CN105808157A - Storage architecture creating method, storage access method and storage system - Google Patents
Storage architecture creating method, storage access method and storage system Download PDFInfo
- Publication number
- CN105808157A CN105808157A CN201410854717.2A CN201410854717A CN105808157A CN 105808157 A CN105808157 A CN 105808157A CN 201410854717 A CN201410854717 A CN 201410854717A CN 105808157 A CN105808157 A CN 105808157A
- Authority
- CN
- China
- Prior art keywords
- physical disks
- disk
- virtual disk
- mapping relations
- 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.)
- Withdrawn
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
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)
Abstract
The present invention discloses a storage architecture creating method, a storage access method and a storage system. The storage architecture creating method disclosed by the present invention specifically comprises: creating a logic disks according to a physical disk, and establishing a mapping relationship of logic disks to physical disks; selecting a logic disk to create a virtual disk, wherein the virtual disk comprises at least one logic disk; creating a storage pool according to the virtual disk, wherein the storage pool comprises at least one virtual disk; and creating a volume, and establishing a mapping relationship of volumes to virtual disks. The creating method disclosed by the present invention is capable of solving the technical problem that metadata existing in the existing virtualization storage system is excessively large in data volume, complex to be managed, and can not be compatible with the conventional storage system.
Description
Technical field
The present invention relates to Computer Storage field, particularly relate to the creation method of a kind of storage architecture, storage access method and storage system.
Background technology
Along with the continuous expansion of user data, the hard disk of system management memory gets more and more.Configuring storage in traditional RAID array mode, configuration process is loaded down with trivial details.After initial configuration completes, tradition RAID array is relatively larger challenge by the dilatation in later stage and business reorganization.Virtualized pool can simplify the configuration of user, can flexible expansion, significantly reduce the maintenance workload of user.
The growth of capacity is far unable to catch up with in the growth of the access rate of hard disk, and from 21 years of 1991 to 2012, HDD hard-disk capacity brings up to 4TB from 40MB, improves 100,000 times, and speed has brought up to 128MB/s from 0.7MB/s, only improves 182 times.This directly results in when hard disk breaks down, and the time that reconstruct spends is more and more longer.Hard disk just looks like a pregnant slot water bottle, will fill water institute's time spent more and more longer.Virtualization can alleviate this problem.The access speed of one hard disk is slow, it is possible to allow the simultaneously concurrent working of multiple hard disk, improves access speed, reduces reconstitution time.
Limitation due to the heritage storage system based on RAID array, having occurred in that the storage system based on virtualization Pool on market, but existing virtual storage system exists some problems, the data volume of metadata is excessive, complex management, it is impossible to compatible heritage storage system etc.
Summary of the invention
The main technical problem to be solved in the present invention is, there is provided the creation method of a kind of storage architecture, storage access method and storage system, it is possible to the data volume solving the metadata that existing virtual storage system exists is excessive, complex management and can not the technical problem of compatible heritage storage system.
For solving above-mentioned technical problem, the present invention provides the creation method of a kind of storage architecture, comprises the steps:
Create Logical Disk according to physical disks, and set up the Logical Disk mapping relations to physical disks；
Selecting Logical Disk to create virtual disk, described virtual disk is made up of at least one Logical Disk；
Creating storage pool according to virtual disk, described storage pool is made up of at least one virtual disk；
Create volume and set up the mapping relations being rolled onto virtual disk.
Further, described create Logical Disk according to physical disks, and set up Logical Disk and include to the step of the mapping relations of physical disks:
At least one physical disks being divided into multiple pieces, selects block to create Logical Disk, described Logical Disk is made up of the block at least one physical disks.
Further, described at least one physical disks is divided into multiple pieces, selects the step that block creates Logical Disk to include:
Physical disks at least one physical disks set is divided into multiple pieces, and described physical disks set is made up of multiple physical disks；
Block composition Logical Disk is selected from a described physical disks set.
Further, the described step creating the mapping relations rolled up and be rolled onto virtual disk includes:
Create volume, described volume is divided into multiple CK；
Set up in volume CK to the mapping relations of virtual disk.
Further, the described step according to virtual disk establishment storage pool includes:
Selecting virtual disk to create accumulation layer, described accumulation layer is made up of at least one virtual disk；
Selecting accumulation layer to create storage pool, described storage pool is made up of accumulation layer at least one described.
It is similarly the technical problem that solution is above-mentioned, present invention also offers a kind of storage access method, based on the storage architecture that method described above creates, comprise the steps:
After cigarette receives access instruction, according to the described mapping relations being rolled onto virtual disk, described access instruction is sent to the virtual disk of correspondence；
After described virtual disk receives described access instruction, carry out RAID and calculate the Logical Disk determining correspondence, according to the mapping relations of the Logical Disk determined to physical disks, described access instruction is sent to corresponding physical disks and conducts interviews.
