CN106951375A - The method and device of snapped volume is deleted within the storage system - Google Patents
The method and device of snapped volume is deleted within the storage system Download PDFInfo
- Publication number
- CN106951375A CN106951375A CN201610048363.1A CN201610048363A CN106951375A CN 106951375 A CN106951375 A CN 106951375A CN 201610048363 A CN201610048363 A CN 201610048363A CN 106951375 A CN106951375 A CN 106951375A
- Authority
- CN
- China
- Prior art keywords
- volume
- mapping set
- snapped
- snapped volume
- mapping
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F12/00—Accessing, addressing or allocating within memory systems or architectures
- G06F12/02—Addressing or allocation; Relocation
- G06F12/0223—User address space allocation, e.g. contiguous or non contiguous base addressing
- G06F12/023—Free address space management
- G06F12/0238—Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory
- G06F12/0246—Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory in block erasable memory, e.g. flash memory
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/14—Error detection or correction of the data by redundancy in operation
- G06F11/1402—Saving, restoring, recovering or retrying
- G06F11/1446—Point-in-time backing up or restoration of persistent data
- G06F11/1448—Management of the data involved in backup or backup restore
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0646—Horizontal data movement in storage systems, i.e. moving data in between storage devices or systems
- G06F3/0647—Migration mechanisms
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Human Computer Interaction (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of device for deleting snapped volume within the storage system, the storage system includes the mapping set of the logical address of the upper adjacent volume of the mapping set and record of the logical address of record snapped volume and the mapping relations of storage location and the mapping relations of storage location, and described device includes：For the module for the delete command for receiving specified snapped volume；For by the element in the mapping set of specified snapped volume, being merged into the module in the mapping set of the upper adjacent volume；The module that the operation of the specified snapped volume is completed is deleted for feeding back；Wherein, the upper adjacent volume creation time is later than and close to the creation time of the specified snapped volume.The present invention be not it is each individually go to change map metadata, but use backstage migrating data, and polymerize newly-generated mapping data into the mode of a bulk, scale-up problem write with avoid solid state hard disc.
Description
Technical field
Deposited the present invention relates to the snapping technique field in Flash storage system, more particularly to one kind
The method and device of snapped volume is deleted in storage system.
Background technology
Snapshot realizes that technology includes Copy on write (COW) and redirects (ROW) when writing.
As shown in figure 1, Copy on write (COW) situation is：It is first when writing new data A ' toward source book
First legacy data A is read from source book, A is secondly write into snapped volume correspondence position, finally will
A ' is write in source book, and so final new data A ' is in source book, and legacy data A is used in snapped volume
Family from source book read data when, read new data A '.(ROW) situation is redirected when writing is：
When writing new data A ' toward source book, A ' is directly write into snapped volume correspondence position, while in order to tie up
The data consistency of source book is protected, it is necessary to the map information of source book be changed, so that source book is known newly
Data A ' in snapped volume, so final new data A ' in snapped volume, legacy data A in source book,
User from source book read data when, still read new data A '.
So as to draw a conclusion：The each write request of Copy on write (COW) technology can bring one
Read and two write operations, and (ROW) technology each write request is redirected when writing and only produces one
Individual write operation.Data on Copy on write (COW) technology, source book are in physical distribution
Continuously, (ROW) technology is redirected when writing, each write operation is in physical distribution
Discrete；All it is new data on Copy on write (COW) technology, source book, and is reset when writing
New data is safeguarded together to (ROW) technology source book and snapped volume.When so deleting snapshot,
(ROW) technology is redirected to safeguard the data consistency of source book, during for writing, it is necessary to enter
Row data are rolled onto the migration of source book from snapshot, and for Copy on write (COW) technology, no
Need migrating data.
The applicable snapping technique of traditional mechanical hard disk storage system：Because traditional mechanical hard disk is deposited
Storage system is by the way of machinery tracking, and data distribution is more continuous, and its reading performance is higher.Root
According to summary above, it is known that Copy on write (COW) algorithm can be provided in physical distribution
Preferable continuity, therefore more applicable Copy on write in traditional mechanical hard disk storage system
(COW) algorithm.
The storage mediums such as flash memory (Flash) are due to can preferably support random read-write, therefore
It is more applicable that (ROW) algorithm is redirected when writing.
