CN107632788A - A kind of method of more controlled storage system I O schedulings and more controlled storage systems - Google Patents
A kind of method of more controlled storage system I O schedulings and more controlled storage systems Download PDFInfo
- Publication number
- CN107632788A CN107632788A CN201710883612.3A CN201710883612A CN107632788A CN 107632788 A CN107632788 A CN 107632788A CN 201710883612 A CN201710883612 A CN 201710883612A CN 107632788 A CN107632788 A CN 107632788A
- Authority
- CN
- China
- Prior art keywords
- instruction
- data block
- target controller
- client
- controller
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of method of more controlled storage system I O schedulings, this method includes client and obtains I/O instruction；The client selects any one virtual route to map the virtual route in storage device obtained by same data block respectively as destination virtual path, the virtual route for multiple controllers according to the I/O instruction from multiple virtual routes prestored；The client sends the I/O instruction to the target controller in the corresponding destination virtual path；The target controller performs corresponding operation to the data block according to the I/O instruction；Method disclosed by the invention, client can perform the operation of corresponding I/O instruction by different controllers to same data block, and cooperate work between multiple controllers, can effectively reduce the operating pressure of controller, the time delay of read-write operation is reduced, improves the performance of whole storage system；The invention also discloses a kind of more controlled storage systems, equally with above-mentioned beneficial effect.
Description
Technical field
The present invention relates to field of data storage, method and more controls more particularly to a kind of more controlled storage system I O schedulings are deposited
Storage system.
Background technology
As science and technology is constantly progressive in recent years, application of more controlled storage systems in enterprise is more and more extensive.
As its name suggests, more controlled storage systems are that a storage device has multiple controllers to be controlled.The storage device
Usually disk array etc..Under normal circumstances, may proceed to mark off multiple data blocks in disk array, for data storage.
User can by client to controller send IO (input/output) instruct, controller according to the I/O instruction to storage
Equipment performs corresponding operation.
In the prior art, for more controlled storage systems, generally use ALUA patterns, i.e. Asymmetric Active/
Active patterns.By taking more controlled storage systems of dual controller as an example, for a specific data block, one of controller
Target port has the initiative/optimized (active/optimized) state, and the target port of another controller has the initiative/and it is non-
Optimize (active/unoptimized) state.Popular says, i.e., for some data block in storage device, can only pass through
Specific one controller is controlled to the data block.
But prior art, during specifically used, the time delay of read-write operation is more long under normal circumstances, entirely
The performance comparision of storage system is low.
The content of the invention
It is an object of the invention to provide a kind of method of more controlled storage system I O schedulings, it can effectively reduce system and carry out
Time delay used during read-write operation；Another object of the present invention is to provide a kind of more controlled storage systems, can effectively carry
The performance of high whole storage system.
In order to solve the above technical problems, the present invention provides a kind of method of more controlled storage system I O schedulings, methods described bag
Include：
Client obtains I/O instruction；
The client selects any one virtual route according to the I/O instruction from multiple virtual routes prestored
As destination virtual path, the virtual route is that multiple controllers map in storage device obtained by same data block respectively
Virtual route；
The client sends the I/O instruction to the target controller in the corresponding destination virtual path；
The target controller performs corresponding operation to the data block according to the I/O instruction.
Optionally, the client selects any one empty according to the I/O instruction from multiple virtual routes prestored
Intend path includes as destination virtual path：
The client is according to the multiple virtual routes of the I/O instruction poll, to select the destination virtual path.
Optionally, institute's client by the I/O instruction send to the corresponding destination virtual path target controller it
Afterwards, methods described also includes：
The target controller performs corresponding operation according to the I/O instruction to the caching pre-set；
The target controller performs corresponding operation to the data block according to the I/O instruction to be included：
When the caching meets the condition pre-set, the target controller is according to the I/O instruction to the data
Block performs corresponding operation.