Further, when in described storage architecture, Logical Disk is made up of the block at least one physical disks, and Logical Disk when including the mapping relations of Logical Disk block to physical disks to the mapping relations of physical disks, described access instruction is sent to the step that the physical disks of correspondence conducts interviews and includes by the described mapping relations according to the Logical Disk determined to physical disks:
Conduct interviews according to the block that described access instruction is sent in the physical disks of correspondence by the mapping relations of block in the Logical Disk determined to physical disks.
Further, when in described storage architecture, volume is divided into multiple CK, and the mapping relations being rolled onto virtual disk are when including that in volume, CK is to the mapping relations of virtual disk, it are rolled onto the mapping relations of virtual disk described in described basis and described access instruction is sent to the step of corresponding virtual disk includes:
Described access instruction is sent to the virtual disk of correspondence by the mapping relations according to CK in described volume to virtual disk.
It is similarly the technical problem that solution is above-mentioned, present invention also offers a kind of storage system, the storage architecture that described storage system creates based on method as above, including: volume processing module and virtual disk processing module；
Described volume processing module, for after cigarette receives access instruction, is sent to the virtual disk of correspondence according to the described mapping relations being rolled onto virtual disk by described access instruction；
Described virtual disk processing module, after receiving described access instruction at described virtual disk, carry out RAID and calculate the Logical Disk determining correspondence, according to the mapping relations of the Logical Disk determined to physical disks, described access instruction is sent to corresponding physical disks and conducts interviews.
Further, described virtual disk processing module, for when in described storage architecture Logical Disk be made up of the block at least one physical disks, and Logical Disk when including the mapping relations of Logical Disk block to physical disks to the mapping relations of physical disks, conduct interviews according to the block that described access instruction is sent in the physical disks of correspondence by the mapping relations of block in the Logical Disk determined to physical disks.
Further, described volume processing module, for when in described storage architecture volume be divided into multiple CK, and when the mapping relations being rolled onto virtual disk include CK in volume to the mapping relations of virtual disk, according to the mapping relations of CK in described volume to virtual disk, described access instruction is sent to corresponding virtual disk.
The invention has the beneficial effects as follows:
The invention provides the creation method of a kind of storage architecture, storage access method and storage system；The creation method of the storage architecture of the present invention specifically includes: creates Logical Disk according to physical disks, and sets up the Logical Disk mapping relations to physical disks；Selecting Logical Disk to create virtual disk, described virtual disk is made up of at least one Logical Disk；Creating storage pool according to virtual disk, described storage pool is made up of at least one virtual disk；Create volume and set up the mapping relations being rolled onto virtual disk；The storage architecture applying creation method of the present invention establishment so that storage system adopts two-layer to map, can decrease space and the access frequency of metadata；If needing the storage architecture that application the inventive method creates can reuse RAID processing module for heritage storage system, therefore the storage system that the storage architecture of creation method of the present invention establishment is compatible traditional, it addition, the storage architecture that creation method of the present invention creates also has the advantage that management is simple, be easy to configuration.
Accompanying drawing explanation
The schematic flow sheet of the creation method of a kind of storage architecture that Fig. 1 provides for the embodiment of the present invention one；
A kind of storage architecture schematic diagram that Fig. 2 provides for the embodiment of the present invention one；
A kind of mapping schematic diagram being rolled onto virtual disk that Fig. 3 provides for the embodiment of the present invention one；
A kind of Logical Disk that Fig. 4 provides for the embodiment of the present invention one is to the mapping schematic diagram of physical disks；
Relation schematic diagram between a kind of physical disks and storage pool that Fig. 5 provides for the embodiment of the present invention one；
A kind of schematic flow sheet storing access method that Fig. 6 provides for the embodiment of the present invention two；
The schematic flow sheet that a kind of I/O that Fig. 7 provides for the embodiment of the present invention two accesses；
A kind of structural representation storing system that Fig. 8 provides for the embodiment of the present invention three.
Detailed description of the invention
The present invention is described in further detail in conjunction with accompanying drawing below by detailed description of the invention.
Embodiment one:
The data volume considering the metadata that existing virtual storage system exists is excessive, complex management and can not the compatible technical problem embodied based on the conventional store of RAID array, present embodiments provide a kind of storage method, as it is shown in figure 1, include as follows in step:
Step 101: create Logical Disk according to physical disks, and set up the Logical Disk mapping relations to physical disks.
Specifically, physical disks is first selected, for instance multiple physical disks can be selected；Then the Logical Disk (LogicDisk) that physical disks is corresponding is created；Logical Disk and physical disks are mapped, sets up the Logical Disk mapping relations to physical disks, specifically, it is possible to set up the Logical Disk mapping relations table to physical disks.In storage accesses, the physical disks corresponding with Logical Disk can be gone out according to the mapping relationship searching of Logical Disk to physical disks, then access instruction is sent to this physical disks conduct interviews, such as can pass through the mapping relations table acquisition physical disks corresponding to Logical Disk searching Logical Disk to physical disks, then this physical disks be conducted interviews.
