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
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
-
- 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
- H04L41/0663—Performing the actions predefined by failover planning, e.g. switching to standby network elements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L61/00—Network arrangements, protocols or services for addressing or naming
- H04L61/50—Address allocation
- H04L61/5007—Internet protocol [IP] addresses
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L61/00—Network arrangements, protocols or services for addressing or naming
- H04L61/50—Address allocation
- H04L61/5084—Providing for device mobility
Abstract
The embodiment of the present application discloses a kind of method and device of distributed memory system disaster tolerance, when the first memory node of client institute carry breaks down, distributed memory system according to client send about the first memory node failure when fault message, judge the first memory node failure, then obtain corresponding first virtual ip address of the first memory node, and a memory node i.e. the second memory node is selected from remaining multiple memory node, first virtual ip address is distributed to the second memory node in system, so that client is written and read the second memory node according to the first virtual ip address, the case where avoiding the first memory node failure and leading to client traffic terminal, improve the reliability of business in client.
Description
Technical field
This application involves field of computer technology, and in particular to a kind of distributed memory system disaster recovery method and device.
Background technique
With cloud computing and the continuous development of big data technology, information data is with exponential increase, this is to storage equipment
Performance proposes further requirement.
The existing storage system data acquired using the storage server storage concentrated, when facing mass data,
Centralised storage system becomes the bottleneck of data storage, is not able to satisfy the demand of Mass storage application.To meet a large amount of numbers
According to storage, distributed memory system is suggested, uses expansible system structure, shared and being deposited using more storage servers
Load is stored up, to meet Mass storage application.
In practical applications, distributed memory system is the fixed storage server of each client configuration, the client
Data read-write operation is carried out using fixed storage server.However, client can not be into when storage server breaks down
The read-write operation of row data causes service disconnection, traffic affecting reliability.
Summary of the invention
In view of this, the embodiment of the present application provides a kind of method and device of distributed memory system disaster tolerance, it is existing to solve
There is the problem of terminal leads to service disconnection because of storage server failure in technology.
To solve the above problems, technical solution provided by the embodiments of the present application is as follows:
In a first aspect, providing a kind of method of distributed memory system disaster tolerance, this method can wrap the embodiment of the present application
It includes:
Receive the fault message that client is sent；The fault message breaks down for identifying the first memory node；Institute
Stating the first memory node is the memory node that the distributed memory system is the client configuration；First memory node
For a memory node in the distributed memory system；
Obtain corresponding first virtual ip address of first memory node；It is virtual that the client is stored with described first
IP address；
First virtual ip address is distributed into the second memory node, so that the client is empty according to described first
Quasi- IP address is written and read second memory node；Second memory node is in the distributed memory system
Another memory node；The distributed memory system includes at least two memory nodes.
In one possible implementation, in the distributed memory system each memory node be corresponding with to
A few virtual ip address, it is equal according to preconfigured load when needing to be written and read there are multiple clients
Weighing apparatus mechanism is that each client distributes a virtual ip address.
In one possible implementation, described when needing to be written and read there are multiple clients, root
It is that each client distributes a virtual ip address according to preconfigured load-balancing mechanism, specifically includes:
Each memory node connection institute is determined according to the quantity of the quantity of the client and the memory node
State the quantity of client；
For any memory node, the corresponding virtual ip address of the memory node is sent to the memory node
The client to be connected.
In one possible implementation, after first memory node restores normal, for the first storage section
Point redistributes virtual ip address.
In one possible implementation, before the fault message for receiving the client transmission, the method
Further include:
Receive the domain name carry request that the client is sent；The request of domain name carry includes the distributed storage system
The domain name of system；Domain name is that the distributed memory system is the client configuration in advance；
First virtual ip address is obtained according to domain name parsing；Described in first virtual ip address uniquely corresponds to
First memory node；
First virtual ip address is sent to the client, so that the client is according to first virtual IP address
Address is written and read first memory node.
The embodiment of the present application second aspect provides a kind of distributed memory system disaster tolerance device, the apparatus may include:
First receiving unit, for receiving the fault message of client transmission；The fault message is deposited for identifying first
Node is stored up to break down；First memory node is the storage section that the distributed memory system is the client configuration
Point；First memory node is a memory node in the distributed memory system；
Acquiring unit, for obtaining corresponding first virtual ip address of first memory node；The client storage
There is first virtual ip address；
First allocation unit, for first virtual ip address to be distributed to the second memory node, so that the visitor
Family end is written and read second memory node according to first virtual ip address；The client is stored with described
First virtual ip address；The distributed memory system includes at least two memory nodes.
