CN108540315A - Distributed memory system, method and apparatus - Google Patents
Distributed memory system, method and apparatus Download PDFInfo
- Publication number
- CN108540315A CN108540315A CN201810269046.1A CN201810269046A CN108540315A CN 108540315 A CN108540315 A CN 108540315A CN 201810269046 A CN201810269046 A CN 201810269046A CN 108540315 A CN108540315 A CN 108540315A
- Authority
- CN
- China
- Prior art keywords
- storage
- protected field
- node
- failure
- domain
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- 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
-
- 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
- H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
- H04L41/508—Network service management, e.g. ensuring proper service fulfilment according to agreements based on type of value added network service under agreement
- H04L41/5096—Network service management, e.g. ensuring proper service fulfilment according to agreements based on type of value added network service under agreement wherein the managed service relates to distributed or central networked applications
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Retry When Errors Occur (AREA)
Abstract
Present disclose provides a kind of distributed memory system, method and apparatus；The system includes multiple protected fields；Each protected field includes multiple failure domains, and each failure domain includes one group of memory node；Each copy of same data is stored on the memory node in the different faults domain in the same protected field; with in failure memory node; it is safeguarded in protected field belonging to failed storage node, in each protected field, the maximum allowable number of faults of failure domain is less than the quantity of copy.The disclosure can improve the safety and reliability of distributed memory system.
Description
Technical field
This disclosure relates to technical field of data storage, more particularly, to a kind of distributed memory system, method and apparatus.
Background technology
It with the fast development of Internet technology and is widely used, there is mass data continually to generate daily, therefore,
Storage system as data storage carrier is also more and more important.Poor, safety that heritage storage system usually has the scalability and
The defects of reliability is poor, the storage demand of user cannot be met by leading to it.Distributed memory system is with its good number
According to processing capacity and reliability, increasingly approved by user.
But since distributed memory system has the characteristics that distributed hardware structure, if one or more in system
Failure has occurred in storage server, can usually influence the data service performance of whole system, makes all services of system sometimes
It interrupts, increasing with the scale of system, the service range of influence is also increasing, causes the service of distributed memory system
Performance is relatively low.
Invention content
In view of this, the disclosure is designed to provide a kind of distributed memory system, method and apparatus, to improve distribution
The service performance of formula storage system.
To achieve the goals above, the technical solution that the disclosure uses is as follows：
In a first aspect, present disclose provides a kind of distributed memory system, which includes multiple protected fields；Each protection
Domain includes multiple failure domains, and each failure domain includes one group of memory node；Each copy of same data is stored in the same guarantor
It protects on the memory node in the different faults domain in domain, in failure memory node, in the guarantor belonging to failed storage node
Shield is safeguarded in domain, and in each protected field, the maximum allowable number of faults of failure domain is less than the quantity of copy.
Second aspect, present disclose provides a kind of distributed storage method, this method is applied to distributed memory system
Node is monitored, system includes multiple protected fields；Each protected field includes multiple failure domains, and each failure domain includes one group of storage section
Point；Method includes：Receive data to be stored；Generate multiple copies of data；Multiple copies of data are stored respectively to same
On the memory node in the different faults domain in one protected field.
The third aspect, present disclose provides a kind of distributed storage devices, including memory and processor, wherein storage
Device is for storing one or more computer instruction, and one or more computer instruction is executed by the processor, in realization
State distributed storage method.
Above-mentioned distributed memory system, method and apparatus can be by same numbers by the way that system is divided into multiple protected fields
According to each copy be stored on the memory node in the different faults domain in the same protected field, in failure memory node
When, it is safeguarded in the protected field belonging to failed storage node, which improves the tolerable maximum malfunctioning node of system
Quantity, the coverage and system for reducing failed storage node stop the probability servicing, make the service of distributed memory system
Performance is more stablized, and the safety and reliability of system is improved.
Other feature and advantage of the disclosure will illustrate in the following description, alternatively, Partial Feature and advantage can be with
Deduce from specification or unambiguously determine, or by implement the disclosure above-mentioned technology it can be learnt that.
To enable the above objects, features, and advantages of the disclosure to be clearer and more comprehensible, better embodiment cited below particularly, and match
Appended attached drawing is closed, is described in detail below.
Description of the drawings
It, below will be to specific in order to illustrate more clearly of disclosure specific implementation mode or technical solution in the prior art
Embodiment or attached drawing needed to be used in the description of the prior art are briefly described, it should be apparent that, in being described below
Attached drawing is some embodiments of the disclosure, for those of ordinary skill in the art, before not making the creative labor
It puts, other drawings may also be obtained based on these drawings.