Preferably, at least one physical disks can be divided into multiple pieces (Block) by the present embodiment, for instance by big blocks such as physical disks divisions；Then selecting Block to form Logical Disk from one or more physical disks, namely a Logical Disk is made up of the Block in one or more physical disks, and the Block in Logical Disk can come from same physical disks, it is also possible to from different physical disks.At this point it is possible to the block in Logical Disk and physical disks are mapped, set up Logical Disk to physical disks the mapping relations of block, specifically can set up the mapping relations table of Logical Disk block to physical disks.Further, the Logical Disk under same virtual disk, the BLK of identical address, it is impossible to select same physical disks.This is for RAID redundancy.When a physical disks fault, it is impossible to cause multiple BLK faults of same address.
In access process, when access instruction (such as I/O) arrives LogicDisk,, then according to look for Logical Disk to physical disks the mapping relations of block, the access instruction (I/O) of LogicDisk is transmitted to block in corresponding physical disks and processes.LogicDisk is main in order that quick reconfiguration when realizing dish fault to mapping of physical disks in the present embodiment.
In the present embodiment this layer of LogicDisk table look-up after the conversion of I/O address.LogicDisk is main in order that quick reconfiguration when realizing dish fault to mapping of physical disks.Due to this layer of Random Maps of LogicDisk to physical disks, when a physical disks fault, and the Block in the physical disks of fault has the Block of RAID relation to be dispersed on multiple hard disk.During physical disks fault, Block fault corresponding for LogicDisk, in LogicDisk Random select system, the Block of other normal condition remaps.And start reconstruct.So from system level, a reconstruct during physical disks fault has reformed into multiple physical disks being reconstructed to multiple physical disks.Substantially increase the speed of reconstruct.
Preferably, in order to physical disks to be divided into different failure domains so that not influencing each other between the physical disks in different faults territory, the present embodiment is all right: physical disks is divided into physical disks set, and described physical disks set is made up of multiple physical disks；The Logical Disk of a same virtual disk physical disks set from which selects block composition.It is to say, first physical disks is divided, multiple physical disks are divided at least one PDG (physicdiskgroup, physical disks set)；Then physical disks in PDG is divided into multiple BLK, then selects the block composition Logical Disk in a PDG, set up Logical Disk to physical disks in PDG the mapping relations of block.
Step 102: selecting Logical Disk to create virtual disk, described virtual disk is made up of at least one Logical Disk.
The present embodiment is after creating Logical Disk (VD), one or more Logical Disk can be chosen to form virtual disk, such as, when needing to create 3 virtual disks, Logical Disk first virtual disk of composition can be chosen, select 2 Logical Disk second virtual disks of composition, select 5 Logical Disk to form the 3rd virtual disk.
LogicDisk is the RAID member disk of virtual disk in the present embodiment, and virtual disk carries out RAID calculating using Logical Disk LogicDisk as member disk, can pass through virtual disk when accessing and determine Logical Disk by RAID calculating；Therefore in same virtual disk, the Block being on the same address of LogicDisk can not point to same physical disks, and otherwise the member disk of virtual disk just loses RAID protective capability in same physical disks.
Step 103: creating storage pool according to virtual disk, described storage pool is made up of at least one virtual disk.
In the present embodiment, storage pool Pool is a disk storage resource virtualizing management unit, and during establishment, disk is joined pool management by user, and Pool comprises all physical disks in Pool and all object logics set up on physical disk.
Specifically, this step may include that selection virtual disk creates accumulation layer (Tier), and described accumulation layer is made up of at least one virtual disk；Selecting accumulation layer to create storage pool, described storage pool is made up of accumulation layer at least one described.
In the present embodiment, Tier layer is the concept in classification storage, and a Tier is exactly a level of classification storage, represents the combination of the hard disk of a kind of special speed.When supporting classification storage, a usual pool can create three Tier: high performance SSD layer is Tier0, and the SAS layer of medium-performance is Tier1, and jumbo SATA layer is Tier2.
The RAID group that virtual disk is made up of multiple LogicDisk, virtual disk primary responsibility RAID calculates.Hard disk in same Tier has identical speed.Tier and virtual disk logically belong to same layer.Tier and virtual disk are to comprise and involved relation.The RAID that one Tier is created by multiple same type dishes forms.
Management can create Tier and RAID on path.But on I/O path, the access instruction (such as I/O) that Volume issues directly is issued on virtual disk, after doing RAID calculating, it is handed down to LogicDisk.Tier is on I/O path and is absent from.Therefore, the present embodiment provides facility for storage management during Tier.
Step 104: create volume and set up the mapping relations being rolled onto virtual disk.
The present embodiment is rolled up Volume, for a user configured capacity management unit, generally corresponds to a LUN (such as C, D dish) on main frame.In access process, such as, when Volume receives access instruction (such as I/O), according to the mapping relations being rolled onto virtual disk, access instruction can be sent to virtual disk, such as can pass through to search the mapping relations table being rolled onto virtual disk and carry out address conversion, the address that the address of Volume is converted in pool on virtual disk, and access instruction (such as I/O) is issued on virtual disk corresponding in pool.
