CN106527978A - Loop-shaped virtual double controller-based multi-controller realization method - Google Patents
Loop-shaped virtual double controller-based multi-controller realization method Download PDFInfo
- Publication number
- CN106527978A CN106527978A CN201610912391.3A CN201610912391A CN106527978A CN 106527978 A CN106527978 A CN 106527978A CN 201610912391 A CN201610912391 A CN 201610912391A CN 106527978 A CN106527978 A CN 106527978A
- Authority
- CN
- China
- Prior art keywords
- controller
- virtual
- request
- physical disk
- controllers
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0614—Improving the reliability of storage systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0604—Improving or facilitating administration, e.g. storage management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0604—Improving or facilitating administration, e.g. storage management
- G06F3/0607—Improving or facilitating administration, e.g. storage management by facilitating the process of upgrading existing storage systems, e.g. for improving compatibility between host and storage device
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/061—Improving I/O performance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0629—Configuration or reconfiguration of storage systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0655—Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices
- G06F3/0658—Controller construction arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0662—Virtualisation aspects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0683—Plurality of storage devices
- G06F3/0689—Disk arrays, e.g. RAID, JBOD
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Memory System Of A Hierarchy Structure (AREA)
Abstract
The invention discloses a loop-shaped virtual double controller-based multi-controller realization method. According to the method, a physical disk controller is mapped into two virtual controllers; virtual controllers mapped by a plurality of physical disk controllers form a virtual controller loop; the two virtual controllers which are adjacent and do not belong to the same physical disk controller form a group of virtual double controllers, and then loop-shaped virtual double controllers are formed; a storage volume is accessed through mapping of multiple groups of virtual double controllers; each group of virtual double controllers is subjected to classification processing according to the type of a received request; the two virtual controllers in each group of virtual double controllers mutually form mirror images; and when the physical disk controllers are added or removed, a system can quickly reconstruct a new virtual controller loop and perform re-distribution on loads of the controllers. According to the method, the reading/writing performance of the system can be improved, and the data reliability and the load balance can be ensured; and the method has good expansibility and fault processing capability.
Description
Technical field
A kind of the invention belongs to technical field of information storage, more specifically, multi-controller reality based on the virtual dual control of ring-type
Existing method.
Background technology
With the high speed development of information technology, the explosive growth that information content occurs, the storage system of data center is by face
Face more acute challenge.As a example by 2002, the information of about 5EB, wherein 92% fresh information big portion is generated one year
Divide and be stored in disk, large capacity disc has become current requisite storage medium.
Disk array is that multiple disks are constituted an array, externally provides Large Copacity, and highly reliable, performance is of a relatively high
Data storage service.Memory technology based on raid-array RAID is just obtained from its birth within the storage system
Application widely, because can ensure that the reliability of system using the storage system of RAID structure and provide due to disk
Fail and the caused recovery for losing data.The high-reliability storage of important information is very crucial, according to external IT Policy
Compliance Group reports show that loss of data can cause tremendous influence to business event, publicly disclose loss of data
Enterprise estimated will cause that its client measures and related income reduces by 8%;For marketing enterprises, per share share price can decline 8%;
It is average often to lose the extraneoas loss that a user logging cause 100 dollars.
It is in prior art, although common disk array can provide jumbo data storage service, systemic
Can cannot meet present information industry, so storage array is gradually evolved into blending agent storage array by disk array.
Using RAID structure storage system to internal data have higher reliability, but its controller read or write speed and reliability
The high I/O bandwidth and high reliability demand of blending agent storage array cannot still be met.So, it would be desirable to it is a kind of to have more
The storage array framework of high I/O performances and higher reliability.
The content of the invention
Disadvantages described above or Improvement requirement for prior art, the invention provides a kind of based on many of the virtual dual control of ring-type
Controller implementation method, its object is to meet the high I/O bandwidth of blending agent storage array and high reliability demand, fully send out
Wave blending agent storage array performance.