In one possible implementation, described device further include: the second allocation unit, for when there are multiple described
When client needs to be written and read, an institute is distributed for each client according to preconfigured load-balancing mechanism
State virtual ip address；Each memory node is corresponding at least one virtual ip address in the distributed memory system.
In one possible implementation, second allocation unit includes:
Determine subelement, it is each described for being determined according to the quantity of the client and the quantity of the memory node
Memory node connects the quantity of the client；
Transmission sub-unit, for for any memory node, the corresponding virtual ip address of the memory node to be sent out
Give the memory node client to be connected.
In one possible implementation, described device further include: third allocation unit, for when first storage
After node restores normal, virtual ip address is redistributed for first memory node.
In one possible implementation, described device further include:
Second receiving unit, for receiving the domain name carry that the client is sent before the execution of the first receiving unit
Request；The request of domain name carry includes the domain name of the distributed memory system；Domain name is the distributed storage system
System configures in advance for the client；
Resolution unit, for obtaining first virtual ip address according to domain name parsing；First virtual IP address
Location uniquely corresponds to first memory node；
First virtual ip address is sent to the client for this, so that the client root by transmission unit
First memory node is written and read according to first virtual ip address.
It can be seen that the embodiment of the present application has the following beneficial effects:
The embodiment of the present application when client institute carry the first memory node break down when, distributed memory system according to
Client send about the first memory node failure when fault message, judge the first memory node failure, then obtain the
Corresponding first virtual ip address of one memory node, and select a memory node i.e. second to deposit from remaining multiple memory node
Node is stored up, the first virtual ip address is distributed into the second memory node in system, so that client is virtual according to first
IP address is written and read the second memory node, avoids the first memory node failure and leads to the feelings of client traffic terminal
Condition improves the reliability of business in client.
Detailed description of the invention
Fig. 1 is the block schematic illustration of exemplary application scene provided by the embodiments of the present application；
Fig. 2 is a kind of flow chart of distributed memory system disaster recovery method provided by the embodiments of the present application；
Fig. 3 is a kind of flow chart of the method for client carry memory node provided by the embodiments of the present application；
Fig. 4 is a kind of structure chart of distributed memory system disaster tolerance device provided by the embodiments of the present application.
Specific embodiment
In order to make the above objects, features, and advantages of the present application more apparent, with reference to the accompanying drawing and it is specific real
Mode is applied to be described in further detail the embodiment of the present application.
Technical solution provided herein for ease of understanding is below first illustrated the background technique of the application.
Inventor has found that traditional storage strategy is deposited for distribution in traditional client storing data technique study
Storage system is the fixed memory node of each client configuration, and each client is written and read behaviour to the memory node distributed
Make, when the memory node failure that certain client is distributed, which can not continue data read-write operation, lead to business
Terminal affects user experience.In addition, causing the memory node to load when a certain memory node configures multiple client
It is heavier, influence memory node service performance.
Based on this, the embodiment of the present application provides a kind of disaster recovery method of distributed memory system, corresponding to client
The first memory node break down when, the information of the first memory node failure is sent to storage system by client, the storage
System judges the first memory node according to fault message, obtains corresponding first virtual ip address of the first memory node, and from
The second memory node is selected in remaining multiple memory node, the first virtual ip address is distributed into the second memory node, due to visitor
Family end has been stored with the first virtual ip address, therefore so that client by the first virtual ip address to the second memory node into
Row read-write operation, avoids service disconnection, promotes the reliability and user experience of business.In addition, the embodiment of the present application also mentions
A kind of load-balancing mechanism is supplied, so that multiple client balancedly to be distributed to multiple memory nodes in storage system.
Referring to Fig. 1, which is the block schematic illustration of exemplary application scene provided by the embodiments of the present application, wherein this Shen
Please be in the storage and disaster tolerance method application that provides of embodiment in distributed memory system 30, in specific implementation, distributed storage system
System 30 may include multiple memory nodes, select a memory node as primary storage node as control from multiple memory nodes
Node processed, with monitor and configuration system in other memory nodes, execute method provided by the embodiments of the present application.