Preferably, first the present embodiment can create Volume, then Volume is divided into multiple CHUNK (being called for short CK), the big CK such as such as can be divided into, then the virtual disk that these CK are mapped in pool is set up the CK mapping relations to virtual disk, for instance the CK mapping relations table to virtual disk can be set up.In access process, after cigarette receives access instruction, by searching the CK mapping relations table to virtual disk, access instruction is sent to virtual disk corresponding in pool.
The present embodiment method makes the I/O on Volume be dispersed in all of physical disks by setting up one layer of mapping, creates a volume and just can utilize the performance of all physical disks in pool.By this layer of mapping can also make that the dilatation of Volume becomes very easy, the dilatation of Volume is just to increase the entry of CK mapping table, simplifies the management planning cost of user.Can also realizing dynamic classification storage by this layer of mapping, the temperature of statistics CK selects to be mapped to by CK in the physical disks corresponding with temperature, improves storage efficiency
The storage architecture that application the present embodiment creation method creates so that storage system virtualization is only with two-layer mapping table, can decrease space and the access frequency of metadata；
If needing the storage architecture that application the inventive method creates can retain RAID processing module for heritage storage system, compatible traditional storage system, because: on I/O level, the I/O level virtualizing pool in the present embodiment is consistent with tradition RAID array I/O level, virtualization Pool and tradition RAID differ only in virtualization pool volume to virtual disk map adopt be mapping table, and traditional virtual dish employing be direct calculating；Specifically, in the present embodiment virtualize pool be rolled onto virtual disk adopt be mapping table, and tradition be rolled onto RAID array adopt be one_to_one corresponding.Owing to only having fine distinction on I/O level, traditional RAID array can be revised as the storage system of virtualization Pool by the algorithm of fixing calculating being revised as mapping table, the mode of multi-level mapping table is adopted relative to existing virtualization pool system, the framework of the present invention is more suitable for being upgraded to virtualization pool from traditional RAID array, and can be compatible that support virtualization pool and tradition RAID array easily and deposit, use same set of RAID to process code；
It addition, the storage architecture that the present embodiment creation method creates also has the advantage that management is simple, be easy to configuration.
As in figure 2 it is shown, the storage architecture being created that for application the present embodiment creation method, including: the several logical level of Volume/Pool/Tier/RAID/LogicDisk/Block；
Tier layer:
Wherein Tier is the concept in classification storage, and a Tier is exactly a level of classification storage, represents the combination of the hard disk of a kind of special speed.When supporting classification storage, a usual pool can create three Tier: high performance SSD layer is Tier0, and the SAS layer of medium-performance is Tier1, and jumbo SATA layer is Tier2.
The Logical Disk group that virtual disk is made up of multiple LogicDisk, virtual disk primary responsibility RAID calculates.Hard disk in same Tier has identical speed.Tier and virtual disk logically belong to same layer.Tier and virtual disk are to comprise and involved relation.The RAID that one Tier is created by multiple same type dishes forms.
Management can create Tier and virtual disk on path.But on I/O path, the I/O that Volume issues directly is issued on virtual disk, after doing RAID calculating, it is handed down to LogicDisk.Tier is on I/O path and is absent from.
As it is shown on figure 3, for Volume in storage architecture shown in Fig. 2 to the mapping of virtual disk, in Fig. 3, Volume such as is divided at the big CK, and is mapped to virtual disk.
As shown in Figure 4, for LogicDisk in storage architecture described in Fig. 2 to the mapping of physical disks, in Fig. 4, physical disks is divided into BLK, BLK composition Logical Disk, and in Logical Disk, BLK is mapped to physical disks.
Introduce the creation method of the present embodiment in detail below, including four-stage:
1, user creates PDG (set of physical disks)
User selects physical disks, physical disks is joined and can comprise multiple physical disks in PDG, PDG.
2, user creates pool
The handling process that user creates pool is as follows:
User initiates storage pool request to create, it is intended that following parameter:
The title of Pool,
A, from one or more PDG select Pool capacity
Configuration RAID parameter, level of redundancy (RAID0, RAID1, RAID5 (2+1), RAID5 (4+1), RAID5 (8+1), RAID6 (4+2), RAID6 (8+2))
Residual capacity alarm threshold (%): the remaining space of storage pool is comprehensively less than or equal to producing an alarm information after this threshold value
Retaining space strategy: nothing, low, high；
A, storage system receive the establishment pool request of user, start the establishment of pool, select BLK from PDG, create LD, create VD, write metadata.
C, Pool have created.
3, user creates volume
The handling process that user creates volume is as follows:
A, user initiate storage pool request to create, it is intended that parameter, including: the name of volume, capacity etc.
B, storage system receive user and create the request of volume, start the establishment of volume, big CK such as being divided into will be rolled up, and distribution CK maps from the RAID of Pool, when the CK number deficiency of Pool, start the LogickDisk mapping to physical disks, create CK, until all spaces all the allocated (distribution can also be postponed) on volume.
C, the storage complete volume of system creation mapping after, write metadata, and return the result creating volume to user.
D, volume establishment complete.