For achieving the above object, according to one aspect of the present invention, there is provided a kind of many controls based on the virtual dual control of ring-type
Device implementation method processed, comprises the steps:
(1) two or more physical disk controllers are adopted, each physical disk controller is mapped to into two
Virtual Controller;The Virtual Controller of multiple physical disk controllers mapping constitutes Virtual Controller ring, adjacent and be not belonging to same
Two Virtual Controllers of one physical disk controller form one group of virtual dual control;Multigroup virtual dual control joins end to end and becomes ring
The virtual dual control of shape;
(2) data that client is accessed as needed, select corresponding virtual dual control to set up virtual link, by described
Multigroup virtual dual control maps the data of access storage volumes;
(3) per group of virtual dual control carries out classification process according to the type of the request for receiving:For physical disk controller is passed
The write request come, is responsible for receiving and processing request with a Virtual Controller of the virtual dual control of group, and another Virtual Controller is standby
Part is responsible for the write buffer of the Virtual Controller for receiving and processing request;The mutual shape of two Virtual Controllers in per group of virtual dual control
It is mirrored into;One group of virtual dual control is made up of two Virtual Controllers, during one of Virtual Controller processing write requests, in addition one
What individual Virtual Controller will back up the Virtual Controller writes data;After one of Virtual Controller failure, to failure
Virtual Controller enters the Virtual Controller of row write backup and can substitute the request that the Virtual Controller process of failure is received, and right
Data are recovered;For the read request that physical disk controller is transmitted, virtual dual control first it is respective it is data cached in look into
Look for, if cache hit directly returns request data, read request is issued to into disk memory array process if being not hit by.
Further, the virtual dual control of the ring-type can be reconstructed into one newly when increasing or remove physical disk controller
Virtual Controller ring, and the load of controller is carried out again balanced.The more performances of physical disk controller number are better, because
Load for each system can be evenly distributed in each virtual dual control.
Further, described step (3) include following sub-step:
(3-1) judge the request type for receiving, if read request then enters step (3-2), if write request then enters step
Suddenly (3-5);
(3-2) judge whether read request is hit in caching, step (3-4) is entered if hit, enter step if being not hit by
Suddenly (3-3);
(3-3) read request is issued to into disk memory array, Magnetic Disk Controler can be deposited by storage array indirect control
Storage volume, disk memory array read requests data be located data strip in caching, into step (3-4);
(3-4) request data returning result of caching, ending request are read;
(3-5) construct mirrored write request and be sent to backup controller, carry out mirror image write buffer backup, then process to write and ask
Ask;
(3-6) judge whether write request is hit in caching, step (3-7) is entered if hit, enter step if being not hit by
Suddenly (3-8);
(3-7) request data is write direct caching, into step (3-9);
(3-8) write request is issued to into disk array, disk array reads the data strip at request data place to caching
In, into step (3-7);
(3-9) after caching write success and the process mirrored write request success of standby virtual controller, return is responded and is tied
Beam is asked.
Further, step (3-5) includes following sub-step:
(3-5-1) after a Virtual Controller in one group of virtual dual control receives write request, as master controller process
Request, and with another Virtual Controller being organized as backup controller, process the mirrored write request that master controller sends;
(3-5-2) standby virtual controller receives the mirrored write request that main Virtual Controller sends, by mirrored write request plus
Enter the process queue of standby virtual controller;
(3-5-3) backup controller processes mirrored write request, and result is returned to main Virtual Controller.
Further, the present invention can flexibly increase or decrease physical disk controller, increase or decrease physical disk control
Device operation includes following sub-step:
(4-1) judge action type, to increase physical disk controller, into step (4-2), to remove physics magnetic
Disk controller, then into step (4-5).