Fig. 1 is a kind of structural schematic diagram for distributed memory system that disclosure embodiment provides；
Fig. 2 is the structural schematic diagram for another distributed memory system that disclosure embodiment provides；
Fig. 3 is the structural schematic diagram for another distributed memory system that disclosure embodiment provides；
Fig. 4 is the structural schematic diagram for another distributed memory system that disclosure embodiment provides；
Fig. 5 is the structural schematic diagram for another distributed memory system that disclosure embodiment provides；
Fig. 6 is the structural schematic diagram for another distributed memory system that disclosure embodiment provides；
Fig. 7 is a kind of flow chart for distributed storage method that disclosure embodiment provides；
Fig. 8 is a kind of structural schematic diagram for distributed storage devices that disclosure embodiment provides.
Specific implementation mode
To keep the purpose, technical scheme and advantage of disclosure embodiment clearer, below in conjunction with attached drawing to this public affairs
The technical solution opened is clearly and completely described, it is clear that and described embodiment is a part of embodiment of the disclosure,
Rather than whole embodiment.Based on the embodiment in the disclosure, those of ordinary skill in the art are not making creation
Property labour under the premise of the every other embodiment that is obtained, belong to the range of disclosure protection.
Distributed memory system is that data dispersion is stored in more independent equipment, utilizes more storage servers point
Load storage load；Relative to direct-connected additional storage, network attached storage, storage area network these traditional network storage framves
Structure, distributed memory system have preferable reliability, availability and access efficiency, are also easy to system scale extension.
For the application platform of enterprise-level, especially Internet Service Provider is (for example, shopping, retrieval, Map Services
The supplier of website or application program), with the continuous expansion of business scale, business becomes to become increasingly complex, and concurrent user asks
Ask more and more, data to be processed are also more and more；Distributed memory system can constantly expand with the business demand of enterprise
Exhibition enhances the processing capacity of distributed system entirety, to support at high concurrent and mass data by increasing number of servers
Reason ability meets business event and increases the calculating demand brought.
A kind of structural schematic diagram of distributed memory system shown in Figure 1；Management server (also referred to as monitoring section
Point) by data distributed and saved in the storage server (also referred to as memory node) being attached thereto, and record the guarantor of Various types of data
Deposit position, at the same monitor each storage server working condition (such as, if break down)；Client and management server
Communication connection.For example, user sends a data inquiry request, after management server receives the request, root by client
The storage location that corresponding data is searched according to the request, for example, the mark of storage server, the mark of storage hard disk in storage server
Know etc., to read data from the storage location, and it is back to client.
In distributed memory system, if wherein failure, management service has occurred in one or more storage server
Device provides the data service preserved in the storage server without normal direction client, usually reads failure to client returned data
Information；Especially for enterprise-class tools, even the storage server failure of very short time, the storage of this low reliability
A large amount of business service can be caused to fail, cause economic loss.
In order to make storage system that there is certain safety and reliability, it will usually which storage system is divided into multiple failures
One storage server can be divided into a failure domain by domain for example, as in Fig. 1；Data to be stored are generated again multiple
Copy, multiple copies are stored in respectively in different failure domains, when the storage server in one of failure domain breaks down
When, identical data can be obtained from other failure domains, make storage system that there is higher fault-resistant ability, reliability is higher,
To ensure that being smoothed out for data service, while the safety of data will not be affected.
The copy amount of above-mentioned generation is usually determined by the redundancy strategy of the business demand of enterprise or current storage system
It is fixed；Copy amount is bigger, and storage system reliability is stronger, but the data volume that can be stored is smaller；Copy amount is smaller, storage
System reliability is lower, but the data volume that can be stored is larger.
Furthermore it is also possible to be handled data to be stored using correcting and eleting codes technology；Specifically, first will be to be stored
Data are divided into n original data block, obtain m verification data block further according to this n original data block operation, n+m is obtained
A data block；When wherein arbitrarily no more than the error of the data block (including original data block and verification data block) of m blocks when, can be with
By remaining data block, that n original data block is recovered using corresponding restructing algorithm.In storage system, by this n+m
Data block is distributed in different failure domains；It, can be in this way, when the storage server in one of failure domain breaks down
Data block is obtained from other failure domains, and reconstructs original data block.
The division range and rank of above-mentioned failure domain, can according to the scale of storage system, business demand of enterprise etc. because
Element is adjusted, changes；For example, when storage system scale is smaller, one or more storage server can be divided into one
A failure domain is usually provided with multiple storage hard disks in each storage server；When system scale is larger, it will usually use base
The form stood, more storage servers of a BTS management, at this point it is possible to by one or more base stations, and its what is be managed deposit
Storage server is divided into a failure domain.In failure domain partition process, it will usually keep each failure domain consistent, i.e., each
Storage server quantity in failure domain is identical, and the performance configuration of each storage server in each failure domain is identical.
In general, if there are one memory nodes (for example, storage server or storage hard disk) to occur in a failure domain
Failure, the failure domain just need to carry out fault recovery；In failover procedure, storage system can continue offer service, but
Whole operational efficiency and performance can all be affected, and lead to the reduction of business service efficiency；If had in storage system multiple
Failure domain all breaks down node, and the failure domain quantity of failure node has reached the copy amount of above-mentioned data, at this time
Memory node where probably there are all copies of certain data all breaks down, and causes these loss of data, in order to
The business service failure for being related to these data is avoided, the usual storage system needs to interrupt service, to the failure to break down
It is repaired in domain.