4, volume is mapped to main frame access
The flow process that volume is mapped to main frame is consistent with the flow process of traditional storage system, is not discussed herein.
PDG is made up of multiple physical disks in the present embodiment, also just says and multiple physical disks can be divided into multiple PDG.As it is shown in figure 5, the set PDG (physicdiskgroup) of multiple physical disks one physical disks of composition, a PDG can create 1 or multiple pool, pool distribute BLK resource from PDG.Different Pool on PDG can be differently configured parameter, multiple pool share the BLK resource on PDG.One Pool can also cross over multiple PDG to be set up.Such benefit is the division by PDG, it is possible to physical disks is divided into different failure domains, belong to the physical disks of different PDG be each other do not have influential.And inside same PDG, it is possible to configure the Pool of multiple different parameters, to meet the different demands of user.One pool across PDG, so can make full use of the resource of PDG.
The data volume that the virtualization architecture adopting the establishment of the present embodiment creation method has metadata is little, and management is simple, it is possible to the advantage of the compatible heritage storage system based on RAID array.
Embodiment two:
As shown in Figure 6, present embodiments provide a kind of storage access method, based on the storage architecture that method as shown in Figure 1 creates, comprise the steps:
Step 601: after cigarette receives access instruction, is sent to the virtual disk of correspondence according to the described mapping relations being rolled onto virtual disk by described access instruction.
Preferentially, in the present embodiment, access instruction includes I/O instruction.
Preferentially, when in storage architecture, Logical Disk is made up of the block at least one physical disks, and when Logical Disk includes the mapping relations of Logical Disk block to physical disks to the mapping relations of physical disks；In this step, described access instruction is sent to the process that the physical disks of correspondence conducts interviews and may include that the block being sent in the physical disks of correspondence by described access instruction according to the mapping relations of block in the Logical Disk determined to physical disks conducts interviews by mapping relations according to the Logical Disk determined to physical disks.
When creating storage architecture, volume can be divided into multiple CHUNK, it is called for short CK and such as can will roll up big CK such as being divided into, now volume includes multiple CK, then set up the CK mapping relations to virtual disk, when accessing, it is possible to according to the mapping relations of CK to virtual disk, access instruction is sent to the virtual disk of correspondence.
In the present embodiment after tabling look-up, if it find that when needing to split access instruction, described access instruction is split into many sub-access instructions and is sent to CK corresponding in volume, then described sub-access instruction being sent to the disk array in storage pool, this disk array is corresponding with the CK receiving instruction in described volume.
Step 602: after described virtual disk receives described access instruction, carries out RAID and calculates the Logical Disk determining correspondence, according to the mapping table of the Logical Disk determined to physical disks, described access instruction is sent to corresponding physical disks and conducts interviews.
Preferably, when in described storage architecture, volume is divided into multiple CK, and the mapping relations being rolled onto virtual disk are when including that in volume, CK is to the mapping relations of virtual disk, in this step according to described in be rolled onto the process of the virtual disk that described access instruction is sent to correspondence by the mapping relations of virtual disk and may include that described access instruction is sent to corresponding virtual disk by the mapping relations according to CK in described volume to virtual disk.
Create storage architecture time, it is possible to physical disks is divided into multiple BLOCK, choose BLOCK form Logical Disk, then set up Logical Disk to physical disks the mapping relations of BLK；Can conduct interviews according to the block that access instruction is issued in corresponding physical disks by the mapping relations of BLK in Logical Disk to physical disks when accessing.
The present embodiment storage access method adopts two-layer to map, and decreases the expense searching metadata, has bigger advantage in access performance.
Describe a kind of I/O browsing process as described in Figure 7, based on the storage architecture shown in Fig. 2, including:
The I/O request that step 701:Volume layer Receiving Host sends；
Step 702: search the mapping table being rolled onto virtual disk, the address of Volume is converted to the address of virtual disk in pool；
Step 703: judge that I/O is the need of fractionation, if so, then performs step 704, then performs step 705 if not；
Step 704: I/O is split as multiple I/O and is issued to CK corresponding in Volume；
Step 705: I/O is issued to the virtual disk of corresponding address, carries out RAID process and determines Logical Disk；
Step 706: by searching Logical Disk to the mapping table of block in physical disks, obtaining the address of block in the physical disks corresponding with Logical Disk, the block that I/O is issued to corresponding address carries out I/O access.
Embodiment three:
As shown in Figure 8, present embodiments provide a kind of storage system, the storage architecture that described storage system creates based on the method described in Fig. 1, including: volume processing module and virtual disk processing module；
Described volume processing module, for after cigarette receives access instruction, is sent to the virtual disk of correspondence according to the described mapping relations being rolled onto virtual disk by described access instruction；
Described virtual disk processing module, after receiving described access instruction at described virtual disk, carry out RAID and calculate the Logical Disk determining correspondence, according to the mapping relations of the Logical Disk determined to physical disks, described access instruction is sent to corresponding physical disks and conducts interviews.