The content of the invention
On the one hand Flash storage system can be compared using (ROW) algorithm is redirected when writing
Preferably support write performance (compared with COW, operation is reduced, without extra read-write)；
On the other hand due to the characteristic of flash media, there is preferable reading performance under physical accidental distribution.
But ROW algorithms need, when deleting snapshot, the new data of snapshot volume maintenance to be moved into source
Volume.
The map information method of prior art typically use modification source book, but this method meeting
The problem of bringing new.The each map information smaller (several bytes~a few crossed joints) of one side,
The problem of another aspect solid state hard disc medium has write-in amplification in itself.Both combines and led
Substantial amounts of write-in scale-up problem on solid state hard disc when causing to delete snapped volume.
The logical address that the map information of volume is used to record belonging to the volume is corresponding with storage location
Relation.The association that map information is additionally operable to record between source book and one or more snapshots is closed
System.
According to an aspect of the invention, there is provided a kind of delete snapped volume within the storage system
Method, the storage system includes the mapping of the logical address and storage location of record snapped volume
The mapping relations of mapping set and the record upper adjacent logical address rolled up and storage location of relation
Mapping set, methods described includes：Receive the delete command for specifying snapped volume；By specified snapshot
In element in the mapping set of volume, the mapping set for being merged into the upper adjacent volume；Feedback deletes institute
The operation for stating specified snapped volume is completed；Wherein, the upper adjacent volume creation time is later than and close to described
Specify the creation time of snapped volume.
According to an embodiment of the invention, wherein by the member in the mapping set of specified snapped volume
In element, the mapping set for being merged into the upper adjacent volume, including：Mapping ensemblen for specifying snapped volume
Each element in conjunction, judges its logical address whether in the mapping set of upper adjacent volume, if institute
The logical address for stating the element in the mapping set of specified snapped volume is not present in the upper adjacent volume
In mapping set, then the element in the mapping set of the specified snapped volume is added to the upper neighbour
In the mapping set of volume.
According to an embodiment of the invention, further comprise：Described specify is deleted in feedback
Before the operation of snapped volume is completed, record represents the day deleted and specify snapped volume in NVDIMM
Will.
According to an embodiment of the invention, the storage system includes data block mapping set,
The whether effective information of data block of data block mapping set record storage position, methods described enters one
Step includes：If the logical address of the first element in the mapping set of the specified snapped volume is present in
In the mapping set of the upper adjacent volume, then in data block mapping set the element of mark first storage
The data block of position is invalid.
According to an embodiment of the invention, further comprise：Delete the specified snapped volume
Mapping set.
According to an embodiment of the invention, wherein, the storage medium in the storage system
For nonvolatile memory.
According to an embodiment of the invention, further comprise：According to data block mapping
Set performs garbage collection operation.
According to an embodiment of the invention, wherein, perform rubbish according to data block mapping set
Rubbish reclaimer operation includes：It is the first data block to obtain mark in data block mapping set；
Judge fast indicated by the volume identification data (i.e. volume identifier) corresponding with the first data block
It whether there is according to volume, wherein, if the volume identification data institute corresponding with first data block
The snapped volume of instruction is not present, and is updated and the first data with the volume identification data of the upper adjacent volume
The corresponding volume identification data of block, and by and by corresponding with the first data block after renewal
Volume identification data write NVDIMM；If the volume identification number corresponding with first data block
Exist according to indicated snapped volume, the volume identification data corresponding with the first data block are write
NVDIMM。
According to an embodiment of the invention, further comprise, will write NVDIMM's
Data aggregate simultaneously writes nonvolatile memory.
According to an embodiment of the invention, further comprise, update and the first data block phase
The storage location of the element of the mapping set of association.
According to an embodiment of the invention, wherein, the storage system further comprises：
Record the mapping set of the logical address of adjacent volume and the mapping relations of storage location, the side
Method further comprises, the access pointer of the adjacent mapping set rolled up is pointed to the mapping ensemblen of lower adjacent volume
Close, wherein, the lower adjacent volume creation time earlier than and close to the specified snapped volume establishment when
Between.
