CN101976181A - Management method and device of storage resources - Google Patents
Management method and device of storage resources Download PDFInfo
- Publication number
- CN101976181A CN101976181A CN2010105080409A CN201010508040A CN101976181A CN 101976181 A CN101976181 A CN 101976181A CN 2010105080409 A CN2010105080409 A CN 2010105080409A CN 201010508040 A CN201010508040 A CN 201010508040A CN 101976181 A CN101976181 A CN 101976181A
- Authority
- CN
- China
- Prior art keywords
- resource
- logical
- resource block
- logical resource
- raid array
- 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
Images
Abstract
The invention provides a management method of storage resources and a network storage system, wherein the network storage system comprises a storage controller and a plurality of magnetic discs, wherein at least one Raid array is established by using the magnetic discs; a port at the front end of the storage controller is connected to a client mainframe through a network; and the port at the back end of the magnetic disc is connected with the magnetic disc. A management mechanism comprises the following steps of: dividing an array space into blocks through presetting a resource block different in sizes, and establishing a corresponding block table used for recording whether the resource blocks are occupied or not; allocating the resource blocks and establishing a logical resource index table while establishing a logical resource so as to record a corresponding relation between the logical resources and the resource blocks; and carrying out management operation on the logical resources by taking the resource blocks as units, and updating the block table and the logical resource index table after the management operation is finished. The management method is simple, flexible, strong in expansibility and succinct in realization and has lower cost.
Description
Technical field
The present invention relates to Network storage technology, relate in particular to a kind of administrative skill by different level of storage resources.
Background technology
In the network environment of the data storage that relates to numerous main frames, in order to improve reliability of data storage and security, simultaneously for expansion of storage capacity and dirigibility, Network storage technology is arisen at the historic moment.As a rule, the effect of network store system is to provide available storage space for client PC or server (generally being referred to as main frame or Host).
The core of network store system is memory controller (Controller), and memory controller carries out data processing and writes data in the back-end physical disk.For raising is write the performance of physical disk and data redundancy is provided, memory controller is supported Redundant Array of Independent Disks (RAID) (Raid) technology usually, can provide various other performances of level and fiduciary level according to other difference of Raid level, promptly can guarantee in most cases, under the situation of the fault of one or more disk, data in it can be recovered promptly usually said Raid array rebuild by corresponding checking algorithm.The front end of network store system can link to each other with main frame (Host) by IP network or FC network, for main frame provides data storage service.In data transmission, be example with network store system based on the IP carrying, main frame can measured ISCSI (Internet Small Computer Systems Interface) protocol network storage system carry out the data write operation.
For making things convenient for the reading and writing data of main frame, to main frame shielding storage resources difference physically, can carry out abstract to the physical store resource by Intel Virtualization Technology, more than physical layer, form the logical resource of user oriented main frame, and corresponding formation logic resource address mapping table is with the mapping relations of record logical address and physical address.
For user's various application, every kind of application need storage and the data of visiting all have life cycle, such as mobile charge system, the telephone expenses data of every month mobile phone may repeatedly be visited in three months after data generate, accessed once in a while in 1 year, may just nobody shows any interest in again backward, become filing data, but have only the time limit that has arrived regulation when it, could be deleted.
For network store system, even same network store system, disk type, disk running speed etc. influence the performance of disk, on this basis, form disk type difference, number of disks difference, the array Raid rank difference of Raid array, the Performance And Reliability of Raid array also has nothing in common with each other.Therefore the notion that the data hierarchy storage occurred." layering " is meant according to the actual count data of client to data storage and visit, the data of user storage in network store system are moved on different Raid arrays, promptly the long-time data Autonomic Migration Framework of not visiting of handle is to slower, the lower-cost Raid array of speed, as the SATA disk, reduce hardware cost with this; And those frequent accessed or important data are placed on the Raid array that speed is very fast, cost is higher, as SAS disk, SSD disk, promote performance with this.The migration of data simultaneously is that the aspect below logical resource is carried out, and whole process is transparent for the user, and the logical mappings unit in the system can not change the mapping of logical address to physical address according to the result of migration.
The layer-management technology of more common network store system resource has two kinds at present.A kind of segment management that is based on the Raid array, its hierarchical structure as shown in Figure 1.Basic management process is as follows: the first step: use disk to create the Raid array; Second step: use the free segment of Raid array to create logical resource, can create a plurality of logical resources on the Raid array, a logical resource can be distributed in a plurality of RAID.