The storage system of the present embodiment can enable the heritage storage system based on RAID array upgrade to virtualized pool on the basis of maintenance RAID process code and store system, this storage system has simplification user's configuration, makes full use of the performance of all hard disks, the advantage of the fast quick-recovery of energy when dish fault.Owing to adopting the Intel Virtualization Technology of two-layer, decrease the expense searching metadata, access performance has bigger advantage.
Preferably, described virtual disk processing module, for when in described storage architecture Logical Disk be made up of the block at least one physical disks, and Logical Disk when including the mapping relations of Logical Disk block to physical disks to the mapping relations of physical disks, conduct interviews according to the block that described access instruction is sent in the physical disks of correspondence by the mapping relations of block in the Logical Disk determined to physical disks.
Preferably, described volume processing module, for when in described storage architecture volume be divided into multiple CK, and when the mapping relations being rolled onto virtual disk include CK in volume to the mapping relations of virtual disk, according to the mapping relations of CK in described volume to virtual disk, described access instruction is sent to corresponding virtual disk.
Above content is in conjunction with specific embodiment further description made for the present invention, it is impossible to assert that specific embodiment of the invention is confined to these explanations.For general technical staff of the technical field of the invention, without departing from the inventive concept of the premise, it is also possible to make some simple deduction or replace, protection scope of the present invention all should be considered as belonging to.
Claims (11)
1. the creation method of a storage architecture, it is characterised in that comprise the steps:
Create Logical Disk according to physical disks, and set up the Logical Disk mapping relations to physical disks；
Selecting Logical Disk to create virtual disk, described virtual disk is made up of at least one Logical Disk；
Creating storage pool according to virtual disk, described storage pool is made up of at least one virtual disk；
Create volume and set up the mapping relations being rolled onto virtual disk.
2. the method for claim 1, it is characterised in that described create Logical Disk according to physical disks, and set up Logical Disk and include to the step of the mapping relations of physical disks:
At least one physical disks being divided into multiple pieces, selects block to create Logical Disk, described Logical Disk is made up of the block at least one physical disks；
Set up Logical Disk to physical disks the mapping relations of block.
3. method as claimed in claim 2, it is characterised in that described at least one physical disks is divided into multiple pieces, selects the step that block creates Logical Disk to include:
Physical disks at least one physical disks set is divided into multiple pieces, and described physical disks set is made up of multiple physical disks；
Block composition Logical Disk is selected from a described physical disks set.
4. the method as described in any one of claim 1-3, it is characterised in that the step of the mapping relations that described establishment is rolled up and is rolled onto virtual disk includes:
Create volume, described volume is divided into multiple CK；
Set up in volume CK to the mapping relations of virtual disk.
5. the method as described in any one of claim 1-3, it is characterised in that the described step according to virtual disk establishment storage pool includes:
Selecting virtual disk to create accumulation layer, described accumulation layer is made up of at least one virtual disk；
Selecting accumulation layer to create storage pool, described storage pool is made up of accumulation layer at least one described.
6. a storage access method, it is characterised in that the storage architecture that described storage access method creates based on method as claimed in claim 1, described storage access method comprises the steps:
After cigarette receives access instruction, according to the described mapping relations being rolled onto virtual disk, described access instruction is sent to the virtual disk of correspondence；
After described virtual disk receives described access instruction, carry out RAID and calculate the Logical Disk determining correspondence, according to the mapping relations of the Logical Disk determined to physical disks, described access instruction is sent to corresponding physical disks and conducts interviews.
7. method as claimed in claim 6, it is characterized in that, when in described storage architecture, Logical Disk is made up of the block at least one physical disks, and Logical Disk when including the mapping relations of Logical Disk block to physical disks to the mapping relations of physical disks, described access instruction is sent to the step that the physical disks of correspondence conducts interviews and includes by the described mapping relations according to the Logical Disk determined to physical disks:
Conduct interviews according to the block that described access instruction is sent in the physical disks of correspondence by the mapping relations of block in the Logical Disk determined to physical disks.
8. method as claimed in claims 6 or 7, it is characterized in that, when in described storage architecture, volume is divided into multiple CK, and the mapping relations being rolled onto virtual disk are when including that in volume, CK is to the mapping relations of virtual disk, it are rolled onto the mapping relations of virtual disk described in described basis and described access instruction is sent to the step of corresponding virtual disk includes:
Described access instruction is sent to the virtual disk of correspondence by the mapping relations according to CK in described volume to virtual disk.
9. a storage system, it is characterised in that the storage architecture that described storage system creates based on the method for claim 1, including: volume processing module and virtual disk processing module；
Described volume processing module, for after cigarette receives access instruction, is sent to the virtual disk of correspondence according to the described mapping relations being rolled onto virtual disk by described access instruction；
Described virtual disk processing module, after receiving described access instruction at described virtual disk, carry out RAID and calculate the Logical Disk determining correspondence, according to the mapping relations of the Logical Disk determined to physical disks, described access instruction is sent to corresponding physical disks and conducts interviews.