According to the second aspect of the invention, additionally provide one kind and delete snapped volume within the storage system
Device, the storage system includes the record logical address of snapped volume and reflecting for storage location
Penetrate the logical address and the mapping relations of storage location of the upper adjacent volume of mapping set and record of relation
Mapping set, described device includes：For the module for the delete command for receiving specified snapped volume；
For by the element in the mapping set of specified snapped volume, being merged into the mapping set of the upper adjacent volume
In module；The module that the operation of the specified snapped volume is completed is deleted for feeding back；Wherein, institute
Adjacent volume creation time is stated to be later than and close to the creation time of the specified snapped volume.
According to the third aspect of the invention we, additionally provide one kind and create snapped volume within the storage system
Method, the storage system include record source book logical address and storage location mapping
First mapping set of relation, methods described includes：Receive the instruction for creating the first snapped volume；
By logical address and storage location of first mapping set labeled as the first snapped volume of record
Mapping relations mapping set；The second empty mapping set is created, record source book is used as
Second mapping set of the mapping relations of logical address space and storage location；Feedback creates the
The operation of one snapped volume is completed.
An embodiment according to the third aspect of the invention we, also includes：In response to receiving more
First data recorded the first storage location by the request of the first data of new first logical address,
And reflecting for the first logical address and first time storage location is recorded in second mapping set
Penetrate relation.
An embodiment according to the third aspect of the invention we, also includes：In response to accessing first
The request of logical address, if being found in second mapping set with first logical address
The first element, according to the first element record storage location obtain data；And if described
The element with first logical address is not found in two mapping sets, then in the described first mapping
The second element with first logical address is searched in set, and according to second element note
The storage location of record obtains data.
There is provided a kind of computer journey comprising computer program code according to the fourth aspect of the invention
Sequence, when being loaded into computer system and performing on the computer systems, the computer program generation
Code makes the computer system perform the deletion within the storage system provided according to an aspect of the present invention
The method of snapped volume or the establishment within the storage system provided according to a third aspect of the present invention are fast
According to the method for volume.
According to the fifth aspect of the invention there is provided a kind of program including program code, when being loaded into
Storage device and when performing on a storage device, the meter program code performs the storage device
There is provided according to an aspect of the present invention within the storage system delete snapped volume method or according to
The method for creating snapped volume within the storage system that third aspect present invention is provided.
The present invention delete snapped volume when, be not it is each individually go change map metadata, but use
Backstage migrating data, and it polymerize newly-generated mapping data into the mode of a bulk, to avoid
Solid state hard disc writes scale-up problem.
Brief description of the drawings
By reading the detailed description of hereafter preferred embodiment, various other advantage and benefit pair
It will be clear understanding in those of ordinary skill in the art.Accompanying drawing is only used for showing preferred embodiment
Purpose, and be not considered as limitation of the present invention.And in whole accompanying drawing, use identical
Reference symbol represents identical part.Wherein in the accompanying drawings, the alphabetic flag after reference numeral refers to
Show multiple identical parts, when referring to these parts, its last alphabetic flag will be omitted.
In accompanying drawing：
Fig. 1 redirects (ROW) for Copy on write (COW) technology in the prior art with when writing
The schematic diagram of technology；
Fig. 2 shows the structural representation of the snapshot chain according to an embodiment of the invention；
Fig. 3 shows the structural representation of the mapping set according to an embodiment of the invention；
Fig. 4 A show the establishment within the storage system according to an embodiment of the invention
The schematic diagram of snapped volume；
Fig. 4 B show in the storage system according to an embodiment of the invention and created soon
According to the flow chart of the method for volume；
Fig. 5 A show the schematic diagram of the deletion snapped volume according to an embodiment of the invention；
Fig. 5 B show the stream of the method for the deletion snapped volume according to an embodiment of the invention
Cheng Tu；
Fig. 6 shows deletion snapshot within the storage system according to another implementation of the invention
The flow chart of the method for volume；
Fig. 7 A have gone out in the nonvolatile memory (NVM) according to one embodiment of the present invention
Data tissue schematic diagram；
Fig. 7 B show the data in the nonvolatile memory according to one embodiment of the present invention
The schematic diagram of tissue；
Fig. 8, which is shown in power up according to another aspect of the present invention, recovers snapshot chain
And its flow chart of the method for mapping set；
Fig. 9 A show the signal of the execution garbage collection operation according to one embodiment of the present invention
Figure；
Fig. 9 B show the flow of the method for the garbage reclamation according to one embodiment of the present invention
Figure；
Figure 10 shows deletion snapshot within the storage system according to another aspect of the present invention
The structural representation of the device of volume.