(4-2) position of the physical disk controller of addition is selected, and control is sent to two adjacent physical disk controllers
Device addition request processed;
(4-3), after adjacent physical disk controller receives controller addition request, cancel the virtual dual control combination of correspondence,
Physical disk controller to new addition sends connection request;
(4-4) after the physical disk controller of new addition receives the connection request of contiguous physical Magnetic Disk Controler, mapping
Two Virtual Controllers will form new virtual double respectively with the virtual controlling of adjacent two physical disk controllers mapping
Control, into step (4-8);
(4-5) physical disk controller to be removed is selected, and controller is sent to two adjacent physical disk controllers
Remove request;
(4-6) contiguous physical Magnetic Disk Controler is received after controller removes request, cancels correspondence virtual dual control combination, together
When send connection request to another adjacent physical disk controller of removed controller;
(4-7) after two adjacent physical disk controllers receive mutual connection request, corresponding Virtual Controller
New virtual dual control is formed, all I/O requests of physical disk controller to be removed are taken over;
(4-8) remaining physical disk controller re-forms a ring-type Virtual Controller, and to each physical disk control
I/O loads between device processed carry out again balanced.
In general, by the contemplated above technical scheme of the present invention compared with prior art, can obtain down and show
Beneficial effect;
(1) the multi-controller implementation method based on the virtual dual control of ring-type that the present invention is provided, using Virtual Controller technology,
Each physical disk controller is mapped to into two Virtual Controllers, between controller in the form of the virtual dual control of ring-type pipe
Reason, two Virtual Controllers in per group of virtual dual control mutually form mirror image, not only improve system read-write performance, while also ensureing
The reliability of data;
(2) the multi-controller implementation method based on the virtual dual control of ring-type that the present invention is provided, storage volume is by multigroup virtual
Dual control mapping is accessed, and the load between active balance controller improves the degree of parallelism that request is processed, there is provided higher I/O services
Ability;
(3) the multi-controller implementation method based on the virtual dual control of ring-type that the present invention is provided, by ring-type Virtual Controller
Structure forms multigroup virtual dual control, and when increasing or removing physical disk controller, system can be a new void with quick reconfiguration
Intend controller ring, be with good expansibility and troubleshooting capability.
Description of the drawings
Fig. 1 is the virtual dual control structural representation of ring-type in the embodiment of the present invention;
Fig. 2 is that the mapping of the storage volume and virtual dual control in the embodiment of the present invention accesses schematic diagram;
Fig. 3 is the read-write operation flow chart in the embodiment of the present invention;
Fig. 4 is the Virtual Controller mirror back-up write buffer schematic diagram in the embodiment of the present invention;
Fig. 5 is removing in the embodiment of the present invention and adds physics Magnetic Disk Controler schematic diagram.
Specific embodiment
In order that the objects, technical solutions and advantages of the present invention become more apparent, it is below in conjunction with drawings and Examples, right
The present invention is further elaborated.It should be appreciated that specific embodiment described herein is only to explain the present invention, and
It is not used in the restriction present invention.As long as additionally, technical characteristic involved in invention described below each embodiment
Do not constitute conflict each other can just be mutually combined.
The present invention Integral Thought be:Using Virtual Controller technology, each physical disk controller is mapped to into two
Individual Virtual Controller, forms ring-type Virtual Controller structure, constitutes multiple virtual dual controls;It is virtually double using ring-type between controller
The format management of control, two Virtual Controllers in per group of virtual dual control mutually form mirror image, improve readwrite performance and data can
By property;The addition of physical disk controller rapidly and efficiently can be realized by the multiple virtual dual control that ring-type Virtual Controller is constituted
Or remove, it is with good expansibility and troubleshooting capability.
Shown in Fig. 1, it is the virtual dual control structural representation of ring-type in the embodiment of the present invention;Wherein, A, B, C and D are four things
Reason Magnetic Disk Controler, each physical disk controller map out two Virtual Controllers, and the Virtual Controller that such as A is mapped out is A1
And A2.Then A1 and D1, D2 and C1 are formed, C2 and B2, B1 and A2 this four groups of virtual dual controls, two in per group of virtual dual control are empty
Intend controller mutually form mirror image (each Virtual Controller has a logical place, by logical place table be known that with
The adjacent Virtual Controller of the Virtual Controller, adjacent and two virtually controls by a physical disk controller virtually out
Device meeting Auto-matching processed becomes one group of virtual dual control).The design of virtual dual control can be prevented effectively from Single Point of Faliure, and empty with group
Intend double aerial Virtual Controllers to be mutually mirrored into, back up mutually, further improve the reliability of data.