Optionally, institute's client by the I/O instruction send to the corresponding destination virtual path target controller it
Afterwards, methods described also includes：
The target controller sends the I/O instruction to each controller.
Optionally, before the target controller performs corresponding operation according to the I/O instruction to the data block,
Methods described also includes：
The target controller obtains the distributed lock of the corresponding data block；
The target controller performs corresponding operation to the data block according to the I/O instruction to be included：
After the target controller gets the distributed lock, the target controller is according to the I/O instruction to institute
State data block and perform corresponding operation.
Present invention also offers a kind of more controlled storage systems, the system is set including client, multiple controllers and storage
Standby, the client is connected with each other with multiple controllers, and multiple controllers mutually interconnect with the storage device respectively
Connect；
Multiple controllers are stored with the client and map same data block institute in the storage device respectively
Obtained virtual route；The client is used to select any one empty from multiple virtual routes according to the I/O instruction of acquisition
Intend path as destination virtual path, and the I/O instruction is sent to the target controller in the corresponding destination virtual path；
The target controller is used to perform the data block corresponding operation according to the I/O instruction.
Optionally, the client is specifically used for the multiple virtual routes of poll, to select the destination virtual path.
Optionally, the target controller is used for：
Corresponding operation is performed to the caching pre-set according to the I/O instruction；
When the caching meets the condition pre-set, the target controller is according to the I/O instruction to the data
Block performs corresponding operation.
Optionally, the target controller is used for：
The I/O instruction is sent to each controller.
Optionally, the target controller is used for：
Obtain the distributed lock of the corresponding data block；
After the distributed lock is got, the data are write into the data block.
A kind of method of more controlled storage system I O schedulings provided by the present invention, client can be from corresponding same data
Any one virtual route is selected as destination virtual path in multiple virtual routes prestored of block, according to the virtual of selection
Path is different, the operation of corresponding I/O instruction can be performed to same data block by different controllers, hence for a spy
Fixed data block, have multiple controllers and be controlled by mulitpath.Cooperate work between multiple controllers, can be with
The operating pressure of controller is effectively reduced, reduces the time delay of read-write operation, improves the performance of whole storage system.
Present invention also offers a kind of more controlled storage systems, equally with above-mentioned beneficial effect, no longer repeated herein.
Brief description of the drawings
, below will be to embodiment or existing for the clearer explanation embodiment of the present invention or the technical scheme of prior art
The required accompanying drawing used is briefly described in technology description, it should be apparent that, drawings in the following description are only this hair
Some bright embodiments, for those of ordinary skill in the art, on the premise of not paying creative work, can be with root
Other accompanying drawings are obtained according to these accompanying drawings.
The flow chart for the first more controlled storage system I O scheduling method that Fig. 1 is provided by the embodiment of the present invention；
The flow chart for the I O scheduling method of controlled storage system more than second that Fig. 2 is provided by the embodiment of the present invention；
A kind of structured flowchart for more controlled storage systems that Fig. 3 is provided by the embodiment of the present invention.
Embodiment
The core of the present invention is to provide a kind of method of more controlled storage system I O schedulings.In the prior art, for more controls
Storage system, generally use ALUA patterns, i.e., for some data block in storage device, specific one control can only be passed through
Device processed is controlled to the data block.It is corresponding when a certain moment needing to carry out substantial amounts of read-write operation to specific data block
The controller of the data block just needs to be operated in high load condition, and to be typically only capable to serial execution each for controller
Read-write operation, when workload is very big inevitably resulting in read-write operation there is obvious delay.When connecting some controllers
When path is broken down, mistake occurs certainly when toward data storage in data block corresponding to the controller, and due to not having
There are other paths for connecting the data block, cause the time delay for completing the I/O instruction would generally be very long.
And a kind of method of more controlled storage system I O schedulings provided by the present invention, client can be from corresponding same numbers
According to any one virtual route is selected as destination virtual path in multiple virtual routes prestored of block, according to the void of selection
Intend path difference, the operation of corresponding I/O instruction can be performed to same data block by different controllers, hence for one
Specific data block, have multiple controllers and be controlled by mulitpath.Cooperate work between multiple controllers, can
Effectively to reduce the operating pressure of controller, the time delay of read-write operation is reduced, improves the performance of whole storage system.