In the accompanying drawings, same or similar element is referred to using same or similar label.
Embodiment
The invention will be further described with specific embodiment below in conjunction with the accompanying drawings.
Fig. 2 shows the structural representation of the snapshot chain according to an embodiment of the invention.
As shown in Fig. 2 storage system includes snapshot chain.Snapshot chain includes source book 210, snapshot
Roll up Tn220th, snapped volume Tn-1230 ... snapped volume T1240 and snapped volume T0250.With regard to the time
For order, snapped volume TnNewly in snapped volume Tn-1, snapped volume T0Data are oldest.That is, snapped volume
T0It is created earliest, and in Fig. 2 multiple snapped volumes, snapped volume TnIt is created the latest.In figure
The chain type that arrows show data is accessed in 2 is sequentially.In an embodiment according to the present invention, snapshot chain
Upper to include multiple snapped volumes, the quantity to snapped volume is not limited, and n can be positive integer.
When some data is read from source book 210, if will without record in the mapping set of source book 210
The logical address of access, then the order shown in Fig. 2 arrow, accesses snapped volume Tn, attempt
Snapped volume TnMapping set in search the logical address to be accessed.If snapped volume TnDo not have still
The logical address to be accessed, then access Tn-1.On the snapshot chain for accessing Fig. 2 successively in this way
Each node.If if final T0On the also not no data, then tell user or the requesting party data not
In the presence of.In one example, mapping set access pointer is provided for each volume, to indicate Fig. 2
Arrow indicated by access order.
Fig. 3 shows the structural representation of the mapping set according to an embodiment of the invention.
As shown in figure 3, being each volume (source book in Fig. 2 in an embodiment according to the present invention
/ snapped volume) mapping set is provided.The logical address of the volume is provided in the element of mapping set with depositing
The corresponding relation that storage space is put.Generally, mapping set is maintained in internal memory so that visited
Quick response can be obtained by asking during mapping set.Partly or entirely also being present in for mapping set is non-
In volatile memory, so as to reduce to the occupancy of Installed System Memory or reduction because the factors such as power down are led
The risk for causing map information to lose.
As shown in figure 3, mapping set according to embodiments of the present invention include map information element 310,
Map information element 320, map information element 330 and map information element 340.In mapping ensemblen
Map information element can have multiple in conjunction.In figure 3, map information element 310 provides logic
Address and storage location, so as to indicate the logical address and storage location of map information element 310
Corresponding relation, such as in the volume, the data storage at logical address L is in storage location P.
In one example, record has logical address in map information element.In another example, logic
Address can be provided by memory address is implicit, and prevent take up the space of mapping information element element.
Storage location in mapping set can be the storage money of physical address or virtualization
The address in source.Mapping set can safeguard volume mapping information using plurality of data structures, such as array,
Chained list, tree, Hash table etc..
Fig. 4 A show the establishment within the storage system according to an embodiment of the invention
The structural representation of snapped volume, Fig. 4 B show depositing according to an embodiment of the invention
The flow chart of the method for snapped volume is created in storage system.
Referring to Fig. 4 A, create before snapshot, under initial situation, snapshot chain 400 includes volume 1
Node 420 and empty (NULL) node 430.1 node 420 is rolled up with mapping set, is used for
Record the logical address of source book and the mapping relations of storage location.Roll up the mapping ensemblen of 1 node 420
Close access pointer and point to empty (NULL) node 430.Source book pointer points to 1 node 420 of volume, with
It is source book to indicate volume 1.When updating the data, the node 420 (being source book based on volume 1) of modification volume 1
Mapping set, pair of updated the data logical address and storage location is recorded in mapping set
It should be related to.