Shown in Fig. 2, it is that the mapping of storage volume and virtual dual control in the embodiment of the present invention accesses schematic diagram;Have in figure 2 it is objective
Family end, each client may be connected to arbitrary virtual dual control, and such as in figure, each client is connected with 2 virtual dual controls.
Storage pool is divided into 10 volumes, and each virtual dual control connects several storage volume, and such client can be by selecting corresponding void
Intend dual control to access the data of the overall situation.
The multi-controller implementation method based on the virtual dual control of ring-type provided by embodiment, comprises the following steps:
(1) two or more physical disk controllers are adopted, each physical disk controller is mapped to into two
Software Virtual Controller;The Virtual Controller of multiple physical disk controllers mapping constitutes Virtual Controller ring, adjacent and do not belong to
One group of virtual dual control is formed in two Virtual Controllers of same physical disk controller;Multigroup virtual dual control is joined end to end into
For the virtual dual control of ring-type;
(2) client selects corresponding virtual dual control to set up virtual link, by institute by the data for accessing as needed
The multigroup virtual dual control stated maps the data of access storage volumes;
(3) per group of virtual dual control carries out classification process according to the type of the request for receiving:For physical disk controller is passed
The write request come, is responsible for receiving and processing request with a Virtual Controller of the virtual dual control of group, another Virtual Controller mirror
The write buffer of the Virtual Controller for receiving and processing request as backup is responsible;Two Virtual Controller phases in per group of virtual dual control
Mirror image is formed mutually;For the read request that physical disk controller is transmitted, virtual dual control first in respective data cached middle lookup,
If cache hit directly returns request data, read request is issued to into disk memory array process if being not hit by;
(4) the virtual dual control of ring-type can be reconstructed into a new virtual controlling when increasing or remove physical disk controller
Device ring, and the load to controller carries out equilibrium again.The more performances of physical disk controller number are better, because each system
Load can be evenly distributed in each virtual dual control.
Shown in Fig. 3, it is read-write operation flow chart in the embodiment of the present invention;Memory array system can select suitable control first
Device processed sends request, after request is issued to cache module, carries out classification process to request.
In embodiment, step (3) includes following sub-step:
(3-1) judge the request type for issuing, if read request then enters step (3-2), if write request then enters step
Suddenly (3-5)
(3-2) judge whether read request is hit in caching, step (3-4) is entered if hit, enter step if being not hit by
Suddenly (3-3)
(3-3) read request is issued to into disk memory array, Magnetic Disk Controler can be deposited by storage array indirect control
Storage volume, disk memory array read requests data be located data strip in caching, into step (3-4);
(3-4) request data returning result of caching, ending request are read;
(3-5) construct mirrored write request and be sent to standby virtual controller, carry out mirror image write buffer backup, then process and write
Request
(3-6) judge whether write request is hit in caching, step (3-7) is entered if hit, enter step if being not hit by
Suddenly (3-8)
(3-7) request data is write direct caching, into step (3-9);
(3-8) request is issued to disk array, and disk array reads request bar and takes in caching, into step (3-7)
(3-9) after caching write success and backup server process mirrored write request success, return responds and terminates to ask
Ask.
Wherein, the flow process of step (3-5) is as shown in figure 4, including following sub-step:
(3-5-1), after a Virtual Controller in one group of virtual dual control receives write request, understand at as master controller
Reason request, and another Virtual Controller will process the mirrored write request that master controller sends as backup controller;
(3-5-2) standby virtual controller receives the mirrored write request that main Virtual Controller sends, by mirrored write request plus
Enter the process queue of standby virtual controller;
(3-5-3) backup controller processes mirrored write request, and result is returned to main Virtual Controller.