In order that those skilled in the art more fully understand the present invention program, with reference to the accompanying drawings and detailed description
The present invention is described in further detail.Obviously, described embodiment is only part of the embodiment of the present invention, rather than
Whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art are not making creative work premise
Lower obtained every other embodiment, belongs to the scope of protection of the invention.
It refer to Fig. 1, the flow for the first more controlled storage system I O scheduling method that Fig. 1 is provided by the embodiment of the present invention
Figure, this method include：
S101：Client obtains I/O instruction.
The I/O instruction is broadly divided into two kinds, and one kind is to read data, and another kind is write-in data.In the embodiment of the present invention
In, user can be read by sending I/O instruction or write data, and now client and controller are needed according to the IO
Instruction performs corresponding operation.
S102：Client selects any one virtual route conduct according to I/O instruction from multiple virtual routes prestored
Destination virtual path.
In embodiments of the present invention, the virtual route is that multiple controllers map same data in storage device respectively
Virtual route obtained by block.
Above-mentioned client can be write data into data block by mulitpath or data are read from data block, its
In per paths a corresponding controller.In embodiments of the present invention, for same data block in storage device, in client
It is stored with a plurality of virtual route for eventually pointing to the data block.Under normal circumstances, corresponding number is all included in every virtual route
According to the UUID (general unique identifier) of block, UUID purpose is all elements allowed in distributed system, can be had unique
Recognize information.The client can identify which virtual route corresponds to same data block according to the UUID.Usual feelings
Under condition, client is that above-mentioned multiple virtual routes are managed using MPIO (multi-route management software), naturally it is also possible to logical
Cross other software to be managed above-mentioned multiple virtual routes, concrete implementation mode does not do specific limit in embodiments of the present invention
It is fixed.
The virtual route that client selects from a plurality of virtual route of corresponding same data block is as destination virtual road
Footpath, now client the controller corresponding with the destination virtual path can be found according to the destination virtual path, and finally
Data block is operated accordingly according to I/O instruction is obtained in S101.
Wherein, the mode of client selection target virtual route has many kinds, such as at random distributes the I/O instruction of acquisition
Into each virtual route etc..Certainly, the I/O instruction of acquisition is assigned in each virtual route in order to more balanced, this
Step S102 can be specially：Client is according to the multiple virtual routes of the I/O instruction poll, to select the target empty
Intend path.I.e. by way of poll by the balanced distribution of I/O instruction in each virtual route.It is, of course, also possible to pass through other
Mode distribute I/O instruction, such as I/O instruction is distributed in each virtual route by hash algorithm or other modes.
The relevant specific method of salary distribution is not specifically limited in embodiments of the present invention.
S103：Client sends I/O instruction to the target controller in corresponding destination virtual path.
In embodiments of the present invention, the target controller is the controller of the corresponding target virtual address, at this
It is responsible for performing data block corresponding operation according to the I/O instruction in flow.In addition, the target controller and other
Between each controller and indifference.
S104：Target controller performs corresponding operation to data block according to I/O instruction.
In this step, if the I/O instruction that target controller receives is reading data, the target controller will
Corresponding data are read from the data block, and the data of reading are sent to client；What if target controller received
I/O instruction is write-in data, then the target controller can write data into the data block.After aforesaid operations have been carried out,
The target controller would generally collect the result after the completion of operation, and the result is back into above-mentioned client.