As shown in Figure 1, including client 10 and client 20 under the application scenarios, distributed memory system 30 includes the
One memory node 301, the second memory node 303 and third memory node 304, wherein corresponding first memory node of client 10
301, it is configured with corresponding first virtual ip address of the first memory node 301, the corresponding third memory node 303 of client 20 is matched
It is equipped with the corresponding third virtual ip address of third memory node 303.When client 10 detects that event occurs in the first memory node 301
When barrier, fault message is sent to distributed memory system 30, and with obtaining corresponding first virtual IP address of the first memory node 301
It location, will for the heavier loads for avoiding third memory node 303 since third memory node 303 is that client 202 provides service
First virtual ip address is allocated to the second memory node 302, so that client 10 is according to the first virtual ip address to second
Memory node 302 carries out data read-write operation, is not influenced 10 business of client.
It will be understood by those skilled in the art that block schematic illustration shown in FIG. 1 is only that presently filed embodiment can be
An example being wherein achieved.The scope of application of the application embodiment is not limited by any aspect of the frame.
It should be noted that the client 10 and client 20 in the embodiment of the present application can be it is existing, researching and developing
Or in the future research and development, can by it is any type of wiredly and/or wirelessly connection (for example, Wi-Fi, LAN, honeycomb, coaxial electrical
Cable etc.) realize any user equipment interacted with the first memory node 301 and third memory node 303, including but not limited to: existing
Having, researching and developing or research and development in the future smart phones, non-smart mobile phone, tablet computer, laptop PC, desktop
Type personal computer, minicomputer, medium-size computer, mainframe computer etc..It is also to be noted that in the embodiment of the present application
First memory node 301, the second memory node 302, third memory node 303 can be existing, researching and developing or future
Research and development, equipment that can provide a user a Card read/write function example, presently filed embodiment is in this regard not
It is restricted.
The technical solution of the application for ease of understanding, below in conjunction with attached drawing to a kind of distribution provided by the embodiments of the present application
Formula storage system disaster recovery method is illustrated.
Referring to fig. 2, which is a kind of flow chart of distributed memory system disaster recovery method provided by the embodiments of the present application, such as
Shown in Fig. 2, this method can be applied to the host node in distributed memory system, and this method may include:
S201: the fault message that client is sent is received.
In the present embodiment, when client monitors break down to the first memory node itself connected, to distribution
Host node in storage system sends fault message, so that the host node in distributed memory system knows the first memory node
It breaks down.Wherein, the first memory node is a memory node in the distributed memory system, the first memory node failure
It can be the damage of the first memory node and/or the first memory node delay machine.Fault message occurs for identifying the first memory node
Failure；First memory node is the memory node that distributed memory system is client configuration.
It should be noted that host node is for each memory node in monitoring distributed storage system and is client point
With required memory node.In practical applications, distributed memory system includes multiple memory nodes, by certain mechanism from more
A host node is selected in a memory node, for managing other multiple memory nodes, while to need to establish the client of connection
End distribution memory node, so that client is written and read the client distributed.It wherein, is client about host node
The specific implementation of end distribution memory node will be illustrated in subsequent embodiment.
S202: corresponding first virtual ip address of the first memory node is obtained.
It, can be from pre- when host node judges that the first memory node breaks down by the fault message that client is sent
If obtaining corresponding first virtual ip address of the first memory node in configuration information.
In practical applications, host node can configure virtual ip address for each memory node, and the memory node pair
The virtual ip address answered is sent to client, so that client is accessed by virtual ip address, the virtual ip address is corresponding to be deposited
Store up node.In specific implementation, to meet access of the multi-client to memory node, host node can match for a memory node
Multiple virtual ip address are set, and each virtual ip address only corresponds to a memory node, so that memory node can be with
Service is provided for multiple client, to meet the requirements for access of a large amount of clients.
S203: distributing to the second memory node for the first virtual ip address, so that client is according to the first virtual IP address
Location is written and read second memory node.
In the present embodiment, corresponding first virtual ip address of the first memory node is distributed to the second storage section by host o'clock
Point, so that client accesses the second memory node using the first virtual ip address, to complete read-write operation.Wherein, second
Memory node is another memory node in the distributed memory system.
In practical applications, distributed memory system includes at least two memory nodes, when distributed memory system includes
Two memory nodes, when the first memory node breaks down, host node distributes the first virtual ip address of the first memory node
To the second memory node, so that client accesses the second memory node；When distributed memory system includes multiple memory nodes,
When the first memory node breaks down, host node selects the lesser memory node of load as second from multiple memory nodes
Memory node avoids the occurrence of the biggish storage section of load so that the client terminal quantity that each memory node is loaded is balanced
Point.
In addition, host node is the first memory node distribution virtual IP address again when the first memory node fault recovery is normal
Address, so that the first memory node can provide service for other clients.