When the scale of distributed memory system is increasing, in the failure domain quantity in system, and each failure domain
Memory node quantity can all increase；After memory node increase in failure domain, the probability of the failure domain failure node
It can increase；For example, when the memory node in failure domain is 4, then the possibility for the node that breaks down is 4；The failure domain is deposited
When storing up point spread to 10, then the possibility for the node that breaks down has been increased by 10.After failure domain quantity in system, deposit
The failure domain quantity of failure node easily reaches the copy amount of setting in storage system, i.e. storage system needs to interrupt clothes
The probability of business greatly increases；Storage system is frequently interrupted, and causes the stability of storage system poor, and service performance is affected,
It is difficult to meet the business demand of enterprise customer.
Based on the problem that above-mentioned large-scale storage systems service performance is poor, disclosure embodiment provides a kind of distribution
Formula storage system, method and apparatus；The technology can be applied in distributed storage cluster, storage system, it is particularly possible to apply
In the distributed memory system of enterprise customer.As shown in Fig. 2, the distributed memory system includes multiple protected fields；It is each to protect
Shield domain includes multiple failure domains, and each failure domain includes one group of memory node；Each copy of same data is stored in same
On the memory node in the different faults domain in protected field, in failure memory node, belonging to failed storage node
It is safeguarded in protected field.
In addition, in disclosure embodiment in each protected field, the maximum allowable number of faults of failure domain is less than copy
Quantity.
In present embodiment, failure domain quantity that protected field quantity that system includes, each protected field include and each
The quantity for the memory node that failure domain includes is not specifically limited；The multiple copies of data are stored in the same protected field at random
In different faults domain；Alternatively, according to the memory state of each failure domain in protected field, multiple copies are stored in the same protected field
Specified different faults domain in.With system include two protected fields in Fig. 2, each protected field include for three failure domains into
Row explanation；If same data include two copies, the two copies can be stored in two of the same protected field at random
In failure domain, or according to the memory state of each failure domain, it is stored in two specified failure domains of the same protected field.
If same data include three copies, these three copies can be stored in three failure domains of the same protected field respectively
In.
Since each copy of same data is stored in the same protected field, what each protected field preserved has been
Complete different data are not present the intersection of data, i.e., can not possibly preserve the pair of same data between protected field two-by-two each other
This；When system failure memory node needs to be serviced, the operational efficiency and property of the protected field belonging to the memory node
It can be affected；And the operational efficiency of other protected fields and property retention are constant, alleviate system because of a failed storage
The problem of node causes systematic entirety that can reduce improves the stability and reliability of system.
It can be seen from the above, in the case where being not provided with protected field, system usually requires that the maximum allowable number of faults of failure domain
Amount is less than the quantity of copy, since there may be data cross between each two failure domain in system, if there is failure section
The failure domain quantity of point has reached copy amount, and in order to avoid loss of data, the usual storage system needs to interrupt service, to hair
The failure domain of raw failure is repaired.If being provided with protected field through the above way, in each protected field, the maximum of failure domain permits
Perhaps number of faults again smaller than copy quantity, if the failure domain quantity of failure node has reached copy in a certain protected field
Quantity, it is only necessary to interrupt the data service of the protected field, other protected fields are unaffected.In general, only in all protected fields
When the failure domain quantity of interior failure node has reached copy amount, system integrally can just stop servicing；It is protected by being arranged
Domain greatly reduces the probability that system stops service, improves the reliability of system.
By taking distributed memory system shown in Fig. 2 as an example, it is assumed that the copy amount of data is 2；If one of protection
One failure domain in domain breaks down memory node, then the protected field carries out fault restoration, and shadow is not generated to another protected field
It rings；The memory node if two failure domains of one of protected field break down, which stops service, to another
Protected field does not also have an impact；It is another only when the two or more failure domains of one of protected field break down memory node
The two or more failure domains of a protected field also break down memory node when, the system that is equivalent to occur four failure domains occur therefore
Hinder memory node, which just stops service.
As shown in Fig. 2, if the system is not provided with protected field, or the function of closing protected field, in the copy amount of data
In the case of being similarly 2, the memory node if a failure domain breaks down, systematic entirety can be affected；If two
A or more failure domain breaks down memory node, and the system that is equivalent to two failure domains occurs and breaks down memory node, system
It will stop servicing.
By above-mentioned comparison it is found that the distributed memory system that disclosure embodiment provides, more by the way that system to be divided into
Each copy of same data can be stored in the memory node in the different faults domain in the same protected field by a protected field
On, in failure memory node, to be safeguarded in the protected field belonging to failed storage node, which, which improves, is
It unites tolerable maximum malfunctioning node quantity, reduces the coverage of failed storage node and system stops the probability of service,
Improve the safety and reliability of distributed memory system.