Shown in Fig. 5, it is removing in the embodiment of the present invention and adds physics Magnetic Disk Controler schematic diagram, each physical disk
Controller is mainly made up of tri- parts of Cache, CPU and PCIe, and CTL1~4 represent physics Magnetic Disk Controler 1~4 respectively;
Wherein, each physical disk controller is mapped to two Virtual Controllers, adjacent and be not belonging to the control of unified physics disk
Virtual Controller is matched two-by-two, constitutes virtual dual control, and two Virtual Controllers in same group of virtual dual control are mutually mirrored into, mutually
Backup write buffer, so its corresponding physical controller also mutually can be mirrored into, backs up mutually write buffer.When CTL3 breaks down
Afterwards, CTL3 is removed from system, the I/O requests for taking over CTL3 by CTL4, among the data backup of CTL2 goes to CTL4, work as CTL3
After the completion of reparation, again CTL3 is added in system;
In embodiment, step (4) includes following sub-step
(4-1) when needing to increase physical disk controller, into step (4-2), when needing to remove physical disk controller,
Step (4-5) is entered then;
(4-2) position of the physical disk controller of addition is selected, and control is sent to two adjacent physical disk controllers
Device addition request processed;
(4-3), after physical disk controller receives the addition request, cancel the virtual dual control combination of correspondence, to new addition
Physical disk controller send connection request;
(4-4), after the physical disk controller of new addition receives the connection request of neighbor controller, two of mapping are empty
Intend controller and the Virtual Controller with two adjacent physical disk controller mappings is formed into new virtual dual control respectively, enter
Step (4-8)
(4-5) physical disk controller to be removed is selected, and controller is sent to two adjacent physical disk controllers
Remove request;
(4-6) contiguous physical Magnetic Disk Controler is received after physical disk controller removes request, cancels correspondence virtual double
Control combination, while send connection request to another adjacent physical disk controller of removed controller;
(4-7) after two adjacent physical disk controllers receive mutual connection request, corresponding virtual controlling shape
The virtual dual control of Cheng Xin, takes over all I/O requests of the physical disk controller that will be removed;
(4-8) physical disk controller re-forms a ring-type Virtual Controller, and to each physical disk controller
Between I/O load redistributed.
As it will be easily appreciated by one skilled in the art that the foregoing is only presently preferred embodiments of the present invention, not to
The present invention, all any modification, equivalent and improvement made within the spirit and principles in the present invention etc. is limited, all should be included
Within protection scope of the present invention.
Claims (5)
1. a kind of multi-controller implementation method based on the virtual dual control of ring-type, it is characterised in that comprise the steps:
(1) two or more physical disk controllers are adopted, each physical disk controller is mapped to into two virtually
Controller;The Virtual Controller of multiple physical disk controllers mapping constitutes Virtual Controller ring, adjacent and be not belonging to same
Two Virtual Controllers of physical disk controller form one group of virtual dual control;Multigroup virtual dual control joins end to end becomes ring-type void
Intend dual control;
(2) data that client is accessed as needed, select corresponding virtual dual control to set up virtual link, by described multigroup
Virtual dual control maps the data of access storage volumes;
(3) per group of virtual dual control carries out classification process according to the type of the request for receiving:For physical disk controller is transmitted
Write request, is responsible for receiving and processing request with a Virtual Controller of the virtual dual control of group, and another Virtual Controller backup is negative
Blame the write buffer of the Virtual Controller for receiving and processing request;Two Virtual Controllers in per group of virtual dual control mutually form mirror
Picture;For the read request that physical disk controller is transmitted, virtual dual control is first in respective data cached middle lookup, if caching
Hit directly returns request data, and read request is issued to disk memory array process if being not hit by.
2. method according to claim 1, it is characterised in that the virtual dual control of the ring-type is increasing or removing physical disk
During controller, a new Virtual Controller ring can be reconstructed into, and the load to controller carries out equilibrium again.
3. method according to claim 1 and 2, it is characterised in that described step (3) include following sub-step:
(3-1) judge the request type for receiving, if read request then enters step (3-2), if write request then enters step (3-
5);
(3-2) judge whether read request is hit in caching, step (3-4) is entered if hit, enter step (3- if being not hit by
3);
(3-3) read request is issued to into disk memory array, the data strip that disk memory array read requests data are located is arrived
In caching, into step (3-4);
(3-4) request data returning result of caching, ending request are read;
(3-5) construct mirrored write request and be sent to backup controller, carry out mirror image write buffer backup, then processing write requests;
(3-6) judge whether write request is hit in caching, step (3-7) is entered if hit, enter step (3- if being not hit by
8);
(3-7) request data is write direct caching, into step (3-9);
(3-8) write request is issued to into disk array, disk array reads the data strip at request data place in caching, enters
Enter step (3-7);
(3-9) after caching write success and the process mirrored write request success of standby virtual controller, return responds and terminates to ask
Ask.