In specific implementation, host node is that the first memory node distributes virtual ip address by ctdb, and ctdb passes through calling
Nfs order returns to the virtual address for the business that is currently executing, and then host node can be the one or more for the business that is not carried out
Virtual ip address distributes to the first memory node.When nfs order is returned as sky, the side of nfs order can be called by repetition
Formula obtains the virtual address for the business that is currently executing.When nfs order is returned as sky for a long time, which can be existed
In log, so that follow-up maintenance personnel can safeguard memory node according to log.Certainly, in practical applications, when
When one memory node restores normal, the first virtual ip address is directly reassigned to the first storage section by ctdb by host node
Point, so that client accesses the first memory node again.Wherein, the first virtual ip address distributes to first from the second memory node
During memory node, it is possible that of short duration service disconnection, but have no effect on the subsequent progress of business.
By foregoing description, in the embodiment of the present application when the first memory node of client institute carry breaks down, point
Cloth storage system according to client send about the first memory node failure when fault message, judge the first storage section
Point failure then obtains corresponding first virtual ip address of the first memory node, and one is selected from remaining multiple memory node
First virtual ip address is distributed to the second memory node in system, so that objective by memory node i.e. the second memory node
Family end is written and read the second memory node according to the first virtual ip address, avoids the first memory node failure and leads to client
The case where holding service terminal improves the reliability of business and the performance of distributed memory system in client.
In practical applications, client needs and stores section before the memory node in access distributed memory system
Communication connection is established between point, and memory node is written and read with will pass through the communication connection.Based on this, the application is implemented
Example provides the method for a kind of client and memory node foundation communication connection, says below in conjunction with attached drawing to this method
It is bright.
Referring to Fig. 3, which is the side that a kind of client provided by the embodiments of the present application and memory node establish communication connection
Method flow chart, as shown in figure 3, this method may include:
S301: the domain name carry request that client is sent is received.
In the present embodiment, when certain client needs to carry out data read-write operation, sends and connect to distributed memory system
Domain name is sent to client by request, the system, and then client sends domain name carry to distributed memory system according to domain name
Request, so that host node configures virtual ip address according to the request of domain name carry for the client in distributed memory system, thus
So that the corresponding memory node of the virtual ip address is carried out carry by client.Wherein, domain name carry request is deposited including distribution
The domain name of storage system, the domain name are that distributed memory system is what client configured in advance.
S302: the first virtual ip address is obtained according to domain name mapping.
In the present embodiment, host node carries out domain name solution according to the domain name in the request of domain name carry in distributed memory system
Analysis, and the first virtual ip address is obtained according to preconfigured virtual ip address distribution mechanism, first virtual ip address is unique
Corresponding first memory node, that is to say, that when the first virtual ip address distributes to client, client can according to this first
Virtual ip address accesses the first memory node.
In one possible implementation, each memory node is corresponding at least one in distributed memory system
A virtual ip address, when needing to be written and read there are multiple clients, according to preconfigured load balancing machine
It is made as each client and distributes a virtual ip address.
In practical applications, a virtual ip address uniquely corresponds to a memory node, but a memory node can be pre-
Multiple virtual ip address are first configured, so that a memory node can provide read-write service for multiple client, are met not
The disconnected a large number of users demand increased.It is main when needing to access the memory node in distributed memory system there are multiple client
Node can be that each client distributes a virtual ip address according to preconfigured load-balancing mechanism, so that each
Client can access the corresponding memory node of the virtual ip address according to the virtual ip address distributed.
It in the present embodiment, is that each client distributes a virtual IP address according to preconfigured load-balancing mechanism
Address specifically includes: determining each memory node connection client according to the quantity of the quantity of client and memory node
Quantity；For any memory node, the corresponding virtual ip address of memory node is sent to the memory node client to be connected
End.
In specific implementation, it is first determined the client terminal quantity of distributed memory system will be accessed, to guarantee each to deposit
Node load balancing is stored up, the connected client of each memory node is determined according to the quantity of client terminal quantity and memory node
Quantity.
When client terminal quantity is greater than the quantity of memory node, client terminal quantity is divided by memory node quantity, when not remaining
When number, the quotient according to acquisition is the client of memory node distribution connection；It is first to deposit according to the quotient of acquisition when there is remainder
The client for storing up node distribution connection, is then again sequentially allocated the corresponding client terminal quantity of remainder and gives remainder memory node,
The remainder memory node can be the memory node gone out according to the ID sequential selection of memory node.