Referring to the structural schematic diagram of Fig. 3 and Fig. 4 another distributed memory system being respectively shown in；In the embodiment,
System includes 18 hosts (i.e. above-mentioned memory node), totally 6 base stations (being referred to as Rack or rack), each BTS management
3 hosts；In the system, the host of each base station and its management is a failure domain；In Fig. 4, a protected field includes three
Failure domain.
When data copy to be stored is 2, for example, copy A and copy B；Copy A and copy B is stored in protection simultaneously
In domain 1 or protected field 2；If being stored in protected field 1 simultaneously, two of which base can be selected in base station 1, base station 2 and base station 3
It stands and preserves copy A and copy B, for example, copy A is stored in base station 1, copy B is stored in base station 3.At this point, even if protection
There is hostdown there are one base station in domain 1 and protected field 2, system can still be run；Only protected field 1 and protected field
When occurring two base stations generation hostdowns in 2, system break service.
Similarly, as shown in figure 3, if the system is not provided with protected field, copy A and copy B may be maintained in base station 1
Into any two base station in base station 6；When there are mass data, cause the base station two-by-two of base station 1 to base station 6 that may preserve
The copy of same data；In 6 base stations, any one failure domain breaks down memory node, then systematic entirety can be by shadow
It rings；In 6 base stations, any two or more failure domain breaks down memory node, and system will stop servicing.
It can be seen from the above, the copy amount by data is limited, the failed storage number of nodes that allows in storage system or
Failed storage hard disk number in person's difference node is certain；Therefore, the memory node in system or storage hard disk quantity are more,
The system is because node failure causes the affected probability of system business bigger, usual linear relation with increase.By by system
Protected field is divided, this probability, the event in the failed storage node allowed in each protected field or different nodes can be reduced
The event in failed storage number of nodes or different nodes that entire cluster allows when hindering storage hard disk number from without dividing protected field
Barrier storage hard disk number is consistent.Therefore, for the distributed memory system of same scale, after having divided protected field, ensureing
In the case of continual service, it can hold when the number for the maximum malfunctioning node that system can be tolerated is compared to unallocated protected field
The number for the maximum malfunctioning node born has into the increase of multiple, by taking system shown in Fig. 3 and Fig. 4 as an example, when unallocated protected field,
The maximum malfunctioning node number that system can be tolerated is 1, and if it is greater than 1, system then interrupts service；After having divided two protected fields,
The maximum malfunctioning node number that system can be tolerated is 2, and if it is greater than 2, system then interrupts service.Therefore, which is effectively promoted
The safety of the reliability, stability, robustness and data of distributed memory system in abnormal cases.
In another embodiment, it is contemplated that for data itself there are multiple types, different types of data are corresponding
Service request is different；For example, for the data type for needing read or write speed fast, block storage service may be used；For needing altogether
The high data type of performance is enjoyed, file storage service may be used；For not only needing read or write speed fast but also needing shared performance high
Data type, then the larger object storage service of carrying cost may be used.
When distributed memory system is larger, the interface of above-mentioned a variety of storage services may be set, to provide difference
Storage service；But the interface distributing position of various storage services is more random, when storing data, by storage location, deposits
The limitation of the conditions such as space is stored up, storage system is difficult to select matched storage service according to the service request of data, usually will not
Data with service request are stored in same failure domain, same host, even same storage hard disk, lead to different service requests
Data on physical resource cross-distribution.
For example, the data type for needing shared performance high is preserved into block storage service, needing read or write speed fast
Data type is preserved into file storage service, and the storage mode of this specific admixture prevents the storage performance of various data from meeting
Service request causes the storage efficiency of system entirety relatively low.
For these reasons, embodiment further provides another distributed memory systems for the disclosure, as shown in figure 5, should
System includes multiple subregions, and the identical protected field of one or more storage performances is respectively arranged in each subregion；Each protection
The storage performance of memory node in domain is identical.
In order to make the protected field storage performance having the same in same subregion may be used same in same subregion
Kind storage service, the i.e. storage service of the subregion can be one kind in block storage, file storage and object storage；It is every in order to make
The storage performance of memory node in a protected field is identical, and it is identical that physical resource model may be used in the memory node in protected field
Or similar storage device, for example, SSD (Solid State Drive, solid state disk), which may be used, in memory node provides high property
Energy storage service may be used HDD (Hard Disk Drive, hard disk drive) and provide common performance storage service.
To in system memory node carry out subregion when, usually divided from physical layer, with realize each subregion it
Between isolation on physical resource, do not share physical resource between each subregion in addition to monitoring node, each subregion is pre- for storing
The corresponding data of storage service first configured, to ensure that different storage services use different physical resources.In physical layer
On face, the total collection of each memory node in each storage system can be considered as a big storage pool, and in each subregion
Memory node can also be considered as the corresponding storage pool of the subregion, each storage pool is used to provide a kind of storage clothes more segmented
Business；The storage performance for the memory node that different storage pools includes can be different, can be according to service when executing storage service
Requirement select different storage pools.By dividing storage pool, different storage services can be further realized on physical resource
Isolation.