As shown in Figure 4 A and 4 B shown in FIG., when creating a new snapped volume, target is to retain source book
Current state.The current mapping set record of source book is in 1 node 420 is rolled up.In step S450,
New volume (volume 2) node 440 is created, the node 440 of volume 2 has empty mapping set, is used to
Renewal after being created as new source book carrying snapshot to volume.In step S452,2 nodes will be rolled up
440 mapping set access pointer points to the node 420 of snapped volume 1, so that in the mapping ensemblen of volume 2
When the logical address to be accessed is can not find in conjunction, snapped volume is accessed according to mapping set access pointer
1.Snapshot chain 402 in Fig. 5 A shows snapshot chain structure now.Further, will also volume
The mapping set of 1 node 420 be set to it is read-only so that avoid snapshot create during change institute
The snapshot of establishment.And roll up the snapshot that 1 node 420 indicates created storage system.
In step S454, the pointer of source book is pointed into 2 nodes 440 of volume, to show that volume 2 turns into
Current source book, so that when can not find the logical address to be accessed in the mapping set of source book, according to
The node 420 of snapped volume 1 is accessed according to mapping set access pointer.Snapshot chain 406 in Fig. 5 A
Show snapshot chain structure now.
In step S456, the result created as snapshot returns to the conduct of 1 node of snapped volume 420
The instruction of snapped volume.Rolling up 1 node 420 turns into the snapshot of created storage system.Hereafter, exist
When updating the data, the mapping set of the node 440 (being source book based on volume 2) of modification volume 2, in volume
Updated the data logical address is recorded in the mapping set of 2 nodes 440 corresponding with storage location
Relation.
In one example, it is desirable to access the data of source book.Based on source book pointer, 2 nodes are being rolled up
The corresponding relation of logical address and storage location is searched in 440 mapping set, so that in storage position
Put place and access data.Searched logical address is not present in the mapping set of 2 nodes 440 if rolling up,
According to the mapping set access pointer of 2 nodes 440 of volume, in the mapping set for rolling up 1 node 420
The corresponding relation of logical address and storage location is searched, so as to access data at storage location.
In one example, it is desirable to access the data of snapped volume 1.Rolling up the mapping of 1 node 420
The corresponding relation of logical address and storage location is searched in set, so as to access number at storage location
According to.Searched logical address is not present in the mapping set of 1 node 420 if rolling up, according to volume 1
The mapping set access pointer of node 420, finds the mapping set access pointer of 1 node 420 of volume
Empty (NULL) node 430 is pointed to, so that it is determined that the data accessed are not deposited in snapped volume 1
.
In still another embodiment, the structure of snapshot chain is as shown in the 406 of Fig. 4 A.Herein
In the case of, it is desirable to create snapshot.Similar to the method described in figure 4b, new volume (example is created
Such as, snapped volume 3) node and empty mapping set；The mapping set access pointer newly rolled up is referred to
To 2 nodes 440 of volume；The pointer of source book is pointed into new volume node；And the conduct of 2 node 440 will be rolled up
The result that snapshot is created.
According to another embodiment of the invention, after snapped volume 1 is created, also exist
Log in NVDIMM, indicates the establishment to snapped volume 1 in daily record.In NVDIMM
The daily record of middle record can be write with a brush dipped in Chinese ink into NVM.NVDIMM is the storage for computing device
Device, it has the high-speed data of similar DRAM (Dynamic Random Access Memory) concurrently
The data holding ability of access capability and nonvolatile storage, even if occurring unexpected power down,
The data stored in NVDIMM will not still lose.In another example, creating fast
After volume 1, the log also in NVM, without daily record is write into NVDIMM.
The method for deleting snapped volume within the storage system is described in detail below by specific embodiment.
Fig. 5 A show the schematic diagram of the deletion snapped volume according to an embodiment of the invention.
Fig. 5 B show the flow of the method for the deletion snapped volume according to an embodiment of the invention
Figure.
As shown in Figure 5A, snapshot chain 500 includes source book 510, T1The snapped volume T at moment1520、
T0The snapped volume T at moment0522 and sky (NULL) 530.The mapping set of source book 510 is accessed
Pointer points to snapped volume T1520.Snapped volume T1520 mapping set access pointer points to snapped volume
T0522, snapped volume T0522 mapping set access pointer points to empty (NULL) node 530.
As an example, snapped volume T is deleted from snapshot chain 5001520.Show in figure 5b
The method flow diagram of snapped volume is deleted from snapshot chain.
Step S510：Receive specified snapped volume (the i.e. snapped volume T deleted1) order.