4. method according to claim 1 and 2, it is characterised in that step (3-5) includes following sub-step:
(3-5-1), after a Virtual Controller in one group of virtual dual control receives write request, ask as master controller process,
And with another Virtual Controller being organized as backup controller, process the mirrored write request that master controller sends;
(3-5-2) standby virtual controller receives the mirrored write request that main Virtual Controller sends, and mirrored write request is added standby
The process queue of part Virtual Controller;
(3-5-3) backup controller processes mirrored write request, and result is returned to main Virtual Controller.
5. method according to claim 2, it is characterised in that increasing or decreasing physical disk controller includes following sub-step
Suddenly:
(4-1) judge action type, to increase physical disk controller, into step (4-2), to remove physical disk control
Device processed, then into step (4-5);
(4-2) position of the physical disk controller of addition is selected, and controller is sent to two adjacent physical disk controllers
Addition request;
(4-3), after adjacent physical disk controller receives controller addition request, the virtual dual control combination of correspondence, Xiang Xin are cancelled
The physical disk controller of addition sends connection request;
(4-4) after the physical disk controller of new addition receives the connection request of contiguous physical Magnetic Disk Controler, the two of mapping
Virtual controlling with two adjacent physical disk controller mappings is formed new virtual dual control by individual Virtual Controller respectively, is entered
Enter step (4-8);
(4-5) physical disk controller to be removed is selected, and controller is sent to two adjacent physical disk controllers and is removed
Request;
(4-6) contiguous physical Magnetic Disk Controler is received after controller removes request, cancels correspondence virtual dual control combination, while to
Another adjacent physical disk controller of removed controller sends connection request;
(4-7), after two adjacent physical disk controllers receive mutual connection request, corresponding Virtual Controller is formed
New virtual dual control, takes over all I/O requests of physical disk controller to be removed;
(4-8) remaining physical disk controller re-forms a ring-type Virtual Controller, and to each physical disk controller
Between I/O loads carry out again it is balanced.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610912391.3A CN106527978B (en) | 2016-10-19 | 2016-10-19 | A kind of multi-controller implementation method based on cyclic annular virtual dual control |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610912391.3A CN106527978B (en) | 2016-10-19 | 2016-10-19 | A kind of multi-controller implementation method based on cyclic annular virtual dual control |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106527978A true CN106527978A (en) | 2017-03-22 |
CN106527978B CN106527978B (en) | 2019-07-09 |
Family
ID=58332524
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610912391.3A Expired - Fee Related CN106527978B (en) | 2016-10-19 | 2016-10-19 | A kind of multi-controller implementation method based on cyclic annular virtual dual control |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106527978B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107621996A (en) * | 2017-09-26 | 2018-01-23 | 郑州云海信息技术有限公司 | Cache backup method, fault handling method, system, device and storage medium |
CN111124767A (en) * | 2019-12-29 | 2020-05-08 | 北京浪潮数据技术有限公司 | Multi-control storage device implementation method, device, equipment and readable storage medium |
WO2023184907A1 (en) * | 2022-03-30 | 2023-10-05 | 苏州浪潮智能科技有限公司 | Multi-control storage system io processing method and system, and related component |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20060005076A1 (en) * | 1998-09-24 | 2006-01-05 | Yottayotta, Inc. | Methods and systems for implementing shared disk array management functions |
CN1808401A (en) * | 2005-01-20 | 2006-07-26 | 英业达股份有限公司 | Data accessing system and method capable of recognizing disk cache content validity |
CN101118478A (en) * | 2007-09-10 | 2008-02-06 | 杭州华三通信技术有限公司 | Caching management system |