System can be divided into multiple points by above-mentioned distributed memory system according to the performance of each memory node first
Area, then each subregion is divided into the identical protected field of storage performance, and the storage performance phase of the memory node in each protected field
Together；This partitioned mode makes the data distribution of different storage services on different physical resources, by the object for being unable to performance
Manage resource provide different performance storage service, so that data type is matched with storage performance, ensure system safety with
Under the premise of reliability, system storage efficiency is improved.
Embodiment further provides another distributed memory systems for the disclosure, as shown in fig. 6, system can be according to each
The factors such as the practical business demand of subregion, scale choose whether setting protected field；It, directly will be each in subregion 1 as shown in subregion 1
The host of base station and its management is divided into failure domain；And in subregion 2, first by base station 1, base station 2 and base station 3 and its master of management
Machine is divided into protected field 1, and the host of base station 4, base station 5 and base station 6 and its management is divided into protected field 2；It is with protected field 1
Example, then the host of each base station and its management is divided into failure domain.It is appreciated that usually each base station manage there are one or
Multiple main frames show that the host of portion of base stations is not shown in Fig. 6 for convenience.
Above-mentioned subregion can be as the storage region unit in cluster (i.e. above-mentioned distributed memory system), each subregion packet
Containing host in several clusters, different storage service services is provided to the user in different subregions；It can be met not by subregion
With between storage service data isolation and user use different performance block storage demand.On the basis of subregion, Mei Gefen
Area has further divided multiple protected fields, to improve the reliability of cluster；And failure domain is the minimum list of data distribution in cluster
Member, due to having divided failure domain, the several copies or data block of data can be distributed in several different under same subregion
In failure domain, when cluster breaks down, the reliability of cluster can be further increased.
Corresponding to above system embodiment, the disclosure additionally provides a kind of distributed storage method；This method is applied to
The monitoring node of distributed memory system, the monitoring node connect with each memory node in distributed memory system, are used for
Carry out the failure monitoring and Breakdown Maintenance of data storage and memory node.The system includes multiple protected fields；Each protected field packet
Multiple failure domains are included, each failure domain includes one group of memory node；As shown in fig. 7, this method comprises the following steps：
Step S702 receives data to be stored；
Step S704 generates multiple copies of data；
Step S706 stores multiple copies of data respectively to the storage section in the different faults domain in the same protected field
Point on.
The distributed storage method after receiving data to be stored, generates multiple copies of the data, and then will count again
According to multiple copies store respectively to the memory node in the different faults domain in the same protected field；Which can occur
It when failed storage node, is safeguarded in the protected field belonging to failed storage node, improves the tolerable most die of system
Hinder number of nodes, the coverage and system for reducing failed storage node stop the probability servicing, improve distributed storage
The safety and reliability of system.
In above system, the storage performance of the memory node in each protected field is identical；System further includes multiple subregions, often
It is respectively arranged with the identical protected field of one or more storage performances in a subregion, multiple copies are stored respectively to the same guarantor
The step on the memory node in the different faults domain in domain is protected, including：Judge the corresponding storage service of data, the storage service packet
Include block storage, object storage or file storage；Subregion is selected according to the corresponding storage service of data；By multiple copies of data point
It does not store to the memory node in the different faults domain in the same protected field of selected subregion.
The above method further includes：Monitor the memory node in each protected field；If monitoring failed storage node, and protect
Failure domain quantity in shield domain is not up to preset maximum allowable number of faults, in the protected field belonging to failed storage node into
Row is safeguarded；If monitoring failed storage node, and the failure domain quantity in protected field reaches preset maximum allowable number of faults
It measures, the storage service of the protected field belonging to outage memory node.
Present embodiments provide for a kind of a kind of distributed storage devices corresponding with above method embodiment.Fig. 8
For the structural schematic diagram of the distributed storage devices, which can be arranged the prison in distributed memory system
It controls on node；As shown in figure 8, the device includes memory 100 and processor 101；Wherein, memory 100 is for storing one
Or a plurality of computer instruction, one or more computer instruction are executed by processor, it, should to realize above-mentioned distributed storage method
Method may include one or more in above method.
Further, distributed storage devices shown in Fig. 8 further include bus 102 and communication interface 103, and processor 101 leads to
Letter interface 103 and memory 100 are connected by bus 102.
Wherein, memory 100 may include high-speed random access memory (RAM, Random Access Memory),
May further include non-labile memory (non-volatile memory), for example, at least a magnetic disk storage.By extremely
A few communication interface 103 (can be wired or wireless) is realized logical between the system network element and at least one other network element
Letter connection can use internet, wide area network, local network, Metropolitan Area Network (MAN) etc..Bus 102 can be isa bus, pci bus or
Eisa bus etc..The bus can be divided into address bus, data/address bus, controlling bus etc..For ease of indicating, only used in Fig. 8
One four-headed arrow indicates, it is not intended that an only bus or a type of bus.