In this way to manage, Raid array and logical resource are the relations of multi-to-multi, so storage space management is simple relatively flexibly, get final product for logical resource increases new section.Segment management based on the Raid array, logical resource and Raid array all need to safeguard resource placement's mapping relations of multi-to-multi, the size of each section depends on the idle physical segment size of Raid array and the logical resource size that will create, and the length of section is not what fix.Yet there is certain drawback in this administrative skill.On the one hand, the section on logical resource and the Raid array after more repeatedly creating, deleting logical resource, will produce a large amount of fragments; On the other hand, for the demand of AUTOMATIC ZONING, if whole section migration, the granularity of migration can be bigger, target phase that might the size of the section of finding can satisfy the demands in the Raid array; If the migration partial data then may need the section that oneself has is cut into slices again, fragment is more, manages more complicated; In general, support that based on the segment management of Raid array the cost of AUTOMATIC ZONING is bigger.
Another kind of existing administrative skill is the way to manage of logical resource and the unification of Raid array, and its hierarchical structure as shown in Figure 2.Its basic management step is as follows: the first step: use disk to create disk groups, disk groups is represented one group of disk, is not Raid array entity; Second step: create the Raid array based on disk groups, the Raid array entity of establishment also is the storage space (logical resource) that client can directly be visited.
In this way to manage,, array entity and the unification of logical resource entity, data routing is short, and performance is more excellent.Way to manage for logical resource and the unification of Raid array, logical resource is exactly a Raid array, a Raid array can only be created based on a disk groups, though yet this administrative skill is relatively simplified, but there are some tangible technical matterss in it: first, for the logical resource dilatation, no matter be dilatation on same disk groups (disk space that is same Raid is not a continuous distribution), also be to use different disk groups dilatations, all need on logical resource, additionally increase a virtualization layer, realize complicated; The second, for the demand of AUTOMATIC ZONING,, therefore can only move together by whole Raid array because the minimum particle size of logical resource management is exactly the Raid array, the granularity of migration is too big.On this basis, a kind of improvement project is, logic-based resource division volume is realized the migration based on the son volume, this scheme refinement the granularity of migration, but need on existing logical resource, increase sub-volume management layer; Only on existing logical resource, divide the son volume in addition, the free space unified management is not got up.In sum, at present the user to press for a kind of way to manage simple, flexible and migrations data under the situation of AUTOMATIC ZONING of dilatation are easy to realize, simultaneously the lower storage of hierarchically administrative skill of cost.
Summary of the invention
In view of this, the object of the present invention is to provide a kind of way to manage simple, flexible and migrations data under the situation of AUTOMATIC ZONING of dilatation are easy to realize, simultaneously the lower storage of hierarchically administrative skill of cost.The invention provides technical scheme for achieving the above object is:
A kind of management method of storage resources, it is applied in the network store system, wherein said network store system comprises memory controller and polylith disk, and use disk to create at least one raid-array (Raid array), the front end port of described memory controller connects main frame by network, and the rear end disk port of described memory controller links to each other with described disk; Wherein this method comprises:
A, according to default resource block size the available storage space of above-mentioned Raid array is carried out piecemeal, in described memory controller above-mentioned Raid array is set up corresponding piece table, whether described table be occupied in order to each Resource Block that writes down above-mentioned Raid array;
B is that logical resource distributes at least one Resource Block when creating logical resource, and the Resource Block that will distribute to this logical resource in the piece table of the Raid array that logical resource takies is labeled as seizure condition;
C sets up the corresponding relation of logical resource concordance list with record logical resource and the Resource Block that himself is assigned to for above-mentioned logical resource in described memory controller;
D is that unit manages operation to above-mentioned logical resource with the Resource Block, and upgrades described table and logical resource concordance list after bookkeeping is finished.
Preferably, described to be unit with the Resource Block manage operation to logical resource is specially: be that unit is that logical resource to be expanded distributes the more resources piece with the Resource Block, and upgrade the piece table of the Raid array that is used for dilatation and the logical resource concordance list of described logical resource; Perhaps be that unit moves to the data in the logical resource on another Raid array from a Raid array with the Resource Block, and upgrade the piece table and the described logical resource concordance list of two Raid arrays, wherein the resource block size of two Raid arrays is identical.
Preferably, also comprise after the described step D: E, carry out the mapping alter operation of corresponding logical address to physical address to bookkeeping.
Preferably, the Raid array of wherein said network store system is divided in the first and second two storage pools at least, described two storage pools include at least one Raid array, and resource block size and the interior default resource block size of second storage pool default in described first storage pool are identical or different; Be that the logical resource bookkeeping is that boundary is carried out with same storage pool all among logical resource distributes resource blocks and the step D in the described steps A.