Further, in embodiments of the present invention, can be described with the status information of synchronizing data blocks between multiple controllers
Status information includes the size of data block, free space, path status etc..Status information about the data block it is specific
Content, it is not specifically limited in embodiments of the present invention.The status information of synchronizing data blocks can ensure between multiple controllers
The status information of the same data block of correspondence stored in multiple controllers is consistent, no matter that is, client is led to from which paths
Cross which controller access data block, the status information of the obtained data block is all consistent, be not in status information not
The problem of consistent.Certainly, in embodiments of the present invention between multiple controllers can also unsynchronized data block status information,
It is not specifically limited in the embodiment of the present invention.
A kind of method of more controlled storage system I O schedulings provided by the present invention, client can be from corresponding same data
Any one virtual route is selected as destination virtual path in multiple virtual routes prestored of block, according to the virtual of selection
Path is different, the operation of corresponding I/O instruction can be performed to same data block by different controllers, hence for a spy
Fixed data block, have multiple controllers and be controlled by mulitpath.Cooperate work between multiple controllers, can be with
The operating pressure of controller is effectively reduced, reduces the time delay of read-write operation, improves the performance of whole storage system.
In the present invention, further can set in the controller caching with improve the readwrite performance of whole system and
Reduce delay.Detailed content will carry out expansion description in following inventive embodiments.
It refer to Fig. 2, the flow for the I O scheduling method of controlled storage system more than second that Fig. 2 is provided by the embodiment of the present invention
Figure, this method include：
S201：Client obtains I/O instruction.
S202：Client selects any one virtual route conduct according to I/O instruction from multiple virtual routes prestored
Destination virtual path.
S203：Client sends I/O instruction to the target controller in corresponding destination virtual path.
In embodiments of the present invention, S201, S202 and S203 respectively with S101, S102 and S103 in foregoing invention embodiment
Identical, detailed content refer to foregoing invention embodiment, no longer be repeated herein.
S204：Target controller performs corresponding operation according to I/O instruction to the caching pre-set.
In embodiments of the present invention, in order to further improve the readwrite performance of total system and reduce time delay,
Caching can be set in the controller.Caching is exactly the buffering area of data exchange, is obtained because the speed of service of caching is faster than internal memory
It is more, therefore the effect cached is just to aid in hardware and quickly run.Caching often uses RAM (the impermanent storages that power-off is fallen
Deposit), certainly in embodiments of the present invention, particular type for caching is simultaneously not specifically limited.
In this step, data and the write-in major class of data two, accordingly, target control are read because I/O instruction is generally divided into
The operation that device performs to caching can also be divided into two major classes, and one kind is that data are read from caching；Another kind of is from into caching
Write data.It will be described in detail respectively to reading data and write-in data below.
When reading data, the partial data that would generally be previously stored with caching in above-mentioned data block.It is specific relevant
The data stored in caching can be that the hot spot data taken is written infrequently in a period of time, can also be by pre-reading, from data
The data that client may be read are prestored into data in the buffer in advance in block.About the particular content pre-read, due to
The data stored within the data block are typically continuous, and user by client read data when it is generally also continuous,
So adjacent data can be prestored to when user reads certain one piece of data by client from data block
In caching, to increase the hit rate that the data stored in caching are read.
Certainly, the specific data about being stored in caching can also have other specific contents, or pass through its other party
Formula from the data write-in caching in data block, is not especially limited in embodiments of the present invention in advance.
When writing data, during target controller first can cache the data write-in for needing to write, due to being deposited in caching
The data of storage are in units of page, under normal circumstances when the containing dirty pages ratio in caching reaches the threshold value pre-set, such as dirty
Page ratio has reached 60%, it is possible to writes the data stored in caching in above-mentioned data block.
Certainly, the threshold value of above-mentioned containing dirty pages ratio can also be other values, be not specifically limited in embodiments of the present invention.
S205：Target controller sends I/O instruction to each controller.
In embodiments of the present invention, target controller sends I/O instruction to remaining each controller, can be at it
Disaster-tolerant backup is carried out in remaining controller.Such as when data are read, if not reading the number in the caching of target controller
According to, but data read from data block, while when target controller writes related data into caching by pre-reading, this
Step can allow remaining controller to write related data into caching again by pre-reading, if ensureing, next client is to pass through it
When data are read in his path, the same hit rate that can increase the data stored in caching and be read.