10. store system as claimed in claim 8, it is characterized in that, described virtual disk processing module, for when in described storage architecture Logical Disk be made up of the block at least one physical disks, and Logical Disk when including the mapping relations of Logical Disk block to physical disks to the mapping relations of physical disks, conduct interviews according to the block that described access instruction is sent in the physical disks of correspondence by the mapping relations of block in the Logical Disk determined to physical disks.
11. the storage system as described in claim 9 or 10, it is characterized in that, described volume processing module, for when in described storage architecture volume be divided into multiple CK, and the mapping relations being rolled onto virtual disk are when including that in volume, CK is to the mapping relations of virtual disk, according to the mapping relations of CK in described volume to virtual disk, described access instruction is sent to corresponding virtual disk.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410854717.2A CN105808157A (en) | 2014-12-31 | 2014-12-31 | Storage architecture creating method, storage access method and storage system |
| PCT/CN2015/085787 WO2016107167A1 (en) | 2014-12-31 | 2015-07-31 | Storage architecture creation method, storage access method and storage system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410854717.2A CN105808157A (en) | 2014-12-31 | 2014-12-31 | Storage architecture creating method, storage access method and storage system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105808157A true CN105808157A (en) | 2016-07-27 |
Family
ID=56284115
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410854717.2A Withdrawn CN105808157A (en) | 2014-12-31 | 2014-12-31 | Storage architecture creating method, storage access method and storage system |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN105808157A (en) |
| WO (1) | WO2016107167A1 (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107608914A (en) * | 2017-09-29 | 2018-01-19 | 锤子科技（北京）股份有限公司 | A kind of access method, device and the mobile terminal of multichannel storage device |
| CN107888689A (en) * | 2017-11-16 | 2018-04-06 | 无锡地铁集团有限公司 | Locking resource allocation method based on shared storage |
| WO2019127490A1 (en) * | 2017-12-29 | 2019-07-04 | 华为技术有限公司 | Data access method and storage array |
| CN110134338A (en) * | 2019-05-21 | 2019-08-16 | 深信服科技股份有限公司 | A kind of distributed memory system and its data redundancy protection method and relevant device |
| CN112988468A (en) * | 2021-04-27 | 2021-06-18 | 云宏信息科技股份有限公司 | Method for virtualizing operating system using Ceph and computer-readable storage medium |
| CN114968128A (en) * | 2022-07-28 | 2022-08-30 | 云宏信息科技股份有限公司 | Qcow 2-based virtual disk mapping method, system and medium |
| CN116027991A (en) * | 2023-03-29 | 2023-04-28 | 苏州浪潮智能科技有限公司 | RAID array management method and device, RAID card and storage medium |
| CN117369732A (en) * | 2023-12-07 | 2024-01-09 | 苏州元脑智能科技有限公司 | Logic disc processing method and device, electronic equipment and storage medium |
| WO2024119771A1 (en) * | 2022-12-05 | 2024-06-13 | 苏州元脑智能科技有限公司 | Storage virtualization method, system, apparatus, and device based on disk array card |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108932105A (en) * | 2017-05-25 | 2018-12-04 | 中兴通讯股份有限公司 | A kind of configuration method and system of the classification storage of storage gateway |
| CN112162700B (en) * | 2020-09-18 | 2024-07-02 | 北京浪潮数据技术有限公司 | Storage volume creation method and device, electronic equipment and storage medium |
| CN114461138B (en) * | 2021-12-24 | 2024-06-07 | 天翼云科技有限公司 | Scheduling method and device of storage resources, electronic equipment and readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101067779A (en) * | 2006-04-04 | 2007-11-07 | 株式会社东芝 | Storage system for processing access request from host and method for doing the same |
| JP2008269090A (en) * | 2007-04-17 | 2008-11-06 | Fujitsu Ltd | Information processor having virtualization processing function, virtualization processing method, and program for it |
| CN101788889A (en) * | 2010-03-03 | 2010-07-28 | 浪潮(北京)电子信息产业有限公司 | Memory virtualization system and method |
| CN102521010A (en) * | 2010-11-03 | 2012-06-27 | 微软公司 | Virtualization and offload reads and writes |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102405460B (en) * | 2009-02-11 | 2016-05-04 | 艾梵尼达有限公司 | Virtual storage system and operation method thereof |
| CN102255962B (en) * | 2011-07-01 | 2013-11-06 | 华为数字技术（成都）有限公司 | Distributive storage method, device and system |
-
2014
- 2014-12-31 CN CN201410854717.2A patent/CN105808157A/en not_active Withdrawn
-
2015
- 2015-07-31 WO PCT/CN2015/085787 patent/WO2016107167A1/en active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101067779A (en) * | 2006-04-04 | 2007-11-07 | 株式会社东芝 | Storage system for processing access request from host and method for doing the same |
| JP2008269090A (en) * | 2007-04-17 | 2008-11-06 | Fujitsu Ltd | Information processor having virtualization processing function, virtualization processing method, and program for it |