Processor 101 may be a kind of IC chip, the processing capacity with signal.It is above-mentioned during realization
Each step of method can be completed by the integrated logic circuit of the hardware in processor 101 or the instruction of software form.On
The processor 101 stated can be general processor, including central processing unit (Central Processing Unit, abbreviation
CPU), network processing unit (Network Processor, abbreviation NP) etc.；It can also be digital signal processor (Digital
Signal Processing, abbreviation DSP), application-specific integrated circuit (Application Specific Integrated
Circuit, abbreviation ASIC), ready-made programmable gate array (Field-Programmable Gate Array, abbreviation FPGA) or
Person other programmable logic device, discrete gate or transistor logic, discrete hardware components.It may be implemented or execute sheet
Disclosed each method, step and logic diagram in disclosed embodiment.General processor can be microprocessor or this at
It can also be any conventional processor etc. to manage device.The step of method in conjunction with disclosed in disclosure embodiment, can direct body
Be now that hardware decoding processor executes completion, or in decoding processor hardware and software module combine and execute completion.It is soft
Part module can be located at random access memory, and flash memory, read-only memory, programmable read only memory or electrically erasable programmable are deposited
In the storage medium of this fields such as reservoir, register maturation.The storage medium is located at memory 100, and processor 101 reads storage
Information in device 100, in conjunction with its hardware complete aforementioned embodiments method the step of.
Embodiment of the present invention additionally provides a kind of machine readable storage medium, and machine readable storage medium storage is organic
Device executable instruction, for the machine-executable instruction when being called and being executed by processor, machine-executable instruction promotes processor
Realize that above-mentioned distributed storage method, specific implementation can be found in method embodiment, details are not described herein.
The technique effect of the distributed storage devices that embodiment of the present invention is provided, realization principle and generation and aforementioned
Method embodiment is identical, and to briefly describe, device embodiments part does not refer to place, can refer to preceding method embodiment
Middle corresponding contents.
In several embodiments provided herein, it should be understood that disclosed device and method can also lead to
Other modes are crossed to realize.Device embodiments described above are only schematical, for example, the flow chart in attached drawing and
Block diagram show the device of multiple embodiments according to the disclosure, method and computer program product system in the cards
Framework, function and operation.In this regard, each box in flowchart or block diagram can represent a module, program segment or generation
A part for code, the part of the module, section or code include one or more for realizing defined logic function
Executable instruction.It should also be noted that at some as in the realization method replaced, the function that is marked in box can also be with
Occur different from the sequence marked in attached drawing.For example, two continuous boxes can essentially be basically executed in parallel, they
Sometimes it can also execute in the opposite order, this is depended on the functions involved.It is also noted that block diagram and or flow chart
In each box and the box in block diagram and or flow chart combination, the special of function or action as defined in executing can be used
Hardware based system is realized, or can be realized using a combination of dedicated hardware and computer instructions.
In addition, each function module or unit in each embodiment of the disclosure can integrate to form one solely
Vertical part can also be modules individualism, can also two or more modules be integrated to form one it is independent
Part.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer read/write memory medium.Based on this understanding, the technical solution of the disclosure is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be expressed in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be
People's computer, server or network equipment etc.) execute all or part of step of each embodiment the method for the disclosure
Suddenly.And storage medium above-mentioned includes：USB flash disk, read-only memory (ROM, Read-Only Memory), is deposited mobile hard disk at random
The various media that can store program code such as access to memory (RAM, Random Access Memory), magnetic disc or CD.
Finally it should be noted that：Embodiment described above, the only specific implementation mode of the disclosure, to illustrate this public affairs
The technical solution opened, rather than its limitations, the protection domain of the disclosure are not limited thereto, although with reference to aforementioned embodiments pair
The disclosure is described in detail, it will be understood by those of ordinary skill in the art that：Any technology for being familiar with the art
Personnel can still modify to the technical solution recorded in aforementioned embodiments in the technical scope that the disclosure discloses
Or variation or equivalent replacement of some of the technical features can be readily occurred in；And these modifications, variation or replacement,
The spirit and scope for disclosure embodiment technical solution that it does not separate the essence of the corresponding technical solution, should all cover in this public affairs
Within the protection domain opened.Therefore, the protection domain of the disclosure shall be subject to the protection scope of the claim.
Claims (10)
1. a kind of distributed memory system, which is characterized in that the system comprises multiple protected fields；Each the protected field includes
Multiple failure domains, each failure domain include one group of memory node；
Each copy of same data is stored on the memory node in the different faults domain in the same protected field, to go out
When existing failed storage node, safeguarded in the protected field belonging to the failed storage node；
In each protected field, the maximum allowable number of faults of the failure domain is less than the quantity of the copy.
2. system according to claim 1, which is characterized in that the storage performance of the memory node in each protected field
It is identical；
The system also includes multiple subregions, the identical institute of one or more storage performances is respectively arranged in each subregion
State protected field.