Step S520：To delete snapped volume T1, the mapping set access pointer of source book 510 is changed,
It is allowed to point to snapped volume T0522。
In step S520, snapped volume T0522 are and snapped volume T to be deleted1520 in the time
It is upper close to and creation time is earlier than snapped volume T to be deleted1Snapshot.By snapped volume T0Claim
For snapped volume T1Lower adjacent volume.Source book 510 is and snapped volume T to be deleted1520 phases in time
Close to, and creation time is later than snapped volume T to be deleted1520 volume.Source book 510 is referred to as fast
According to volume T1Upper adjacent volume.Thus, in step S520, by snapped volume T to be deleted1Upper adjacent volume
Mapping set access pointer be modified to point to snapped volume T to be deleted1Lower adjacent volume.
If it should be noted that creation time is not present in snapshot chain earlier than snapped volume to be deleted
T1Snapshot, then change source book mapping set access pointer, be allowed to point to empty node (NULL).
And in Fig. 5 A, Fig. 5 B example, source book is and snapped volume T to be deleted1Phase in time
Close to volume, and in source book and snapped volume T1Between be not present other snapshots.In the case,
Deleting snapped volume T1When, the mapping set access pointer of source book 510 is changed, is allowed to point to soon
According to volume T0.So as in snapshot chain 502 so that the mapping set access pointer of source book 510 is pointed to
Snapped volume T0522。
Step S530：By snapped volume T1520 mapping set is merged into the mapping set of source book 510.
In snapshot chain 500, source book 510 is snapped volume T1520 upper adjacent volume.Thus deleting
Snapped volume T1When 520, by snapped volume T1520 mapping set is merged into the mapping ensemblen of source book 510
Close.
In step S530, to merge snapped volume T1520 mapping set reflects with source book 510
Set is penetrated, for snapped volume T1Being arbitrarily not present in the mapping set of source book 510 in 520
Element, by snapped volume T1The element in 520 is added in the mapping set of source book 510.
Merging snapped volume T1520 with the mapping set of source book 510 during, snapped volume T1520
Mapping set in element exist in the mapping set of source book 510, refer to for snapped volume
T1The logical address of the element in 520 mapping set, is deposited in the mapping set of source book 510
There is the element of the logical address in record.And for snapped volume T1It is each in 520 mapping set
Element, if an element exists in the mapping set of source book 510, does not insert source book by the element
510 mapping set.
Step S540：Feedback deletes snapped volume T1520 operation is completed.
Further, snapped volume T is also destroyed1520 and its mapping set in internal memory, obtain as
Snapshot chain 502 shown in Fig. 5 A.In snapshot chain 502, the mapping set of source book 510 is visited
Ask that pointer points to snapped volume T0522, and snapped volume T0522 volume mapping message reference pointer is pointed to
Empty node (NULL) 530.So far, to snapped volume T1520 deletion action is completed.
In another example, if in snapped volume T1After 520 create, snapped volume T is further created3,
Snapped volume T3With snapped volume T1In time close to and rear.By snapped volume T3Referred to as snapped volume
T1Upper adjacent volume.So deleting snapped volume T1When 520, modification and snapped volume T to be deleted1520
In time close to and creation time is later than snapped volume T to be deleted1520 snapped volume T3Mapping
Gather access pointer, be allowed to point to and snapped volume T to be deleted1520 in time close to and create when
Between earlier than snapped volume T to be deleted1520 snapped volume T0522.Deleting snapped volume T1When 520,
By snapped volume T1520 mapping set is merged into snapped volume T3Mapping set.
Referring to Fig. 5 A snapshot chain 502, snapped volume T is being deleted1After 520, if accessing source book 510
Data, then first look for the mapping set of source book 510, search logical address and storage location
Corresponding relation, so as to access data at storage location.If not deposited in the mapping set of source book 510
In the logical address searched, according to the mapping set access pointer of source book 510, snapped volume T0522
Mapping set in search the corresponding relation of logical address and storage location so that at storage location
Access data.
In one example, it is desirable to access snapped volume T0522 data.In snapped volume T0522
The corresponding relation of logical address and storage location is searched in mapping set, so as to be visited at storage location
Ask data.If snapped volume T0Searched logical address, foundation are not present in 522 mapping set
Snapped volume T0522 mapping set access pointer, finds snapped volume T0522 mapping set is accessed
Pointer points to empty (NULL) node 530, so that it is determined that the data accessed are in snapped volume T0522
In be not present.