| CN101788889A (en) * | 2010-03-03 | 2010-07-28 | 浪潮(北京)电子信息产业有限公司 | Memory virtualization system and method |
| CN102521010A (en) * | 2010-11-03 | 2012-06-27 | 微软公司 | Virtualization and offload reads and writes |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107608914B (en) * | 2017-09-29 | 2020-09-29 | 北京字节跳动网络技术有限公司 | Access method and device of multi-channel storage equipment and mobile terminal |
| CN107608914A (en) * | 2017-09-29 | 2018-01-19 | 锤子科技（北京）股份有限公司 | A kind of access method, device and the mobile terminal of multichannel storage device |
| CN107888689A (en) * | 2017-11-16 | 2018-04-06 | 无锡地铁集团有限公司 | Locking resource allocation method based on shared storage |
| CN107888689B (en) * | 2017-11-16 | 2019-04-30 | 无锡地铁集团有限公司 | Locking resource allocation method based on shared storage |
| US11301157B2 (en) | 2017-12-29 | 2022-04-12 | Huawei Technologies Co., Ltd. | Data access method and storage array |
| WO2019127490A1 (en) * | 2017-12-29 | 2019-07-04 | 华为技术有限公司 | Data access method and storage array |
| CN110214305A (en) * | 2017-12-29 | 2019-09-06 | 华为技术有限公司 | A kind of data access method and storage array |
| CN110214305B (en) * | 2017-12-29 | 2020-09-08 | 华为技术有限公司 | Data access method and storage array |
| CN110134338A (en) * | 2019-05-21 | 2019-08-16 | 深信服科技股份有限公司 | A kind of distributed memory system and its data redundancy protection method and relevant device |
| CN112988468A (en) * | 2021-04-27 | 2021-06-18 | 云宏信息科技股份有限公司 | Method for virtualizing operating system using Ceph and computer-readable storage medium |
| CN114968128A (en) * | 2022-07-28 | 2022-08-30 | 云宏信息科技股份有限公司 | Qcow 2-based virtual disk mapping method, system and medium |
| WO2024119771A1 (en) * | 2022-12-05 | 2024-06-13 | 苏州元脑智能科技有限公司 | Storage virtualization method, system, apparatus, and device based on disk array card |
| CN116027991A (en) * | 2023-03-29 | 2023-04-28 | 苏州浪潮智能科技有限公司 | RAID array management method and device, RAID card and storage medium |
| CN116027991B (en) * | 2023-03-29 | 2023-09-19 | 苏州浪潮智能科技有限公司 | RAID array management method and device, RAID card and storage medium |
| CN117369732A (en) * | 2023-12-07 | 2024-01-09 | 苏州元脑智能科技有限公司 | Logic disc processing method and device, electronic equipment and storage medium |
| CN117369732B (en) * | 2023-12-07 | 2024-02-23 | 苏州元脑智能科技有限公司 | Logic disc processing method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2016107167A1 (en) | 2016-07-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105808157A (en) | Storage architecture creating method, storage access method and storage system | |
| CN102449607B (en) | Storage system provided with a plurality of flash packages | |
| US8464003B2 (en) | Method and apparatus to manage object based tier | |
| US20220137849A1 (en) | Fragment Management Method and Fragment Management Apparatus | |
| US9658779B2 (en) | Computer system and control method for computer system | |
| CN104636080B (en) | Storage system and the method for it | |
| CN111158587B (en) | Distributed storage system based on storage pool virtualization management and data read-write method | |
| CN101604226B (en) | Method for building dynamic buffer pool to improve performance of storage system based on virtual RAID | |
| US9009405B2 (en) | Methods and systems for instantaneous online capacity expansion | |
| CN104115109A (en) | Storage apparatus with a plurality of nonvolatile semiconductor storage units and control method thereof to place hot data in storage units with higher residual life and cold data in storage units with lower residual life | |
| CN104123100A (en) | Controlling data storage in an array of storage devices | |
| CN104536903B (en) | A kind of mixing storage method and system stored classifiedly by data attribute | |
| US20140149707A1 (en) | Method and apparatus to manage tier information | |
| US9760292B2 (en) | Storage system and storage control method | |
| US10860260B2 (en) | Method, apparatus and computer program product for managing storage system | |
| CN105988727B (en) | Storage method and storage device based on RAID | |
| US11150990B2 (en) | Method, apparatus and computer readable medium for managing a plurality of disks | |
| CN104731517A (en) | Method and device for allocating capacity of storage pool | |
| US8650358B2 (en) | Storage system providing virtual volume and electrical power saving control method including moving data and changing allocations between real and virtual storage areas | |
| CN102799533A (en) | Method and apparatus for shielding damaged sector of disk | |
| LU102091B1 (en) | Zone hints for zoned namespace storage devices | |
| CN104050014B (en) | Efficient storage management method based on virtual platform | |
| CN104216660A (en) | Method and device for improving disk array performance | |
| US9069471B2 (en) | Passing hint of page allocation of thin provisioning with multiple virtual volumes fit to parallel data access | |
| Qian et al. | Thinstore: Out-of-band virtualization with thin provisioning |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20160727 |