3. system according to claim 2, which is characterized in that each subregion is for storing preconfigured storage clothes
It is engaged in corresponding data；The storage service includes block storage, object storage or file storage.
4. system according to claim 2, which is characterized in that the physical resource of the memory node in each protected field
Model is identical；The physical resource model includes solid-state hard disk SSD hard disk or hard disk drive HDD hard disks.
5. system according to claim 1, which is characterized in that multiple copies are stored in the same protected field at random
In different faults domain；Alternatively, according to the memory state of each failure domain in the protected field, multiple copies are stored in same
In the specified different faults domain of a protected field.
6. system according to claim 2, which is characterized in that the system also includes the prisons being connect with the memory node
Node is controlled, failure monitoring and Breakdown Maintenance for carrying out data storage and memory node.
7. a kind of distributed storage method, which is characterized in that the method is applied to the monitoring node of distributed memory system, institute
The system of stating includes multiple protected fields；Each protected field includes multiple failure domains, and each failure domain includes one group of storage
Node；The method includes：
Receive data to be stored；
Generate multiple copies of the data；
Multiple copies of the data are stored respectively to the memory node in the different faults domain in the same protected field.
8. the method according to the description of claim 7 is characterized in that the storage performance of the memory node in each protected field
It is identical；It is identical it to be respectively arranged with one or more storage performances the system also includes multiple subregions, in each subregion
The protected field, it is described to store multiple copies to the memory node in the different faults domain in the same protected field respectively
The step of, including：
Judge that the corresponding storage service of the data, the storage service include block storage, object storage or file storage；
Subregion is selected according to the corresponding storage service of the data；
Multiple copies of the data are stored into depositing to the different faults domain in the same protected field of selected subregion respectively
It stores up on node.
9. the method according to the description of claim 7 is characterized in that the method further includes：
Memory node in each protected field of monitoring；
If monitoring failed storage node, and the failure domain quantity in the protected field is not up to preset maximum allowable failure
Quantity is safeguarded in the protected field belonging to the failed storage node；
If monitoring failed storage node, and the failure domain quantity in the protected field reaches preset maximum allowable number of faults
Amount, interrupts the storage service of the protected field belonging to the failed storage node.
10. a kind of distributed storage devices, which is characterized in that including memory and processor, wherein the memory is for depositing
One or more computer instruction is stored up, one or more computer instruction is executed by the processor, to realize that right is wanted
Seek 7 to 9 any one of them methods.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810269046.1A CN108540315B (en) | 2018-03-28 | 2018-03-28 | Distributed storage system, method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810269046.1A CN108540315B (en) | 2018-03-28 | 2018-03-28 | Distributed storage system, method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108540315A true CN108540315A (en) | 2018-09-14 |
| CN108540315B CN108540315B (en) | 2021-12-07 |
Family
ID=63482272
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810269046.1A Active CN108540315B (en) | 2018-03-28 | 2018-03-28 | Distributed storage system, method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108540315B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110096301A (en) * | 2019-05-08 | 2019-08-06 | 深信服科技股份有限公司 | A kind of hot upgrade method of storage system, system and electronic equipment and storage medium |
| WO2020082888A1 (en) * | 2018-10-25 | 2020-04-30 | 华为技术有限公司 | Method, system and apparatus for restoring data in storage system |
| CN111104056A (en) * | 2018-10-25 | 2020-05-05 | 华为技术有限公司 | Data recovery method, system and device in storage system |
| CN111381770A (en) * | 2018-12-30 | 2020-07-07 | 浙江宇视科技有限公司 | Data storage switching method, device, equipment and storage medium |
| CN111522675A (en) * | 2019-02-05 | 2020-08-11 | 希捷科技有限公司 | Data allocation within a fault domain tree |
| 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 |
| CN113821165A (en) * | 2021-08-20 | 2021-12-21 | 济南浪潮数据技术有限公司 | Distributed cluster fusion storage method, system and equipment |
| CN114466030A (en) * | 2021-12-27 | 2022-05-10 | 天翼云科技有限公司 | Management method and device for data distribution storage strategy and distribution storage system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010048048A2 (en) * | 2008-10-24 | 2010-04-29 | Microsoft Corporation | Configuration management in distributed data systems |
| CN104735107A (en) * | 2013-12-20 | 2015-06-24 | 中国移动通信集团公司 | Recovery method and device for data copies in distributed storage system |