Fig. 6 shows the deletion snapshot within the storage system according to another embodiment of the invention
The flow chart of the method for volume.
As an example, figure 6 illustrates snapshot is deleted from snapshot chain 500 (referring to Fig. 5 A)
Roll up T1520 (referring to Fig. 5 A) method flow diagram.
Step S610：Receive specified snapped volume (the i.e. snapped volume T deleted1) order.
Step S620：For snapped volume T1Each element in 520 mapping set, judges fast
According to volume T1The logical address of the element of 520 mapping set whether there is in the mapping set of upper adjacent volume
In.Referring to Fig. 5 A example, snapped volume T1520 upper adjacent volume is source book 510.
Step S630：For snapped volume T1The mapping for being arbitrarily not present in adjacent volume in 520
Element in set, by snapped volume T1The element in 520 is added to the mapping set of upper adjacent volume
In.
Referring to Fig. 5 A example, snapped volume T1520 upper adjacent volume is source book 510.It is fast merging
According to volume T1520 with the mapping set of source book 510 during, snapped volume T1520 mapping set
In element exist in the mapping set of source book 510, refer to for snapped volume T1520 mapping
The logical address of the element in set, has this to patrol in the mapping set of source book 510 in the presence of record
Collect the element of address.And for snapped volume T1Each element in 520 mapping set, if unitary
Element exists in the mapping set of source book 510, then the element is not inserted to the mapping ensemblen of source book 510
Close.
In further example, data block mapping set is also provided in storage system, using record
The whether effective information of the data block of storage location.For snapped volume T1In 520 mapping set
Each element, if the logical address of an element exists in the mapping set of source book 510, in number
Data block according to the storage location that the element is marked in block mapping set is invalid.
Step S632：To delete snapped volume T1, the mapping set of adjacent volume (source book 510) in modification
Access pointer, is allowed to point to snapped volume T0522 mapping set.
In step S632, snapped volume T0It is and snapped volume T to be deleted1In time close to,
And creation time is earlier than snapped volume T to be deleted1Snapshot.If it should be noted that in snapshot
Creation time is not present in chain earlier than snapped volume T to be deleted1Snapshot, then change source book mapping ensemblen
Access pointer is closed, is allowed to point to empty node (NULL).And in Fig. 5 A, Fig. 5 B example
In, source book is that creation time is later than and close to snapped volume T to be deleted1Volume, in source book and snapshot
Roll up T1Between be not present other snapshots.In the case, snapped volume T is being deleted1When, change source
The mapping set access pointer of volume 510, is allowed to point to snapped volume T0.So as in snapshot chain 502
In so that the mapping set access pointer of source book 510 points to snapped volume T0522。
Step S634, represents in NVDIMM records and deletes snapped volume T1520 daily record.
Deleting snapped volume T1After 520, the log also in NVDIMM refers in daily record
Show to snapped volume T1520 deletion.The daily record recorded in NVDIMM can be write with a brush dipped in Chinese ink to non-easy
In the property lost memory (NVM).Optionally, snapped volume T is being deleted1After 520, also non-volatile
Log in property memory (NVM), without daily record is write into NVDIMM.
The log in NVDIMM or NVM, is revised as soon with tackling in snapshot chain 500
After chain 502 (referring to Fig. 5 A), the mapping relations of data and snapped volume in physical storage medium
Before being updated, storage system delay machine (such as power down) situations such as.The introduced expense of log compared with
It is small, so that storage system can embody the shorter response time when deleting snapshot.
In further example, the garbage reclamation module of storage system is indicated, to snapped volume T1520
Data in corresponding physical storage medium are migrated.Record has data in physical storage medium
And the volume information belonging to data.Will be to this detailed description below in association with figure xx.Work as deletion
, it is necessary to change the volume information belonging to the data recorded in physical medium during snapped volume.According to this
During the snapped volume of the embodiment of invention is deleted, without the data to being recorded in physical storage medium
Affiliated volume information is modified, you can feedback snapshot deletion action is completed.For this purpose, complete
After into the merging of deleted snapped volume and the mapping set of source book, garbage reclamation module is notified, by
Garbage reclamation module handles together the modification to the volume information of data block in garbage collection procedure.