CN103092786A (en) * | 2013-02-25 | 2013-05-08 | 浪潮(北京)电子信息产业有限公司 | Double-control double-active storage control system and method |
CN105573857A (en) * | 2014-10-10 | 2016-05-11 | 北京计算机技术及应用研究所 | Method and system for buffering mirror image by multi-control disk array |
-
2016
- 2016-10-19 CN CN201610912391.3A patent/CN106527978B/en not_active Expired - Fee Related
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20060005076A1 (en) * | 1998-09-24 | 2006-01-05 | Yottayotta, Inc. | Methods and systems for implementing shared disk array management functions |
CN1808401A (en) * | 2005-01-20 | 2006-07-26 | 英业达股份有限公司 | Data accessing system and method capable of recognizing disk cache content validity |
CN101118478A (en) * | 2007-09-10 | 2008-02-06 | 杭州华三通信技术有限公司 | Caching management system |
CN103092786A (en) * | 2013-02-25 | 2013-05-08 | 浪潮(北京)电子信息产业有限公司 | Double-control double-active storage control system and method |
CN105573857A (en) * | 2014-10-10 | 2016-05-11 | 北京计算机技术及应用研究所 | Method and system for buffering mirror image by multi-control disk array |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107621996A (en) * | 2017-09-26 | 2018-01-23 | 郑州云海信息技术有限公司 | Cache backup method, fault handling method, system, device and storage medium |
CN111124767A (en) * | 2019-12-29 | 2020-05-08 | 北京浪潮数据技术有限公司 | Multi-control storage device implementation method, device, equipment and readable storage medium |
CN111124767B (en) * | 2019-12-29 | 2022-03-22 | 北京浪潮数据技术有限公司 | Multi-control storage device implementation method, device, equipment and readable storage medium |
WO2023184907A1 (en) * | 2022-03-30 | 2023-10-05 | 苏州浪潮智能科技有限公司 | Multi-control storage system io processing method and system, and related component |
Also Published As
Publication number | Publication date |
---|---|
CN106527978B (en) | 2019-07-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP3997061B2 (en) | Storage subsystem and storage subsystem control method | |
US8341199B2 (en) | Storage system, a method of file data back up and a method of copying of file data | |
CN106066890B (en) | Distributed high-performance database all-in-one machine system | |
JP4790372B2 (en) | Computer system for distributing storage access load and control method thereof | |
CN107798130A (en) | A kind of Snapshot Method of distributed storage | |
CN104981788B (en) | The control method of storage system and storage system | |
US20090125678A1 (en) | Method for reading data with storage system, data managing system for storage system and storage system | |
CN103246616A (en) | Global shared cache replacement method for realizing long-short cycle access frequency | |
CN101571815A (en) | Information system and I/O processing method | |
US7260681B2 (en) | Stripe buffer list | |
US10657062B2 (en) | Distributed storage system and distributed storage control method wherein data stored in cache matrix, that is to be stored in said distributed storage, is tracked by columns and rows | |
US8275937B2 (en) | Storage system and processing efficiency improving method of storage system | |
CN106527978B (en) | A kind of multi-controller implementation method based on cyclic annular virtual dual control | |
CN105897859A (en) | Storage system | |
CN104503923A (en) | Asymmetrical disk array caching dispatching method | |
CN104137069B (en) | Network startup system | |
CN106227464A (en) | A kind of double-deck redundant storage system and data write, reading and restoration methods | |
WO2023065654A1 (en) | Data writing method and related device | |
CN105468296A (en) | No-sharing storage management method based on virtualization platform | |
CN112379825B (en) | Distributed data storage method and device based on data feature sub-pools | |
CN100464290C (en) | Caching management system | |
CN113986830B (en) | Cloud data management and task scheduling method and system for distributed CT | |
CN113721849B (en) | Data copying and unloading method based on distributed storage and terminal equipment | |
US8001347B1 (en) | Coherently sharing any form of instant snapshots separately from base volumes | |
CN116893789B (en) | Data management method, system, device, equipment and computer storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | 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 | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190709 Termination date: 20201019 |