When writing data, the data can be write in the buffer in each controller, for disaster-tolerant backup.
This step can be after S204 by no sequencing, i.e. S205 in specific implementation process with above-mentioned S204
Perform, can also be performed before S204, can also simultaneously perform with S204, be not specifically limited in embodiments of the present invention.
S206：Obtain the distributed lock of corresponding data block.
Due to when multiple controllers operate to same data block simultaneously, can mutually be done between multiple controllers
Disturb., in embodiments of the present invention can be by setting the distributed lock corresponding to above-mentioned data block in order to prevent mutual interference
To cause mutual exclusion between each controller, prevent from disturbing between controller with this, to ensure the uniformity of data.
Under normal circumstances, after target controller gets the distributed lock, can just operation be performed to data block；
If target controller does not get the distributed lock, it can only wait and reacquire the distributed lock, until obtaining
To after the distributed lock, can just operation be performed to data block., it is necessary to discharge described after being completed to data block operation
Distributed lock.
In embodiments of the present invention, it is necessary to further typically after having performed S204 and having met default condition
It is to perform this step to obtain distributed lock that operation accordingly is performed to data block.Certainly, this step and S205 are specific real
It can be performed, can also be performed before S205 after S205 by no sequencing, i.e. S206 during applying, can also be with
S204 is performed simultaneously, is not specifically limited in embodiments of the present invention.But under normal circumstances, S206 needs to hold after S204
OK.
S207：Target controller performs corresponding operation to data block according to I/O instruction.
In this step, on the premise of the distributed lock is got in S206, when S204 reads data, if slow
There is no the data in depositing, then target controller can read the data from the data block.Further, target controller can
So that the related data associated with the data is written in caching；When S205 writes data, when the containing dirty pages ratio in caching
When reaching the threshold value pre-set, the data in the caching can be write above-mentioned data block by target controller.
The method of a kind of more controlled storage system I O schedulings provided by the present invention, by setting caching can be with the controller
Greatly improve the readwrite performance of total system and reduce time delay；And target controller I/O instruction can be sent to
Each controller is with to data progress disaster-tolerant backup；Also further prevent from occurring to do between controller by setting distributed lock
Disturb, ensure the uniformity of data.
A kind of more controlled storage systems provided in an embodiment of the present invention are introduced below, more control storage systems described below
System can be mutually to should refer to a kind of above-described more controlled storage system I O scheduling methods.
A kind of structured flowchart for more controlled storage systems that Fig. 3 is provided by the embodiment of the present invention, reference picture 2, more control storages
System can include：
Client 100, multiple controllers 200 and storage device 300, the client 100 and multiple controllers 200
It is connected with each other, multiple controllers 200 are connected with each other with the storage device 300 respectively.
Multiple controllers 200 are stored with the client and map same number in the storage device 300 respectively
According to the virtual route obtained by block 301；The client is used to be selected from multiple virtual routes according to the I/O instruction of acquisition
Any one virtual route is selected as destination virtual path, and the I/O instruction is sent to the mesh in the corresponding destination virtual path
Mark controller 201.
The target controller 201 is used to perform corresponding operation to the data block 301 according to the I/O instruction.
In embodiments of the present invention, the client 100 can be specifically used for：
The multiple virtual routes of poll, to select the destination virtual path.
In embodiments of the present invention, the target controller 201 can be used for：
Corresponding operation is performed to the caching 202 pre-set according to the I/O instruction；
When the caching 202 meets the condition pre-set, the target controller 201 is according to the I/O instruction to institute
State data block 301 and perform corresponding operation.
In embodiments of the present invention, the target controller 201 can be used for：
The I/O instruction is sent to each controller 200.
In embodiments of the present invention, the target controller 201 can be used for：
Obtain the distributed lock of the corresponding data block 301；
After the distributed lock is got, the data are write into the data block 301.
In embodiments of the present invention, under normal circumstances between client 100 and multiple controllers 200, SAN network phase is passed through
Connect.
A kind of more controlled storage systems of the present embodiment are used to realize a kind of foregoing more controlled storage system I O scheduling methods, because
The embodiment part of the visible more controlled storage system I O scheduling methods hereinbefore of embodiment in this more controlled storage system,
So its embodiment is referred to the description of corresponding various pieces embodiment, will not be repeated here.
Each embodiment is described by the way of progressive in this specification, what each embodiment stressed be with it is other
The difference of embodiment, between each embodiment same or similar part mutually referring to.For dress disclosed in embodiment
For putting, because it is corresponded to the method disclosed in Example, so description is fairly simple, related part is referring to method part
Explanation.
Professional further appreciates that, with reference to the unit of each example of the embodiments described herein description
And algorithm steps, can be realized with electronic hardware, computer software or the combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software, the composition and step of each example are generally described according to function in the above description.These
Function is performed with hardware or software mode actually, application-specific and design constraint depending on technical scheme.Specialty
Technical staff can realize described function using distinct methods to each specific application, but this realization should not
Think beyond the scope of this invention.
Directly it can be held with reference to the step of method or algorithm that the embodiments described herein describes with hardware, processor
Capable software module, or the two combination are implemented.Software module can be placed in random access memory (RAM), internal memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
The method to a kind of more controlled storage system I O schedulings provided by the present invention and more controlled storage systems are carried out above
It is discussed in detail.Specific case used herein is set forth to the principle and embodiment of the present invention, above example
Explanation be only intended to help understand the present invention method and its core concept.It should be pointed out that for the common of the art
For technical staff, under the premise without departing from the principles of the invention, some improvement and modification can also be carried out to the present invention, these
Improve and modification is also fallen into the protection domain of the claims in the present invention.
Claims (10)
- A kind of 1. method of more controlled storage system I O schedulings, it is characterised in that methods described includes：Client obtains I/O instruction；The client selects any one virtual route conduct according to the I/O instruction from multiple virtual routes prestored Destination virtual path, the virtual route map the void in storage device obtained by same data block for multiple controllers respectively Intend path；The client sends the I/O instruction to the target controller in the corresponding destination virtual path；The target controller performs corresponding operation to the data block according to the I/O instruction.
- 2. according to the method for claim 1, it is characterised in that the client is deposited in advance according to the I/O instruction from multiple Any one virtual route is selected to include as destination virtual path in the virtual route of storage：The client is according to the multiple virtual routes of the I/O instruction poll, to select the destination virtual path.
- 3. according to the method for claim 1, it is characterised in that send the I/O instruction to described in correspondence in institute's client After the target controller in destination virtual path, methods described also includes：The target controller performs corresponding operation according to the I/O instruction to the caching pre-set；The target controller performs corresponding operation to the data block according to the I/O instruction to be included：When the caching meets the condition pre-set, the target controller is held according to the I/O instruction to the data block The corresponding operation of row.
- 4. according to the method for claim 3, it is characterised in that send the I/O instruction to described in correspondence in institute's client After the target controller in destination virtual path, methods described also includes：The target controller sends the I/O instruction to each controller.
- 5. according to the method described in any one of Claims 1-4 claim, it is characterised in that in the target controller root Before performing corresponding operation to the data block according to the I/O instruction, methods described also includes：The target controller obtains the distributed lock of the corresponding data block；The target controller performs corresponding operation to the data block according to the I/O instruction to be included：After the target controller gets the distributed lock, the target controller is according to the I/O instruction to the number Corresponding operation is performed according to block.
- 6. a kind of more controlled storage systems, it is characterised in that the system includes client, multiple controllers and storage device, institute State client to be connected with each other with multiple controllers, multiple controllers are connected with each other with the storage device respectively；Multiple controllers are stored with the client to map respectively in the storage device obtained by same data block Virtual route；The client is used to select any one virtual road from multiple virtual routes according to the I/O instruction of acquisition Footpath is sent to the target controller in the corresponding destination virtual path as destination virtual path, and by the I/O instruction；The target controller is used to perform the data block corresponding operation according to the I/O instruction.
- 7. system according to claim 6, it is characterised in that the client is specifically used for the multiple virtual roads of poll Footpath, to select the destination virtual path.
- 8. system according to claim 6, it is characterised in that the target controller is used for：Corresponding operation is performed to the caching pre-set according to the I/O instruction；When the caching meets the condition pre-set, the target controller is held according to the I/O instruction to the data block The corresponding operation of row.
- 9. system according to claim 8, it is characterised in that the target controller is used for：The I/O instruction is sent to each controller.
- 10. according to the system described in any one of claim 6 to 9 claim, it is characterised in that the target controller is used In：Obtain the distributed lock of the corresponding data block；After the distributed lock is got, the data are write into the data block.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710883612.3A CN107632788A (en) | 2017-09-26 | 2017-09-26 | A kind of method of more controlled storage system I O schedulings and more controlled storage systems |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710883612.3A CN107632788A (en) | 2017-09-26 | 2017-09-26 | A kind of method of more controlled storage system I O schedulings and more controlled storage systems |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107632788A true CN107632788A (en) | 2018-01-26 |
Family
ID=61101790
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710883612.3A Pending CN107632788A (en) | 2017-09-26 | 2017-09-26 | A kind of method of more controlled storage system I O schedulings and more controlled storage systems |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107632788A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110618785A (en) * | 2018-12-24 | 2019-12-27 | 深圳创新科软件技术有限公司 | Dual-control storage system |
| CN112148206A (en) * | 2019-06-28 | 2020-12-29 | 北京金山云网络技术有限公司 | Data reading and writing method and device, electronic equipment and medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101430636A (en) * | 2007-11-07 | 2009-05-13 | 中兴通讯股份有限公司 | Multi-route management method in redundancy controller surroundings |
| CN101739220A (en) * | 2009-02-25 | 2010-06-16 | 浪潮电子信息产业股份有限公司 | Method for designing multi-controller memory array |
| CN101770346A (en) * | 2010-01-04 | 2010-07-07 | 浪潮电子信息产业股份有限公司 | Method for dispatching multi-controller disk array coordination |
| US20110029730A1 (en) * | 2009-07-31 | 2011-02-03 | Emc Corporation | Data processing system using cache-aware multipath distribution of storage commands among caching storage controllers |
| CN104461733A (en) * | 2014-11-24 | 2015-03-25 | 浪潮电子信息产业股份有限公司 | Parallel IO scheduling method between multiple controllers of high-end disk array |
| CN105430103A (en) * | 2015-12-31 | 2016-03-23 | 浪潮(北京)电子信息产业有限公司 | Dynamic load balancing system based on multi-controller storage |
| CN103135943B (en) * | 2013-02-21 | 2017-02-08 | 浪潮电子信息产业股份有限公司 | Self-adaptive IO (Input Output) scheduling method of multi-control storage system |
| CN106648851A (en) * | 2016-11-07 | 2017-05-10 | 郑州云海信息技术有限公司 | IO management method and device used in multi-controller storage |
| CN106657365A (en) * | 2016-12-30 | 2017-05-10 | 清华大学 | High concurrent data transmission method based on RDMA (Remote Direct Memory Access) |
| CN103324444B (en) * | 2013-05-24 | 2017-09-22 | 浪潮电子信息产业股份有限公司 | A kind of many control I O scheduling methods that host side is cooperateed with storage end |
-
2017
- 2017-09-26 CN CN201710883612.3A patent/CN107632788A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101430636A (en) * | 2007-11-07 | 2009-05-13 | 中兴通讯股份有限公司 | Multi-route management method in redundancy controller surroundings |
| CN101739220A (en) * | 2009-02-25 | 2010-06-16 | 浪潮电子信息产业股份有限公司 | Method for designing multi-controller memory array |
| US20110029730A1 (en) * | 2009-07-31 | 2011-02-03 | Emc Corporation | Data processing system using cache-aware multipath distribution of storage commands among caching storage controllers |
| CN101770346A (en) * | 2010-01-04 | 2010-07-07 | 浪潮电子信息产业股份有限公司 | Method for dispatching multi-controller disk array coordination |
| CN103135943B (en) * | 2013-02-21 | 2017-02-08 | 浪潮电子信息产业股份有限公司 | Self-adaptive IO (Input Output) scheduling method of multi-control storage system |
| CN103324444B (en) * | 2013-05-24 | 2017-09-22 | 浪潮电子信息产业股份有限公司 | A kind of many control I O scheduling methods that host side is cooperateed with storage end |
| CN104461733A (en) * | 2014-11-24 | 2015-03-25 | 浪潮电子信息产业股份有限公司 | Parallel IO scheduling method between multiple controllers of high-end disk array |
| CN105430103A (en) * | 2015-12-31 | 2016-03-23 | 浪潮(北京)电子信息产业有限公司 | Dynamic load balancing system based on multi-controller storage |
| CN106648851A (en) * | 2016-11-07 | 2017-05-10 | 郑州云海信息技术有限公司 | IO management method and device used in multi-controller storage |
| CN106657365A (en) * | 2016-12-30 | 2017-05-10 | 清华大学 | High concurrent data transmission method based on RDMA (Remote Direct Memory Access) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110618785A (en) * | 2018-12-24 | 2019-12-27 | 深圳创新科软件技术有限公司 | Dual-control storage system |
| CN112148206A (en) * | 2019-06-28 | 2020-12-29 | 北京金山云网络技术有限公司 | Data reading and writing method and device, electronic equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9292218B2 (en) | Method and apparatus to manage object based tier | |
| CN103049397B (en) | A kind of solid state hard disc inner buffer management method based on phase transition storage and system | |
| CN109144887B (en) | Memory system and control method for controlling nonvolatile memory | |
| CN104423894B (en) | Data memory device and method for controlling flash memory | |
| JP5315348B2 (en) | Method and apparatus for migration and cancellation of thin provisioning | |
| WO2017000658A1 (en) | Storage system, storage management device, storage device, hybrid storage device, and storage management method | |
| JP5384576B2 (en) | Selective use of multiple disparate solid-state storage locations | |
| US9753668B2 (en) | Method and apparatus to manage tier information | |
| US8639898B2 (en) | Storage apparatus and data copy method | |
| US20190079676A1 (en) | System and method for channel time management in solid state memory drives | |
| CN103384877A (en) | Storage system comprising flash memory, and storage control method | |
| CN107239526A (en) | File system implementation method, scrap cleaning method, operating position localization method | |
| WO2011072538A1 (en) | Method and apparatus for nand flash erase leveling | |
| CN103514249A (en) | Method and system for automatic data reduction and storage device | |
| CN106201335A (en) | Storage system | |
| CN104731713A (en) | Phase change memory abrasion balancing method and system based on random mapping | |
| CN109471843A (en) | A kind of metadata cache method, system and relevant apparatus | |
| CN107391050A (en) | A kind of data migration method, system, device and computer-readable recording medium | |
| CN109491605A (en) | A kind of date storage method based on COW, device and medium | |
| CN105138284A (en) | Virtual machine disk mirror image synchronization operation optimization system and method | |
| CN108132760A (en) | A kind of method and system for promoting SSD reading performances | |
| CN107632788A (en) | A kind of method of more controlled storage system I O schedulings and more controlled storage systems | |
| CN105389134A (en) | Flash memory interface control method and apparatus | |
| US9645946B2 (en) | Encryption for solid state drives (SSDs) | |
| CN102520879B (en) | Priority-based file information storage method, device and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180126 |