| CN105706056A (en) * | 2013-10-03 | 2016-06-22 | 微软技术许可有限责任公司 | Fault domains on modern hardware |
| CN105912612A (en) * | 2016-04-06 | 2016-08-31 | 中广天择传媒股份有限公司 | Distributed file system and data equilibrium distribution method orienting same |
| CN107085546A (en) * | 2016-02-16 | 2017-08-22 | 深圳市深信服电子科技有限公司 | Data managing method and device based on failure field technique |
-
2018
- 2018-03-28 CN CN201810269046.1A patent/CN108540315B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010048048A2 (en) * | 2008-10-24 | 2010-04-29 | Microsoft Corporation | Configuration management in distributed data systems |
| CN105706056A (en) * | 2013-10-03 | 2016-06-22 | 微软技术许可有限责任公司 | Fault domains on modern hardware |
| CN104735107A (en) * | 2013-12-20 | 2015-06-24 | 中国移动通信集团公司 | Recovery method and device for data copies in distributed storage system |
| CN107085546A (en) * | 2016-02-16 | 2017-08-22 | 深圳市深信服电子科技有限公司 | Data managing method and device based on failure field technique |
| CN105912612A (en) * | 2016-04-06 | 2016-08-31 | 中广天择传媒股份有限公司 | Distributed file system and data equilibrium distribution method orienting same |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111104056B (en) * | 2018-10-25 | 2021-12-31 | 华为技术有限公司 | Data recovery method, system and device in storage system |
| WO2020082888A1 (en) * | 2018-10-25 | 2020-04-30 | 华为技术有限公司 | Method, system and apparatus for restoring data in storage system |
| CN111104056A (en) * | 2018-10-25 | 2020-05-05 | 华为技术有限公司 | Data recovery method, system and device in storage system |
| US12111728B2 (en) | 2018-10-25 | 2024-10-08 | Huawei Technologies Co., Ltd. | Data recovery method, system, and apparatus in storage system |
| CN111381770A (en) * | 2018-12-30 | 2020-07-07 | 浙江宇视科技有限公司 | Data storage switching method, device, equipment and storage medium |
| CN111381770B (en) * | 2018-12-30 | 2021-07-06 | 浙江宇视科技有限公司 | Data storage switching method, device, equipment and storage medium |
| CN111522675A (en) * | 2019-02-05 | 2020-08-11 | 希捷科技有限公司 | Data allocation within a fault domain tree |
| CN111522675B (en) * | 2019-02-05 | 2024-04-05 | 希捷科技有限公司 | Data allocation within fault domain trees |
| CN110096301A (en) * | 2019-05-08 | 2019-08-06 | 深信服科技股份有限公司 | A kind of hot upgrade method of storage system, system and electronic equipment and storage medium |
| 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 |
| CN113821165A (en) * | 2021-08-20 | 2021-12-21 | 济南浪潮数据技术有限公司 | Distributed cluster fusion storage method, system and equipment |
| CN113821165B (en) * | 2021-08-20 | 2023-12-22 | 济南浪潮数据技术有限公司 | Distributed cluster fusion storage method, system and equipment |
| CN114466030A (en) * | 2021-12-27 | 2022-05-10 | 天翼云科技有限公司 | Management method and device for data distribution storage strategy and distribution storage system |
| CN114466030B (en) * | 2021-12-27 | 2024-03-12 | 天翼云科技有限公司 | Management method and device of data distributed storage strategy and distributed storage system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108540315B (en) | 2021-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108540315A (en) | Distributed memory system, method and apparatus | |
| CN107943421B (en) | Partition division method and device based on distributed storage system | |
| CN109725822B (en) | Method, apparatus and computer program product for managing a storage system | |
| US20190220379A1 (en) | Troubleshooting Method, Apparatus, and Device | |
| CN103929500A (en) | Method for data fragmentation of distributed storage system | |
| CN108319618B (en) | Data distribution control method, system and device of distributed storage system | |
| WO2017143957A1 (en) | Data redistribution method and device | |
| US11442827B2 (en) | Policy-based hierarchical data protection in distributed storage | |
| CN110737924B (en) | Data protection method and equipment | |
| CN110825704B (en) | Data reading method, data writing method and server | |
| US11023141B2 (en) | Resiliency schemes for distributed storage systems | |
| US8060773B1 (en) | Systems and methods for managing sub-clusters within a multi-cluster computing system subsequent to a network-partition event | |
| CN113625945A (en) | Distributed storage slow disk processing method, system, terminal and storage medium | |
| CN106899654A (en) | A kind of sequence value generation method, apparatus and system | |
| CN111124264A (en) | Method, apparatus and computer program product for reconstructing data | |
| CN105468296A (en) | No-sharing storage management method based on virtualization platform | |
| CN105404565B (en) | A kind of dual-active data guard method and device | |
| CN107422989A (en) | A kind of more copy read methods of Server SAN systems and storage architecture | |
| CN111866210A (en) | Virtual IP balance distribution method, system, terminal and storage medium | |
| CN112748867A (en) | Method, electronic device and computer program product for storage management | |
| CN115756955A (en) | Data backup and data recovery method and device and computer equipment | |
| CN105262640A (en) | System and method for improving reliability of cloud platform server, and disposition framework of system | |
| US11886309B2 (en) | Cell-based storage system with failure isolation | |
| CN109840051B (en) | Data storage method and device of storage system | |
| CN109117292A (en) | A kind of method and apparatus and cluster storage system of cluster-based storage |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |