CN109683815A - A kind of double control disk array bedding storage method - Google Patents
A kind of double control disk array bedding storage method Download PDFInfo
- Publication number
- CN109683815A CN109683815A CN201811522597.0A CN201811522597A CN109683815A CN 109683815 A CN109683815 A CN 109683815A CN 201811522597 A CN201811522597 A CN 201811522597A CN 109683815 A CN109683815 A CN 109683815A
- Authority
- CN
- China
- Prior art keywords
- storage
- data
- bedding
- map information
- bedding storage
- 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
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0646—Horizontal data movement in storage systems, i.e. moving data in between storage devices or systems
- G06F3/0647—Migration mechanisms
-
- 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
- G06F3/0653—Monitoring storage devices or 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
- G06F3/0664—Virtualisation aspects at device level, e.g. emulation of a storage device or system
-
- 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
- G06F3/0665—Virtualisation aspects at area level, e.g. provisioning of virtual or logical volumes
Abstract
The present invention relates to a kind of double control disk array bedding storage methods, are related to computer memory technical field.The present invention provides a kind of bedding storage method of double storage controllers, after realizing that a storage control fails, storage service is taken over by another controller, i.e., bedding storage can still continue to work in another controller, and guarantee that data storage service is not interrupted.Bedding storage configuration and data statistics real-time synchronization in two storage controls, are in mirrored state.When a storage control, which breaks down, to fail, another controller can be in the storage service on the fault point seamless pipe controller.The efficient performance of bedding storage had both been utilized in the present invention, in turn ensured the high reliability of bedding storage.
Description
Technical field
The present invention relates to computer memory technical fields, and in particular to a kind of double control disk array bedding storage method.
Background technique
With solid state storage technologies (SSD), Serial Attached SCSI (SAS) (SAS) technology being constantly progressive and popularizing, storage medium
Type it is more various, there is very big difference using the performance of the storage equipment of different storage mediums and interface.SSD compared to
Traditional mechanical hard disk, due to the mechanical rotation of not no disk, addressing speed is greatly improved, and is especially answered random read-write is more
With under environment, performance can be greatly promoted, but the unit storage price of these high-performance storage equipment relative to traditional tape and
SATA hard disc is also higher by very much.
It is all storage plant commercial cities the problem of paying close attention to so how to store equipment using these to efficient balance.Using
S bedding storage, by data according to cold and hot carry out AUTOMATIC ZONING, hotter data are stored in the higher equipment of access performance, and
Colder data are stored in the lower equipment of access performance, not only available high performance, but also can be with effectively save cost.
Bedding storage technology has been widely used in the storage equipment of enterprise-level, such as the Easy Tier of IBM, EMC's
FAST etc., but these functions are all integrated in inside storage equipment, need the support of storage device firmware.Point realized on host
Layer storage scheme is also fewer.
Realize bedding storage, it is necessary first to virtualize the storage equipment (DEV1, DEV2) with different access performance
At a new storage equipment (VDEV).It is different from caching (Cache), the memory capacity of VDEV be DEV1 and DEV2 capacity it
(need to remove some for storing the space of metadata).The I/O request of all users will issue new storage equipment VDEV,
Then corresponding physical equipment DEV1 or DEV2 is forwarded to according still further to certain address mapping relation.It also needs to count I/ simultaneously
The temperature of O, and according to these temperature data, dynamically in different storage equipment room migrating datas, to reach performance capability
Optimization.Logical volume management LVM provides Storage Virtualization method, multiple physical volume PV can be built up to a volume group VG, then
Virtual volume LV is created in VG.And LVM provides the API of the migrating data between different physical volumes.Therefore, based on LVM's
Bedding storage scheme is by the virtualization of LVM and the ability of Data Migration.
Bedding storage scheme based on LVM is by the virtualization of LVM and the ability of Data Migration, it can be achieved that in single controller
The bedding storage of storage system.But in the storage system of double storage controllers, if the configuration information and statistics of bedding storage
Temperature information is only stored in main controller, once master controller breaks down, storage service is switched to from controller, and counts temperature
Information is not updated to from controller, individual-layer data can not be reorganized from controller, to cause loss of data.Cause
This, to realize that the bedding storage of dual controller will guarantee that the real-time synchronization for counting temperature information updates.
Summary of the invention
(1) technical problems to be solved
The technical problem to be solved by the present invention is how to provide a kind of bedding storage method of double storage controllers, realize
After one storage control failure, the storage operation of data is taken over by another controller, i.e., tiered storage system still may be used
To continue to work in another controller, and guarantee that data storage service is not interrupted, to improve the reliability of storage array, prolongs
Long products service life.
(2) technical solution
In order to solve the above-mentioned technical problems, the present invention provides a kind of double control disk array bedding storage methods, including with
Lower step:
Step 1: the virtualization of storage equipment: carrying out the creation of virtual unit, delete, and safeguard virtual unit to physics
The address mapping relation of equipment；
It wherein, the use of all physical hard disks include SSD and HDD creation into physical volume PV when creation virtual unit；Then
All PV are created into a volume group VG；Finally virtual volume LV is built on being created that the VG come；
Wherein, successively the virtual volume LV of creation, volume group VG, physical volume PV are deleted when deleting virtual unit；
Virtual unit is wherein established into configuration mapping table with corresponding SSD, HDD when maintenance configuration mapping table, and at two
It is synchronous in storage control A, B to save；
Step 2: the monitoring and statistics of I/O performance: the temperature of statistics I/O data and size, the random performance of I/O data
Attribute provides foundation for follow-up data migration, and the hot statistics of I/O data are the various situations for obtaining I/O request queue, packet
Include: the process title that is written and read, executes time, the physical block number of read-write, block size at process number；Pass through statistics I/O data
Distribution and read-write frequency obtain the temperature information of I/O data, and the temperature by these hot statistics information preservations in database is united
It counts in information table；
Step 3: the migration of data: according to these statistical attributes of the temperature of I/O data, the high data of visiting frequency being moved
The high physical equipment storage of readwrite performance is moved to, and the low Data Migration of visiting frequency is stored to the low physical equipment of performance；
Wherein, Data Migration is carried out belonging between the different PV in the same VG, dsc data and cold data distribution are stored in difference
Physical volume on；
Step 4: configuration map information is synchronous with data statistics: configuration map information is to tiered storage system
Some configurations, in the period including bedding storage finger daemon, the path of bedding storage equipment, bedding storage finger daemon opens
Dynamic completed by reading configuration map information, and after having modified configuration map information, new configuration map information is led to
Know bedding storage finger daemon, configuration map information is stored in database configuration map information table, whenever storage control A
After changing configuration information, storage control B configures map information also by real-time synchronization, whenever the hot statistics of storage control A
Information changes, the hot statistics information of storage control B also real-time synchronization.
Preferably, during step 1 is to four, when storage control A breaks down, the bedding storage of storage control B
Finger daemon initializes dependency structure body, opens by carrying out analysis reading to configuration map information table and hot statistics information table
Dynamic migrating processes, bedding storage service continue, and realize the failover of bedding storage.
Preferably, virtual unit is created in step 1 to realize using the virtualization tool that LVM is provided.
Preferably, virtual unit is deleted in step 1 to realize using the virtualization tool that LVM is provided.
Preferably, the migration of data is realized using the physical volume Migration tools that LVM is provided in step 3.
(3) beneficial effect
The present invention provides a kind of bedding storage method of double storage controllers, after realizing a storage control failure,
Storage service is taken over by another controller, i.e., bedding storage can still continue to work in another controller, and guarantee number
It is not interrupted according to storage service.Bedding storage configuration and data statistics real-time synchronization in two storage controls, are in mirror
As state.When a storage control, which breaks down, to fail, another controller can be in the control of fault point seamless pipe
Storage service on device.The efficient performance of bedding storage had both been utilized in the present invention, in turn ensured the high reliability of bedding storage.It should
Method can be applied in double storage controllers mixing storage array product, the business scenario used simultaneously for SSD and HDD.?
On the basis of reducing the total cost of ownership, while ensure that reliability, and the overall performance of product is improved.
Detailed description of the invention
Fig. 1 is the bedding storage schematic diagram of the invention based on LVM；
Fig. 2 is dual controller bedding storage flow chart of the invention.
Specific embodiment
To keep the purpose of the present invention, content and advantage clearer, with reference to the accompanying drawings and examples, to of the invention
Specific embodiment is described in further detail.
As shown in Figure 1 and Figure 2, the dual controller bedding storage method proposed by the present invention based on LVM the following steps are included:
Step 1: the virtualization of storage equipment: being responsible for the creation of virtual unit, delete；Maintenance virtual unit is set to physics
Standby address mapping relation.
It creates virtual unit: being realized using the virtualization tool that LVM is provided.Using all physical hard disks include SSD and
HDD is created into physical volume PV；Then all PV are created into a volume group VG；Finally virtual volume is built on being created that the VG come
LV。
Delete virtual unit: the virtualization tool provided using LVM, successively by the virtual volume LV of creation, volume group VG, physics
PV is rolled up to delete.
Maintenance configuration mapping table: virtual unit is established into configuration mapping table with corresponding SSD, HDD, and is controlled in two storages
It is synchronous in device A, B processed to save.
Step 2: the monitoring and statistics of I/O performance: the temperature of statistics I/O data and size, the random performance of I/O data
Attribute provides foundation for follow-up data migration.The hot statistics of I/O data are the various situations for obtaining I/O request queue, packet
It includes: process title, process number, execution time, the physical block number of read-write, block size for being written and read etc.；By counting I/O data
Distribution and read-write frequency obtain the temperature information of I/O data, by these hot statistics information preservations database temperature
In Statistics table；
Step 3: the migration of data: according to statistical attributes such as the temperatures of I/O data, by the high Data Migration of visiting frequency
The physical equipment storage high to readwrite performance, and the low Data Migration of visiting frequency to the low physical equipment of performance is stored, it visits
Ask that frequency threshold is preset value；Wherein, the physical volume Migration tools provided using LVM, can belong in the same VG not
With Data Migration is carried out between PV, dsc data and cold data distribution are stored on different physical volumes；
Step 4: configuration map information is synchronous with data statistics: configuration map information is to tiered storage system
Some configurations, the period including bedding storage finger daemon, the path etc. of bedding storage equipment, bedding storage finger daemon
Starting is completed by reading configuration map information, after having modified configuration map information, by new configuration map information
Notify bedding storage finger daemon, configuration map information is stored in database configuration map information table, whenever storage controls
After device A changes configuration information, storage control B configures map information also by real-time synchronization, whenever the temperature of storage control A is united
Meter information changes, and the hot statistics information of storage control B also can real-time synchronization；
During step 1 is to four, when storage control A breaks down, the bedding storage of storage control B guard into
Journey initializes dependency structure body, starting migration by carrying out analysis reading to configuration map information table and hot statistics information table
Process, bedding storage service continue, and realize the failover of bedding storage.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, without departing from the technical principles of the invention, several improvement and deformations can also be made, these improvement and deformations
Also it should be regarded as protection scope of the present invention.
Claims (5)
1. a kind of double control disk array bedding storage method, which comprises the following steps:
Step 1: the virtualization of storage equipment: carrying out the creation of virtual unit, delete, and safeguard virtual unit to physical equipment
Address mapping relation；
It wherein, the use of all physical hard disks include SSD and HDD creation into physical volume PV when creation virtual unit；Then by institute
Some PV creations are at a volume group VG；Finally virtual volume LV is built on being created that the VG come；
Wherein, successively the virtual volume LV of creation, volume group VG, physical volume PV are deleted when deleting virtual unit；
Virtual unit is wherein established into configuration mapping table with corresponding SSD, HDD when maintenance configuration mapping table, and is stored at two
It is synchronous in controller A, B to save；
Step 2: the monitoring and statistics of I/O performance: the temperature of statistics I/O data and size, the random attribute of performance of I/O data,
Foundation is provided for follow-up data migration, the hot statistics of I/O data are the various situations for obtaining I/O request queue, comprising: are carried out
The process title of read-write, executes time, the physical block number of read-write, block size at process number；By count I/O data distribution with
And read-write frequency obtains the temperature information of I/O data, by these hot statistics information preservations database hot statistics information
In table；
Step 3: the migration of data: according to these statistical attributes of the temperature of I/O data, extremely by the high Data Migration of visiting frequency
The high physical equipment storage of readwrite performance, and the low Data Migration of visiting frequency is stored to the low physical equipment of performance；Wherein,
Data Migration is carried out belonging between the different PV in the same VG, dsc data and cold data distribution are stored in different physics
On volume；
Step 4: configuration map information is synchronous with data statistics: configuration map information is one to tiered storage system
A little configurations, the starting in the period including bedding storage finger daemon, the path of bedding storage equipment, bedding storage finger daemon are
It is completed by reading configuration map information, after having modified configuration map information, new configuration map information notice is arrived
Bedding storage finger daemon, configuration map information is stored in database configuration map information table, whenever storage control A is changed
After configuration information, storage control B configures map information also by real-time synchronization, whenever the hot statistics information of storage control A
It changes, the hot statistics information of storage control B also real-time synchronization.
2. the method as described in claim 1, which is characterized in that during step 1 is to four, occur as storage control A therefore
Barrier, the bedding storage finger daemon of storage control B is by analyzing configuration map information table and hot statistics information table
It reads, initializes dependency structure body, start migrating processes, bedding storage service continues, and realizes that the failure of bedding storage is cut
It changes.
3. the method as described in claim 1, which is characterized in that create virtual unit in step 1 and use the virtual of LVM offer
Chemical industry tool is realized.
4. the method as described in claim 1, which is characterized in that delete virtual unit in step 1 and use the virtual of LVM offer
Chemical industry tool is realized.
5. method according to any one of claims 1 to 4, which is characterized in that the migration of data is mentioned using LVM in step 3
The physical volume Migration tools of confession are realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811522597.0A CN109683815B (en) | 2018-12-13 | 2018-12-13 | Double-control disk array layered storage method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811522597.0A CN109683815B (en) | 2018-12-13 | 2018-12-13 | Double-control disk array layered storage method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109683815A true CN109683815A (en) | 2019-04-26 |
| CN109683815B CN109683815B (en) | 2021-03-23 |
Family
ID=66186729
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811522597.0A Active CN109683815B (en) | 2018-12-13 | 2018-12-13 | Double-control disk array layered storage method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109683815B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112463438A (en) * | 2020-11-06 | 2021-03-09 | 苏州浪潮智能科技有限公司 | Configuration information generation method and device of storage product and readable storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102023813A (en) * | 2009-09-22 | 2011-04-20 | 株式会社日立制作所 | Application and tier configuration management in dynamic page realloction storage system |
| CN103106050A (en) * | 2013-02-22 | 2013-05-15 | 浪潮电子信息产业股份有限公司 | Method for achieving layered storage and copy of data of storage system |
| CN106462510A (en) * | 2014-03-06 | 2017-02-22 | 伊姆西公司 | Multiprocessor system with independent direct access to bulk solid state memory resources |
| CN106844052A (en) * | 2017-01-22 | 2017-06-13 | 郑州云海信息技术有限公司 | A kind of method and device that fusion cluster is built based on Windows Server |
| CN107219998A (en) * | 2016-03-21 | 2017-09-29 | 爱思开海力士有限公司 | Data storage device and its operating method |
| CN108459978A (en) * | 2017-02-22 | 2018-08-28 | 爱思开海力士有限公司 | Data storage device including non-volatile memory device and its operating method |
| US10126971B1 (en) * | 2017-06-21 | 2018-11-13 | International Business Machines Corporation | Enhanced application performance in multi-tier storage environments |
-
2018
- 2018-12-13 CN CN201811522597.0A patent/CN109683815B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102023813A (en) * | 2009-09-22 | 2011-04-20 | 株式会社日立制作所 | Application and tier configuration management in dynamic page realloction storage system |
| CN103106050A (en) * | 2013-02-22 | 2013-05-15 | 浪潮电子信息产业股份有限公司 | Method for achieving layered storage and copy of data of storage system |
| CN106462510A (en) * | 2014-03-06 | 2017-02-22 | 伊姆西公司 | Multiprocessor system with independent direct access to bulk solid state memory resources |
| CN107219998A (en) * | 2016-03-21 | 2017-09-29 | 爱思开海力士有限公司 | Data storage device and its operating method |
| CN106844052A (en) * | 2017-01-22 | 2017-06-13 | 郑州云海信息技术有限公司 | A kind of method and device that fusion cluster is built based on Windows Server |
| CN108459978A (en) * | 2017-02-22 | 2018-08-28 | 爱思开海力士有限公司 | Data storage device including non-volatile memory device and its operating method |
| US10126971B1 (en) * | 2017-06-21 | 2018-11-13 | International Business Machines Corporation | Enhanced application performance in multi-tier storage environments |
Non-Patent Citations (2)
| Title |
|---|
| SHINICHI HAYASHI，ET AL.: "Data Location Optimization for Improvement of Tiered Storage Performance", 《ELECTRONICS AND COMMUNICATIONS IN JAPAN》 * |
| 董伟，等.: "自动分层存储技术在现代存储中的应用分析与前景展望", 《云南大学学报<自然科学版>》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112463438A (en) * | 2020-11-06 | 2021-03-09 | 苏州浪潮智能科技有限公司 | Configuration information generation method and device of storage product and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109683815B (en) | 2021-03-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8909887B1 (en) | Selective defragmentation based on IO hot spots | |
| US9229653B2 (en) | Write spike performance enhancement in hybrid storage systems | |
| US10176212B1 (en) | Top level tier management | |
| US10031703B1 (en) | Extent-based tiering for virtual storage using full LUNs | |
| JP5944587B2 (en) | Computer system and control method | |
| US9459809B1 (en) | Optimizing data location in data storage arrays | |
| US9477431B1 (en) | Managing storage space of storage tiers | |
| JP5079841B2 (en) | Method and storage apparatus for controlling data write to virtual logical volume according to Thin Provisioning | |
| JP5456063B2 (en) | Method and system for dynamic storage tiering using allocate-on-write snapshots | |
| CN110268391B (en) | System and method for caching data | |
| US8966143B2 (en) | Method for controlling storages and storage control apparatus | |
| US8364858B1 (en) | Normalizing capacity utilization within virtual storage pools | |
| JP4235220B2 (en) | Computer system and data migration method | |
| US8654472B2 (en) | Implementing enhanced fragmented stream handling in a shingled disk drive | |
| US9330009B1 (en) | Managing data storage | |
| WO2013157032A1 (en) | Storage subsystem and data management method of storage subsystem | |
| JP2006293981A (en) | Database storing method, and database storing system | |
| US11042324B2 (en) | Managing a raid group that uses storage devices of different types that provide different data storage characteristics | |
| US20140095789A1 (en) | Management of data using inheritable attributes | |
| US10481820B1 (en) | Managing data in storage systems | |
| US11461287B2 (en) | Managing a file system within multiple LUNS while different LUN level policies are applied to the LUNS | |
| CN103916459A (en) | Big data filing and storing system | |
| CN106326384A (en) | File storage method suitable for high-speed mass storage based on FPGA (Field Programmable Gate Array) | |
| CN110989924B (en) | Metadata storage performance optimization method and storage server | |
| US8713257B2 (en) | Method and system for shared high speed cache in SAS switches |
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 |