When client terminal quantity is less than the quantity of memory node, client can be sequentially allocated and be deposited to client terminal quantity
Node is stored up, in specific implementation, the memory node distributed can be for according to the memory node of ID sequential selection, or presses
According to the memory node of the sequential selection of load from small to large.The mode for being specifically chosen memory node can be according to practical situations
It determines, the present embodiment is herein without limiting.
For ease of understanding, for example, distributed memory system includes 3 memory nodes, there is currently 6 clients to need to visit
Ask distributed memory system, then according to load-balancing mechanism, each memory node distributes 2 clients, can in specific distribution
Client 1 and client 2 are distributed to memory node 1, client 3 and client 4 distribute to memory node 2,5 He of client
Client 6 distributes to memory node 3, alternatively, client 1 distributes to memory node 1, client 2 distributes to memory node 2, visitor
Memory node 3 is distributed at family end 3, and client 4 distributes to memory node 1, client 5 distributes to memory node 2, client 6 is divided
Dispensing memory node 3.When there are 11 clients, according to load-balancing mechanism, each memory node distributes 3 clients,
Remaining 2 clients first can be that each memory node distributes 3 clients according to the above-mentioned method of salary distribution, then in distribution
Client 10 is distributed into memory node 1, client 11 distributes to memory node 2.
S303: being sent to client for the first virtual ip address, so that client is according to the first virtual ip address to first
Memory node is written and read.
In the present embodiment, the first virtual ip address that host node will acquire is sent to client, which receives
When the first virtual ip address, the first memory node is carried out by carry by the first virtual ip address, thus to the first memory node
Read-write operation.
It in practical applications, is that each memory node distributes corresponding client using load-balancing mechanism when host node
Afterwards, the corresponding virtual ip address of the memory node is sent to client.In specific implementation, when memory node corresponds to multiple visitors
When the end of family, since each memory node can configure multiple virtual ip address, then the corresponding multiple client institute of the memory node
The virtual ip address of distribution can be identical, it is, the corresponding multiple client of the memory node is by identical virtual IP address
The memory node is carried out carry by location.It is of course also possible to different virtual ip address be distributed for each client, so that each visitor
Family end passes through the virtual IP address carry memory node itself distributed.
The method provided through this embodiment, distributed memory system can be by load balancings by multiple client
Multiple memory nodes are balancedly distributed to, performance bottleneck caused by single memory node heavier loads is efficiently avoided, mentions
The high performance of distributed memory system.
Based on above method embodiment, the embodiment of the present application also provides a kind of distributed memory system disaster tolerance device, under
Face will be illustrated the device in conjunction with attached drawing.
Referring to fig. 4, which is a kind of distributed memory system disaster tolerance structure drawing of device provided by the embodiments of the present application, is such as schemed
Shown in 4, the apparatus may include:
First receiving unit 401, for receiving the fault message of client transmission；The fault message is for identifying first
Memory node breaks down；First memory node is the storage section that the distributed memory system is the client configuration
Point；First memory node is a memory node in the distributed memory system；
Acquiring unit 402, for obtaining corresponding first virtual ip address of first memory node；
First allocation unit 403, for first virtual ip address to be distributed to the second memory node, so that described
Client is written and read second memory node according to first virtual ip address；The client storage is
State the first virtual ip address；The distributed memory system includes at least two memory nodes.
In one possible implementation, described device further include: the second allocation unit, for when there are multiple described
When client needs to be written and read, an institute is distributed for each client according to preconfigured load-balancing mechanism
State virtual ip address；Each memory node is corresponding at least one virtual ip address in the distributed memory system.
In one possible implementation, second allocation unit includes:
Determine subelement, it is each described for being determined according to the quantity of the client and the quantity of the memory node
Memory node connects the quantity of the client；
Transmission sub-unit, for for any memory node, the corresponding virtual ip address of the memory node to be sent out
Give the memory node client to be connected.
In one possible implementation, described device further include: third allocation unit is deposited for working as described first
After storing up node recovery normally, virtual ip address is redistributed for first memory node.
In one possible implementation, described device further include:
Second receiving unit, for receiving the domain name carry that the client is sent before the execution of the first receiving unit
Request；The request of domain name carry includes the domain name of the distributed memory system；Domain name is the distributed storage system
System configures in advance for the client；
Resolution unit, for obtaining first virtual ip address according to domain name parsing；First virtual IP address
Location uniquely corresponds to first memory node；
First virtual ip address is sent to the client for this, so that the client root by transmission unit
First memory node is written and read according to first virtual ip address.
The embodiment of the present application when client institute carry the first memory node break down when, distributed memory system according to
Client send about the first memory node failure when fault message, judge the first memory node failure, then obtain the
Corresponding first virtual ip address of one memory node, and select a memory node i.e. second to deposit from remaining multiple memory node
Node is stored up, the first virtual ip address is distributed into the second memory node in system, so that client is virtual according to first
IP address is written and read the second memory node, avoids the first memory node failure and leads to the feelings of client traffic terminal
Condition improves the reliability of business in client.
It should be noted that each embodiment in this specification is described in a progressive manner, each embodiment emphasis is said
Bright is the difference from other embodiments, and the same or similar parts in each embodiment may refer to each other.For reality
For applying system or device disclosed in example, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, phase
Place is closed referring to method part illustration.
It should be appreciated that in this application, " at least one (item) " refers to one or more, and " multiple " refer to two or two
More than a."and/or" indicates may exist three kinds of relationships, for example, " A and/or B " for describing the incidence relation of affiliated partner
It can indicate: only exist A, only exist B and exist simultaneously tri- kinds of situations of A and B, wherein A, B can be odd number or plural number.Word
Symbol "/" typicallys represent the relationship that forward-backward correlation object is a kind of "or"." at least one of following (a) " or its similar expression, refers to
Any combination in these, any combination including individual event (a) or complex item (a).At least one of for example, in a, b or c
(a) can indicate: a, b, c, " a and b ", " a and c ", " b and c ", or " a and b and c ", and wherein a, b, c can be individually, can also
To be multiple.
It should also be noted that, herein, relational terms such as first and second and the like are used merely to one
Entity or operation are distinguished with another entity or operation, without necessarily requiring or implying between these entities or operation
There are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant are intended to contain
Lid non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
The foregoing description of the disclosed embodiments makes professional and technical personnel in the field can be realized or use the application.
Various modifications to these embodiments will be readily apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the application.Therefore, the application
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest scope of cause.
Claims (10)
1. a kind of disaster recovery method of distributed memory system, which is characterized in that the described method includes:
Receive the fault message that client is sent；The fault message breaks down for identifying the first memory node；Described
One memory node is the memory node that the distributed memory system is the client configuration；First memory node is institute
State a memory node in distributed memory system；
Obtain corresponding first virtual ip address of first memory node；The client is with being stored with first virtual IP address
Location；
First virtual ip address is distributed into the second memory node, so that the client is according to first virtual IP address
Address is written and read second memory node；Second memory node is another in the distributed memory system
Memory node；The distributed memory system includes at least two memory nodes.
2. the method according to claim 1, wherein each memory node in the distributed memory system
It is corresponding at least one virtual ip address, when needing to be written and read there are multiple clients, according to matching in advance
The load-balancing mechanism set is that each client distributes a virtual ip address.
3. according to the method described in claim 2, it is characterized in that, described when there are multiple client needs to be written and read
It is that each client distributes a virtual ip address according to preconfigured load-balancing mechanism, specifically when operation
Include:
Determine that each memory node connects the visitor according to the quantity of the quantity of the client and the memory node
The quantity at family end；
For any memory node, the corresponding virtual ip address of the memory node is sent to the memory node and is wanted
The client of connection.
4. the method according to claim 1, wherein being described after first memory node restores normal
First memory node redistributes virtual ip address.
5. method according to claim 1-4, which is characterized in that the failure for receiving the client and sending
Before information, the method also includes:
Receive the domain name carry request that the client is sent；The request of domain name carry includes the distributed memory system
Domain name；Domain name is that the distributed memory system is the client configuration in advance；
First virtual ip address is obtained according to domain name parsing；First virtual ip address uniquely corresponds to described first
Memory node；
First virtual ip address is sent to the client, so that the client is according to first virtual ip address
First memory node is written and read.
6. a kind of distributed memory system disaster tolerance device, which is characterized in that described device includes:
First receiving unit, for receiving the fault message of client transmission；The fault message is for identifying the first storage section
Point breaks down；First memory node is the memory node that the distributed memory system is the client configuration；Institute
Stating the first memory node is a memory node in the distributed memory system；
Acquiring unit, for obtaining corresponding first virtual ip address of first memory node；The client storage is
State the first virtual ip address；
First allocation unit, for first virtual ip address to be distributed to the second memory node, so that the client
Second memory node is written and read according to first virtual ip address；The client is stored with described first
Virtual ip address；The distributed memory system includes at least two memory nodes.
7. device according to claim 6, which is characterized in that described device further include: the second allocation unit is deposited for working as
It is each client according to preconfigured load-balancing mechanism when multiple clients need to be written and read
Distribute a virtual ip address；Each memory node is corresponding at least one void in the distributed memory system
Quasi- IP address.
8. device according to claim 7, which is characterized in that second allocation unit includes:
Subelement is determined, for determining each storage according to the quantity of the client and the quantity of the memory node
Node connects the quantity of the client；
Transmission sub-unit, for for any memory node, the corresponding virtual ip address of the memory node to be sent to
The memory node client to be connected.
9. device according to claim 6, which is characterized in that described device further include: third allocation unit, for working as institute
After stating the recovery normally of the first memory node, virtual ip address is redistributed for first memory node.
10. according to the described in any item devices of claim 6-9, which is characterized in that described device further include:
Second receiving unit, the domain name carry request sent for before the execution of the first receiving unit, receiving the client；
The request of domain name carry includes the domain name of the distributed memory system；Domain name is that the distributed memory system is preparatory
For client configuration；
Resolution unit, for obtaining first virtual ip address according to domain name parsing；First virtual ip address is only
One corresponding first memory node；
First virtual ip address is sent to the client for this, so that the client is according to institute by transmission unit
The first virtual ip address is stated to be written and read first memory node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810968427.9A CN109151028A (en) | 2018-08-23 | 2018-08-23 | A kind of distributed memory system disaster recovery method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810968427.9A CN109151028A (en) | 2018-08-23 | 2018-08-23 | A kind of distributed memory system disaster recovery method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109151028A true CN109151028A (en) | 2019-01-04 |
Family
ID=64791258
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810968427.9A Pending CN109151028A (en) | 2018-08-23 | 2018-08-23 | A kind of distributed memory system disaster recovery method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109151028A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110391933A (en) * | 2019-06-17 | 2019-10-29 | 浙江工商大学 | A kind of embedded type equipment parameter configuration recovery method and embedded device |
| CN110493060A (en) * | 2019-08-28 | 2019-11-22 | 北京浪潮数据技术有限公司 | A kind of virtual IP address distribution method and relevant apparatus |
| CN110750213A (en) * | 2019-09-09 | 2020-02-04 | 华为技术有限公司 | Hard disk management method and device |
| CN111130899A (en) * | 2019-12-30 | 2020-05-08 | 威创集团股份有限公司 | Service recovery method and system for distributed system |
| CN111209260A (en) * | 2019-12-30 | 2020-05-29 | 创新科技术有限公司 | NFS cluster based on distributed storage and method for providing NFS service |
| CN111258958A (en) * | 2020-01-10 | 2020-06-09 | 北京猎豹移动科技有限公司 | Data acquisition method, data providing method and device |
| CN111625421A (en) * | 2020-05-26 | 2020-09-04 | 云和恩墨(北京)信息技术有限公司 | Method and device for monitoring distributed storage system, storage medium and processor |
| CN111787113A (en) * | 2020-07-03 | 2020-10-16 | 北京大道云行科技有限公司 | Node fault processing method and device, storage medium and electronic equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102694863A (en) * | 2012-05-30 | 2012-09-26 | 电子科技大学 | Realization method of distributed storage system on basis of load adjustment and system fault tolerance |
| CN103475732A (en) * | 2013-09-25 | 2013-12-25 | 浪潮电子信息产业股份有限公司 | Distributed file system data volume deployment method based on virtual address pool |
| CN104283710A (en) * | 2014-08-18 | 2015-01-14 | 四川长虹电器股份有限公司 | Database cluster fault handling method and management server |
| CN105450775A (en) * | 2015-12-28 | 2016-03-30 | 曙光信息产业(北京)有限公司 | Load balancing method and load balancing device for DNS |
| CN107645421A (en) * | 2017-11-09 | 2018-01-30 | 郑州云海信息技术有限公司 | A kind of ISCSI protocol implementing methods of distributed storage |
| CN108259239A (en) * | 2018-01-11 | 2018-07-06 | 郑州云海信息技术有限公司 | A kind of database high availability support method and system |
-
2018
- 2018-08-23 CN CN201810968427.9A patent/CN109151028A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102694863A (en) * | 2012-05-30 | 2012-09-26 | 电子科技大学 | Realization method of distributed storage system on basis of load adjustment and system fault tolerance |