The present invention also provides a kind of network store system management devices, be used to carry out the resource management action in the network store system, wherein network store system comprises memory controller and polylith disk, and use disk to create at least one raid-array (Raid array), the front end port of described memory controller connects main frame by network, and the rear end disk port of described memory controller links to each other with described disk; It is characterized in that this management devices comprises:
The Resource Block administrative unit, be used for the available storage space of above-mentioned Raid array being carried out piecemeal according to default resource block size, in described memory controller above-mentioned Raid array is set up corresponding piece table, whether described table be occupied in order to write down each Resource Block;
The management initialization unit is that logical resource distributes at least one Resource Block when being used to create logical resource, and the Resource Block that will distribute to this logical resource in the piece table of the Raid array that logical resource takies is labeled as seizure condition; And in described memory controller, set up the corresponding relation of logical resource concordance list with record logical resource and the Resource Block that himself is assigned to;
The bookkeeping processing unit, being used for the Resource Block is that unit manages operation to logical resource, and upgrades described table and logical resource concordance list after bookkeeping is finished.
Preferably, to be unit with the Resource Block manage operation to logical resource to described bookkeeping processing unit is specially: be that unit is that logical resource to be expanded distributes the more resources piece with the Resource Block, and upgrade the piece table of the Raid array that is used for dilatation and the logical resource concordance list of described logical resource; Perhaps be that unit moves to the data in the logical resource on another Raid array from a Raid array with the Resource Block, and upgrade the piece table and the described logical resource concordance list of two Raid arrays, the resource block size of wherein said two Raid arrays is identical.
Preferably, described management devices also comprises the logical mappings unit, is used for bookkeeping is carried out the mapping alter operation of corresponding logical address to physical address.
Preferably, the Raid array of wherein said network store system is divided in the first and second two storage pools at least, described two storage pools include at least one Raid array, resource block size and the interior default resource block size of second storage pool default in described first storage pool are identical or different, described Resource Block administrative unit is that the Resource Block that logical resource distributes is that boundary is carried out with same storage pool, and described bookkeeping processing module is that boundary is carried out bookkeeping with same storage pool.
Description of drawings
Fig. 1 is existing segment management layer of structure figure.
Fig. 2 is the layer of structure figure of existing logical resource and Raid array unification way to manage.
Fig. 3 is a network store system management devices building-block of logic of the present invention.
Fig. 4 is a resource block assignments synoptic diagram in the network store system of the present invention.
Fig. 5 is another synoptic diagram of resource block assignments in the network store system of the present invention.
Fig. 6 is the synoptic diagram of logical resource concordance list of the present invention.
Fig. 7 is a logical resource dilation process synoptic diagram of the present invention.
Fig. 8 is a logical resource data migration synoptic diagram of the present invention.
Embodiment
The present invention focuses on the administrative skill of storage resources, and it can solve the practical problem on Storage Resource Management (SRM) in the prior art, can catch up with the application of layer and the seamless fusion of technology of bottom simultaneously.Preferred implementation of the present invention is to adopt software engineering to realize, the mode of not getting rid of some software and hardware combining or firmware certainly realizes, introduces the typical treatment scheme of the present invention below in conjunction with some examples.In the present invention if no special instructions, and storage system is network store system, and if not otherwise specified, storage space is meant storage space on the logic aspect.
Please refer to Fig. 3, network store system management devices of the present invention moves in network store system and mainly may further comprise the steps:
Step 101 utilizes a plurality of disks to create at least one Raid array in network store system; Please contrast Fig. 3, this step is carried out by array management unit 10; Comprise memory controller and first and second raid-array (Raid array) at least in the present embodiment in the network store system, the front end port of described memory controller connects main frame by network, and the rear end port of described memory controller links to each other with each physical disk in the described Raid array; The process of creating the Raid array can be with reference to existing Raid technology, and wherein the present invention does not have special requirement to the Raid rank, can be different, as long as the Raid module support in the memory controller.For dilatation, might not need a plurality of arrays in the system; And for data migtation, in general relate to two arrays at least, and two exist one or more differences at least on reliability, cost or three kinds of parameters of access speed; Certainly may exist above-mentioned three parameters of a plurality of other arrays identical equally in the storage system.Need explanation in the present invention, the same with prior art, the Raid array of different stage is when identical storage space that user or administration and supervision authorities provide, be the difference that needs on the verification expense of shielding different stage Raid array, that is to say that the disk space size that in fact takies is also inequality, but this species diversity is invisible for logical resource bookkeeping or user's operation.
Step 102 is carried out piecemeal according to default resource block size to the available storage space of above-mentioned each Raid array, sets up corresponding piece table 20 in described memory controller, and whether described table be occupied in order to write down each Resource Block; This step is carried out by Resource Block administrative unit 30.Here Yu She resource block size can be that the user passes through the configuration of Resource Block administrative unit, also can be the good default parameters of software configured in advance.Please refer to Fig. 4 and in conjunction with Fig. 5, system need all safeguard corresponding piece table for each Raid array that array management unit 10 provides, because piece is a regular length, and because each piece characteristic relevant with resources allocation is to have taken and vacant (promptly idle), therefore can use the bitmap table on realizing, promptly piece of a bit mark can improve recall precision on the one hand, can save piece table occupation space on the other hand.Such as the Raid array 1 of a 64GB is divided into 64 according to the block size of 1GB, the piece table of Raid array 1 only needs the space of 8 bytes.Each bit position is that this Resource Block of expression of 1 is occupied, 0 then expression do not have occupied.
Step 103 is that LUN distributes at least one Resource Block when creating logical resource (LUN), and the Resource Block that will distribute to this logical resource in the piece table of the Raid array that logical resource takies is labeled as seizure condition;
Step 104 is after LUN creates, for each LUN sets up the corresponding relation of logical resource concordance list 40 with record logical resource and the Resource Block that himself is assigned to; Step 103 and step 104 are carried out by management initialization unit 50.
Please refer to Fig. 6, which Resource Block that has write down Raid array 1 in the logical resource concordance list has been assigned to LUN1, with index marker A0B10 is example, the tenth Resource Block of its expression Raid array 1, i.e. second capable from left to right the 3rd represented Resource Block in position in the piece table.Though situation shown in Figure 6 is that all Resource Block of LUN1 correspondence are all in the Raid array.What be readily appreciated that walks, and the Resource Block of LUN1 correspondence can be distributed in the different Raid arrays.
Step 105, D is that unit manages operation to logical resource with the Resource Block, and upgrades described table and logical resource concordance list after bookkeeping is finished.
Step 105 is carried out by bookkeeping processing unit 60.Specifically, after Raid array block table and logic index table creation, can begin with the Resource Block is that unit has managed logical resource.The present invention is that example describes with most typical logical resource dilatation and logical resource data migration.For the logical resource dilatation need be that unit is that logical resource to be expanded distributes the more resources piece with the Resource Block, and upgrades the piece table of the Raid array that is used for dilatation and the logical resource concordance list of described logical resource; For data migtation, same is that unit moves to the data in the logical resource on another Raid array from a Raid array with the Resource Block, and upgrade the piece table and the described logical resource concordance list of two Raid arrays, the resource block size of two Raid arrays that wherein participates in migration is identical.
No matter be the logical resource dilatation or the migration of data, it follows the basic management process of the present invention, is the management mode of operation of unit with above-mentioned Resource Block.
Please refer to Fig. 7, is example with the dilatation of logical resource:
All space is according to the piece unified management of fixed length in lower network storage system preferred embodiment, and the bookkeeping processing unit at first will will judge whether the remaining free space of network store system satisfies the demand of dilatation in this step when therefore the dilatation demand being arranged; If satisfy, continue to handle; If do not satisfy, the prompting mistake; It is evident that said free space is meant the free space of all Raid arrays here.
Step I, the capacity that increases according to this dilatation needs, select to use which Raid array to carry out dilatation, retrieve the bitmap table of selected Raid array, find free block (being unappropriated Resource Block), revise the index area of logical resource simultaneously, upgrade described table and logical resource concordance list, should be labeled as again for the Resource Block that is used and to take, then need to increase logical resource and the mapping relations of the Resource Block selected for use for the renewal of logical resource table; There is no specific (special) requirements in this step for the selection of Raid array, and the selection of each piece there is not specific (special) requirements yet in the Raid array.
Step II, repeating step I are up to the demand of finishing this dilatation.
What need supplementary notes is, the bookkeeping processing unit can be done some optimization processes in the process of carrying out dilatation, if such as find that there is continuous free block in certain Raid array, can be in a step a plurality of of dilatations simultaneously, optimize efficient.Fig. 7 dotted portion is clear and expressed to the LUN1 dilatation 8GB visually, and the LU capacity is handed over from 16GB and is 24GB.
Please refer to Fig. 8, the data migtation with logical resource is an example again:
As mentioned in the background technology, in general the basis of AUTOMATIC ZONING is the IO statistics, find the focus (being the data of frequent access) and the cold spot (i.e. the data of not visiting for a long time) of visit according to predetermined strategy or algorithm, carry out data migtation according to predetermined strategy then.Need to prove, statistics based on IO is the popular data hierarchy theory of industry, yet the user may not be cold and hot based on data for the layering of data, therefore performance statistics only is a kind of common data migtation foundation, yet the user also can not carry out data migtation based on this, because the user has control completely to data.With piece be equally but the following this bookkeeping of migration of carrying out data, present embodiment bookkeeping performance element in step 105 need be that unit carries out performance statistics according to piece, distinguish cold spot data and hot spot data, be that data migtation is carried out in the unit according to Resource Block then, Qian Yi granularity is less like this.Specifically in the trigger data migration:, find the target Raid array of migration at first according to predetermined strategy; Such as can select readwrite performance and/or lower-cost Raid array as target Raid array for the cold spot data, and can select the higher Raid array of readwrite performance as target Raid array for hot spot data.
The bitmap of searched targets Raid array table finds free block then, and this piece of mark is available, and is fast to target from source piece synchrodata then; The same with the dilatation example is, if there is continuous free block in target Raid array, can move a plurality of simultaneously, optimizes efficient.
After data sync is finished, upgrade described table and logical resource concordance list, be specially the piece table of modification source Raid array and target Raid array, the bitmap table of piece place, modification source Raid array, mark source piece is a free block, just the space of this piece is discharged, and continues to use.Make amendment for the logical resource concordance list, use the target resource piece to replace the mapping relations of source Resource Block.Please refer to shown in Figure 8ly,, judge that the hot spot data of LUN1 is A1B11 and A1B15, therefore these data of two are moved on the 2nd Raid array (Array2) Block1 of corresponding A rray2 and Block2 from a Raid array (Array1) according to statistics.The dotted portion of Fig. 8 has shown the process of data migtation very visually.The processing of data migtation is except synchrodata, only need to revise the mark in the bitmap table and the index area of logical resource, do not introduce fragment, do not introduce new management entity yet, can realize the unified management in whole network store system stored space according to the performance statistic and the predetermined strategy of network store system.
The same with existing administrative skill, bookkeeping can trigger the more logical mappings operation of bottom, and the mapping alter operation of corresponding logical address to physical address all can be carried out for bookkeeping in the logical mappings unit 70 of general network store system.Such as when migration took place data, logical address need change to the mapping of physical address so that the process of data migtation be sightless for main frame on the network, do not influence user's normal read write operation.
Above step 101 to step 105 has provided general treatment scheme and the corresponding logical block of the present invention.In order further to optimize the solution of the present invention, for realizing that way to manage is in the management mode of administrative mechanism realization aspect introducing storage pool of the present invention more flexibly.The introducing of storage pool mainly brings the variation of two aspects:
The division of storage resources:
At first need the Raid array of described network store system is divided in first and second storage pools, two storage pools all need to comprise at least one Raid array, and resource block size and the interior default resource block size of second storage pool default in described first storage pool are identical or different.The user who has wishes that resource block size all in total system is identical convenient management, but the user may wish simultaneously whole Raid array resource division is two parts at least, respectively towards different types of business, such as divide according to the importance of business, preferentially guarantee the bookkeeping of important service.The user who has then wishes to become whole Raid array partition two parts towards different business, consider towards the reason of professional self-characteristic difference, the user wishes that resource block size need carry out different settings according to the difference of business, promotes the efficient of bookkeeping.
Distributes resource blocks and bookkeeping:
Because storage resources is divided, when needs are employed storage resources, need be with storage pool as management domain.Specifically, be that the logical resource bookkeeping is that boundary is carried out with same storage pool in logical resource distributes resource blocks and the step 105 in the step 103.With a storage pool is that the boundary execution can be understood as, the Resource Block of distributing to logical resource all belongs to same storage pool, and the Resource Block that bookkeeping relates to also all belongs to same storage pool, and the Resource Block that logical resource had before and after management method was carried out is all in same storage pool.
Below introduce the storage pool management mode by an example.For instance, suppose that storage system comprises four Raid arrays, the first and second Raid arrays belong to first storage pool, and the third and fourth Raid array belongs to second storage pool, the Raid rank of each storage pool local array can be different, two arrays that each storage pool comprises have a kind of there are differences at least on reliability, cost or three kinds of parameters of access speed, can comprise that a plurality of arrays are identical in the storage pool on parameter certainly.The resource block size of first storage pool is 1G, and the resource block size of second storage pool is 2G.So-called is that unit manages all data of logical resource in other words all in same storage pool according to storage pool, described Resource Block administrative unit be the Resource Block that distributes of logical resource in same storage pool, described bookkeeping processing module is carried out corresponding bookkeeping according to the storage pool under the logical resource.Need when creating logical resource, step 102 select a storage pool specifically, and from this storage pool, select the resource block assignments of logical resource to give logical resource, manage operation when correspondingly the bookkeeping processing unit is to logical resource in step 105, not not from all Raid arrays of whole storage system, to select Resource Block, but selecting Resource Block to carry out corresponding management in the included Raid array of the storage pool under logical resource.Perhaps understand storage pool with more popular mode, all resources that just look like network store system are divided two parts in management layer, and are identical to this two-part way to manage, and only wherein Resource Block varies in size.
The above only is the preferable implementation of the present invention, and any modification of making based on spirit of the present invention that is equal to all should be covered by in the claim scope of the present invention.
Claims (8)
1. the management method of a storage resources, it is applied in the network store system, wherein said network store system comprises memory controller and polylith disk, and use disk to create at least one raid-array (Raid array), the front end port of described memory controller connects main frame by network, and the rear end disk port of described memory controller links to each other with described disk; It is characterized in that this method comprises:
A, according to default resource block size the available storage space of above-mentioned Raid array is carried out piecemeal, in described memory controller above-mentioned Raid array is set up corresponding piece table, whether described table be occupied in order to each Resource Block that writes down above-mentioned Raid array;
B is that logical resource distributes at least one Resource Block when creating logical resource, and the Resource Block that will distribute to this logical resource in the piece table of the Raid array that logical resource takies is labeled as seizure condition;
C sets up the corresponding relation of logical resource concordance list with record logical resource and the Resource Block that himself is assigned to for above-mentioned logical resource in described memory controller;
D is that unit manages operation to above-mentioned logical resource with the Resource Block, and upgrades described table and logical resource concordance list after bookkeeping is finished.
2. method according to claim 1, it is characterized in that, described to be unit with the Resource Block manage operation to logical resource is specially: be that unit is that logical resource to be expanded distributes the more resources piece with the Resource Block, and upgrade the piece table of the Raid array that is used for dilatation and the logical resource concordance list of described logical resource; Perhaps be that unit moves to the data in the logical resource on another Raid array from a Raid array with the Resource Block, and upgrade the piece table and the described logical resource concordance list of two Raid arrays, wherein the resource block size of two Raid arrays is identical.
3. method according to claim 1 is characterized in that, also comprises after the described step D: E, carry out the mapping alter operation of corresponding logical address to physical address to bookkeeping.
4. according to the described method of claim 1-4, it is characterized in that, the Raid array of wherein said network store system is divided in the first and second two storage pools at least, described two storage pools include at least one Raid array, and resource block size and the interior default resource block size of second storage pool default in described first storage pool are identical or different; Be that the logical resource bookkeeping is that boundary is carried out with same storage pool all among logical resource distributes resource blocks and the step D in the described steps A.
5. network store system management devices, be used to carry out the resource management action in the network store system, wherein network store system comprises memory controller and polylith disk, and use disk to create at least one raid-array (Raid array), the front end port of described memory controller connects main frame by network, and the rear end disk port of described memory controller links to each other with described disk; It is characterized in that this management devices comprises:
The Resource Block administrative unit, be used for the available storage space of above-mentioned Raid array being carried out piecemeal according to default resource block size, in described memory controller above-mentioned Raid array is set up corresponding piece table, whether described table be occupied in order to write down each Resource Block;
The management initialization unit is that logical resource distributes at least one Resource Block when being used to create logical resource, and the Resource Block that will distribute to this logical resource in the piece table of the Raid array that logical resource takies is labeled as seizure condition; And in described memory controller, set up the corresponding relation of logical resource concordance list with record logical resource and the Resource Block that himself is assigned to;
The bookkeeping processing unit, being used for the Resource Block is that unit manages operation to logical resource, and upgrades described table and logical resource concordance list after bookkeeping is finished.
6. management devices according to claim 5, it is characterized in that, to be unit with the Resource Block manage operation to logical resource to described bookkeeping processing unit is specially: be that unit is that logical resource to be expanded distributes the more resources piece with the Resource Block, and upgrade the piece table of the Raid array that is used for dilatation and the logical resource concordance list of described logical resource; Perhaps be that unit moves to the data in the logical resource on another Raid array from a Raid array with the Resource Block, and upgrade the piece table and the described logical resource concordance list of two Raid arrays, the resource block size of wherein said two Raid arrays is identical.
7. management devices according to claim 5 is characterized in that, also comprises: the logical mappings unit is used for bookkeeping is carried out the mapping alter operation of corresponding logical address to physical address.
8. according to the described management devices of claim 5-7, it is characterized in that, the Raid array of wherein said network store system is divided in the first and second two storage pools at least, described two storage pools include at least one Raid array, resource block size and the interior default resource block size of second storage pool default in described first storage pool are identical or different, described Resource Block administrative unit is that the Resource Block that logical resource distributes is that boundary is carried out with same storage pool, and described bookkeeping processing module is that boundary is carried out bookkeeping with same storage pool.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010105080409A CN101976181A (en) | 2010-10-15 | 2010-10-15 | Management method and device of storage resources |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010105080409A CN101976181A (en) | 2010-10-15 | 2010-10-15 | Management method and device of storage resources |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101976181A true CN101976181A (en) | 2011-02-16 |
Family
ID=43576069
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010105080409A Pending CN101976181A (en) | 2010-10-15 | 2010-10-15 | Management method and device of storage resources |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101976181A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102123176A (en) * | 2011-03-17 | 2011-07-13 | 杭州宏杉科技有限公司 | Space distribution and management method and device for network storage system |
| CN102147714A (en) * | 2011-03-17 | 2011-08-10 | 杭州宏杉科技有限公司 | Management method and management device of network storage system |
| CN102158538A (en) * | 2011-02-18 | 2011-08-17 | 杭州宏杉科技有限公司 | Management method and device of network storage system |
| CN102164165A (en) * | 2011-02-18 | 2011-08-24 | 杭州宏杉科技有限公司 | Management method and device for network storage system |
| CN102982182A (en) * | 2012-12-21 | 2013-03-20 | 浙江宇视科技有限公司 | Data storage planning method and device |
| CN103034592A (en) * | 2012-12-05 | 2013-04-10 | 华为技术有限公司 | Data processing method and device |
| CN103186350A (en) * | 2011-12-31 | 2013-07-03 | 北京快网科技有限公司 | Hybrid storage system and hot spot data block migration method |
| CN103218173A (en) * | 2013-03-27 | 2013-07-24 | 华为技术有限公司 | Method and device for storage control |
| CN104020964A (en) * | 2014-06-18 | 2014-09-03 | 浪潮集团有限公司 | Embedded video storage method |
| CN107122136A (en) * | 2017-04-25 | 2017-09-01 | 浙江宇视科技有限公司 | A kind of capacity acquiring method and device |
| CN107168793A (en) * | 2017-05-10 | 2017-09-15 | 郑州云海信息技术有限公司 | A kind of dynamic current method of virtual memory space |
| CN107800730A (en) * | 2016-08-30 | 2018-03-13 | 阿里巴巴集团控股有限公司 | The expansion method and device of a kind of virtual disk |
| CN108121600A (en) * | 2016-11-30 | 2018-06-05 | 中兴通讯股份有限公司 | Disk array controller, input and output I/O data processing method and processing device |
| WO2022028208A1 (en) * | 2020-08-07 | 2022-02-10 | 华为技术有限公司 | Redundant array of independent disks card, command processing method, and storage device and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6173377B1 (en) * | 1993-04-23 | 2001-01-09 | Emc Corporation | Remote data mirroring |
| CN100349139C (en) * | 2004-02-19 | 2007-11-14 | 华为技术有限公司 | Method for writing data on source film in pistolgraph process |
| CN101620569A (en) * | 2008-07-03 | 2010-01-06 | 英业达股份有限公司 | Expansion method of logical volume storage space |
| CN101840308A (en) * | 2009-10-28 | 2010-09-22 | 创新科存储技术有限公司 | Hierarchical memory system and logical volume management method thereof |
-
2010
- 2010-10-15 CN CN2010105080409A patent/CN101976181A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6173377B1 (en) * | 1993-04-23 | 2001-01-09 | Emc Corporation | Remote data mirroring |
| CN100349139C (en) * | 2004-02-19 | 2007-11-14 | 华为技术有限公司 | Method for writing data on source film in pistolgraph process |
| CN101620569A (en) * | 2008-07-03 | 2010-01-06 | 英业达股份有限公司 | Expansion method of logical volume storage space |
| CN101840308A (en) * | 2009-10-28 | 2010-09-22 | 创新科存储技术有限公司 | Hierarchical memory system and logical volume management method thereof |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102164165B (en) * | 2011-02-18 | 2013-06-12 | 杭州宏杉科技有限公司 | Management method and device for network storage system |
| CN102158538B (en) * | 2011-02-18 | 2013-06-12 | 杭州宏杉科技有限公司 | Management method and device of network storage system |
| CN102158538A (en) * | 2011-02-18 | 2011-08-17 | 杭州宏杉科技有限公司 | Management method and device of network storage system |
| CN102164165A (en) * | 2011-02-18 | 2011-08-24 | 杭州宏杉科技有限公司 | Management method and device for network storage system |
| CN102123176B (en) * | 2011-03-17 | 2015-05-06 | 杭州宏杉科技有限公司 | Space distribution and management method and device for network storage system |
| CN102123176A (en) * | 2011-03-17 | 2011-07-13 | 杭州宏杉科技有限公司 | Space distribution and management method and device for network storage system |
| CN102147714A (en) * | 2011-03-17 | 2011-08-10 | 杭州宏杉科技有限公司 | Management method and management device of network storage system |
| CN102147714B (en) * | 2011-03-17 | 2015-12-16 | 杭州宏杉科技有限公司 | A kind of management method of network store system and device |
| CN103186350B (en) * | 2011-12-31 | 2016-03-30 | 北京快网科技有限公司 | The moving method of mixing storage system and hot spot data block |
| CN103186350A (en) * | 2011-12-31 | 2013-07-03 | 北京快网科技有限公司 | Hybrid storage system and hot spot data block migration method |
| CN103034592A (en) * | 2012-12-05 | 2013-04-10 | 华为技术有限公司 | Data processing method and device |
| WO2014094568A1 (en) * | 2012-12-21 | 2014-06-26 | 浙江宇视科技有限公司 | Data storage planning method and device |
| CN102982182A (en) * | 2012-12-21 | 2013-03-20 | 浙江宇视科技有限公司 | Data storage planning method and device |
| CN102982182B (en) * | 2012-12-21 | 2017-02-08 | 浙江宇视科技有限公司 | Data storage planning method and device |
| US9852168B2 (en) | 2012-12-21 | 2017-12-26 | Zhejiang Uniview Technologies Co., Ltd | Data storage planning |
| CN103218173A (en) * | 2013-03-27 | 2013-07-24 | 华为技术有限公司 | Method and device for storage control |
| CN103218173B (en) * | 2013-03-27 | 2016-03-09 | 华为技术有限公司 | Storage controlling method and device |
| CN104020964A (en) * | 2014-06-18 | 2014-09-03 | 浪潮集团有限公司 | Embedded video storage method |
| CN107800730A (en) * | 2016-08-30 | 2018-03-13 | 阿里巴巴集团控股有限公司 | The expansion method and device of a kind of virtual disk |
| CN107800730B (en) * | 2016-08-30 | 2021-01-29 | 阿里巴巴集团控股有限公司 | Capacity expansion method and device for virtual disk |
| CN108121600A (en) * | 2016-11-30 | 2018-06-05 | 中兴通讯股份有限公司 | Disk array controller, input and output I/O data processing method and processing device |
| WO2018099241A1 (en) * | 2016-11-30 | 2018-06-07 | 中兴通讯股份有限公司 | Disk array controller, and data processing method and device thereof |
| CN108121600B (en) * | 2016-11-30 | 2022-11-29 | 中兴通讯股份有限公司 | Disk array controller, input/output (IO) data processing method and device |
| CN107122136A (en) * | 2017-04-25 | 2017-09-01 | 浙江宇视科技有限公司 | A kind of capacity acquiring method and device |
| CN107168793A (en) * | 2017-05-10 | 2017-09-15 | 郑州云海信息技术有限公司 | A kind of dynamic current method of virtual memory space |
| WO2022028208A1 (en) * | 2020-08-07 | 2022-02-10 | 华为技术有限公司 | Redundant array of independent disks card, command processing method, and storage device and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101976181A (en) | Management method and device of storage resources | |
| CN106687911B (en) | Online data movement without compromising data integrity | |
| US9600181B2 (en) | Live configurable storage | |
| US6895467B2 (en) | System and method for atomizing storage | |
| CN100419664C (en) | Incremental backup operations in storage networks | |
| US7478177B2 (en) | System and method for automatic reassignment of shared storage on blade replacement | |
| US8566550B2 (en) | Application and tier configuration management in dynamic page reallocation storage system | |
| CN100416508C (en) | Copy operations in storage networks | |
| JP5944587B2 (en) | Computer system and control method | |
| CN107092442B (en) | Storage system resource allocation method and device | |
| US20170177224A1 (en) | Dynamic storage transitions employing tiered range volumes | |
| US20210173579A1 (en) | Data migration method and apparatus | |
| JP2017091548A (en) | Dispersion storage system, and dispersion storage method, and dispersion facility | |
| CN102221982B (en) | Method and system for implementing deletion of repeated data on block-level virtual storage equipment | |
| CN101840308A (en) | Hierarchical memory system and logical volume management method thereof | |
| CN101223498A (en) | Virtualisation engine and method, system, and computer program product for managing the storage of data | |
| CN104395904A (en) | Efficient data object storage and retrieval | |
| CN105657066A (en) | Load rebalance method and device used for storage system | |
| CN103946846A (en) | Use of virtual drive as hot spare for RAID group | |
| CN103064765A (en) | Method and device for data recovery and cluster storage system | |
| US20120278560A1 (en) | Pre-fetching in a storage system that maintains a mapping tree | |
| CN101147118A (en) | Methods and apparatus for reconfiguring a storage system | |
| CN103761059A (en) | Multi-disk storage method and system for mass data management | |
| JP2016118821A (en) | Storage management device, storage management method and storage management program | |
| CN101997919B (en) | Storage resource management method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20110216 |
|
| RJ01 | Rejection of invention patent application after publication |