| CN103475732A (en) * | 2013-09-25 | 2013-12-25 | 浪潮电子信息产业股份有限公司 | Distributed file system data volume deployment method based on virtual address pool |
| CN104283710A (en) * | 2014-08-18 | 2015-01-14 | 四川长虹电器股份有限公司 | Database cluster fault handling method and management server |
| CN105450775A (en) * | 2015-12-28 | 2016-03-30 | 曙光信息产业(北京)有限公司 | Load balancing method and load balancing device for DNS |
| CN107645421A (en) * | 2017-11-09 | 2018-01-30 | 郑州云海信息技术有限公司 | A kind of ISCSI protocol implementing methods of distributed storage |
| CN108259239A (en) * | 2018-01-11 | 2018-07-06 | 郑州云海信息技术有限公司 | A kind of database high availability support method and system |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110391933A (en) * | 2019-06-17 | 2019-10-29 | 浙江工商大学 | A kind of embedded type equipment parameter configuration recovery method and embedded device |
| CN110493060A (en) * | 2019-08-28 | 2019-11-22 | 北京浪潮数据技术有限公司 | A kind of virtual IP address distribution method and relevant apparatus |
| CN110750213A (en) * | 2019-09-09 | 2020-02-04 | 华为技术有限公司 | Hard disk management method and device |
| CN111130899A (en) * | 2019-12-30 | 2020-05-08 | 威创集团股份有限公司 | Service recovery method and system for distributed system |
| CN111209260A (en) * | 2019-12-30 | 2020-05-29 | 创新科技术有限公司 | NFS cluster based on distributed storage and method for providing NFS service |
| WO2021136370A1 (en) * | 2019-12-30 | 2021-07-08 | 威创集团股份有限公司 | Service restoration method and system for distributed system |
| CN111258958A (en) * | 2020-01-10 | 2020-06-09 | 北京猎豹移动科技有限公司 | Data acquisition method, data providing method and device |
| CN111625421A (en) * | 2020-05-26 | 2020-09-04 | 云和恩墨(北京)信息技术有限公司 | Method and device for monitoring distributed storage system, storage medium and processor |
| CN111625421B (en) * | 2020-05-26 | 2021-07-16 | 云和恩墨(北京)信息技术有限公司 | Method and device for monitoring distributed storage system, storage medium and processor |
| CN111787113A (en) * | 2020-07-03 | 2020-10-16 | 北京大道云行科技有限公司 | Node fault processing method and device, storage medium and electronic equipment |
| CN111787113B (en) * | 2020-07-03 | 2021-09-03 | 北京大道云行科技有限公司 | Node fault processing method and device, storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109151028A (en) | A kind of distributed memory system disaster recovery method and device | |
| EP3832578A1 (en) | Electronic invoice identifier allocation method, and electronic ticket generating method, device and system | |
| JP7127010B2 (en) | Resource allocation methods, apparatus, electronic equipment, computer readable media and computer programs | |
| CN101699825B (en) | Network file system access method, device and system | |
| CN108023967B (en) | Data balancing method and device and management equipment in distributed storage system | |
| CN108023953B (en) | High-availability implementation method and device for FTP service | |
| CN104735108B (en) | A kind of method, apparatus and cloud desktop system using cloud desktop | |
| CN106790629A (en) | Data synchronization unit and its realize the method for data syn-chronization, client access system | |
| CN107656937B (en) | Method and device for realizing consistency of read-write data | |
| CN109933431B (en) | Intelligent client load balancing method and system | |
| CN108696581A (en) | Caching method, device, computer equipment and the storage medium of distributed information | |
| US8832215B2 (en) | Load-balancing in replication engine of directory server | |
| CN113806300B (en) | Data storage method, system, device, equipment and storage medium | |
| CN113645071A (en) | Cluster deployment method, system, medium and electronic terminal | |
| CN109446062B (en) | Method and device for debugging software in cloud computing service | |
| CN110633046A (en) | Storage method and device of distributed system, storage equipment and storage medium | |
| CN112087401A (en) | Method and device for realizing service quality in distributed storage | |
| CN111225003B (en) | NFS node configuration method and device | |
| CN110740155A (en) | Request processing method and device in distributed system | |
| CN116302352A (en) | Cluster disaster recovery processing method and device, electronic equipment and storage medium | |
| CN107092556B (en) | Test method, device and equipment | |
| CN107491551B (en) | Method and device for realizing client access of distributed file system | |
| CN110188140A (en) | Data pull method, apparatus, storage medium and computer equipment | |
| US10481963B1 (en) | Load-balancing for achieving transaction fault tolerance | |
| CN114827177A (en) | Deployment method and device of distributed file system and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190104 |