The process that Data Migration is carried out in garbage reclamation will be described in detail hereinafter.
Step S640：Feedback deletes snapped volume T1520 operation is completed.
Fig. 7 A show the nonvolatile memory (NVM) according to one embodiment of the present invention
In data tissue schematic diagram.
Nonvolatile memory is such as can be SSD, flash memory, NVDIMM.In NVM
In, basic unit of storage is used as using memory block or page.Basic unit of storage can be NVM thing
Manage page, physical block or superblock.Optionally, multiple Physical Page of data block including NVM, many
Individual physical block or multiple superblocks.In fig. 7a it is shown that data block 710, data block 720,
Data block 730 and data block 740.NVM each data block includes：Data, the data
The volume identifier of volume belonging to logical address, the data.The logical address shown in fig. 7
It is identical logical address with the logical address that is shown in Fig. 3, thus from the NVM shown in Fig. 7 A
In can rebuild mapping set as shown in Figure 3.Volume identifier is used for the unique mark data block institute
The volume (source book or snapped volume) belonged to, for example using sequence number or general unique identifier UUID as
Volume identifier.
Referring back to Fig. 5 A, Fig. 5 B, in step S530, merging snapped volume T1520 with
During the mapping set of source book 510, for snapped volume T1It is each in 520 mapping set
Element, if an element is not present in the mapping set of source book 510, source book is inserted by the element
510 mapping set.In the case, belonging to the data of the storage location indicated by the element
Volume changes.In Fig. 5 A example, the volume belonging to the data of the storage location indicated by the element
From belonging to originally in snapped volume T1520, it is changed to belong to source book 510.The element thus can also be changed
Volume identifier in the data block of indicated storage location (referring to Fig. 7 A).Further
In embodiment, the modification of data block is recorded in NVDIMM, and in NVDIMM
It has polymerize after multiple modification operations to NVM, has been then written in NVM.
According to another embodiment of the invention, version information is also included in data block.It can make
With the timestamp for recording the time that the data block is generated, it is generated for recording the data block
The serial number of order etc. be used as version information.It is general by data block in addition to data division in
Hold (information such as logical address, volume identifier) and be referred to as metadata.
In still another embodiment, data are separately stored with metadata.For example,
In NVM, a basic unit of storage is stored data in, and the metadata of the data is stored
In another basic unit of storage.The example of metadata includes logical address, volume identifier, version
Information identifier etc..
Fig. 7 B show the data in the nonvolatile memory according to one embodiment of the present invention
The schematic diagram of tissue.
Fig. 7 B shows are according to the storage resource organizer on the NVM of further embodiment of this invention
Formula.In an embodiment according to the present invention, the storage resource that storage device is provided is organized as one
Or multiple storage objects.In one example, by SSD storage resource (for example, block) pond
Change, and constitute one or more storage objects.Storage object is otherwise known as container.Write data into
Storage device or from storage device read data when, it is substantially single using storage object as access
Member.The quantity of storage object is determined by the capacity of storage system.
Referring to Fig. 7 B, disk 0, disk 1, disk 2 and the storage device that disk 3 is such as SSD.At it
In his example, disk 0, disk 1, disk 2 and disk 3 can also be mechanical hard disks.As an example, disk 0
Address space is provided as storage device.The address space of disk 0 is divided into multiple bulks (chunk),
Such as bulk 01410, bulk 11412 ... bulk n in Fig. 7 B.Bulk may include multiple
Block.Similarly, each address space of disk 1, disk 2 and disk 3 is divided into multiple bulks, example
As bulk 1420 is provided with bulk 1422 by disk 3.
In Fig. 7 B example, formed by disk 0, disk 1, disk 2 and multiple bulks that disk 3 is provided
Memory resource pool.From memory resource pool by several bulk tissues be storage object (in Fig. 7 B
In be shown as container).For example, container 0 include the bulk 0 (bulk 0-1) from disk 0,
Bulk 1 (bulk 1-1) and the bulk 0 (bulk 2-0) from disk 2, container from disk 1
1 includes the bulk 2 (bulk 0-2) from disk 0 and the bulk 2 (bulk 1-2) from disk 1.
Storage system according to embodiments of the present invention provides multiple containers, and each container may include identical or not
With the bulk of quantity.In container, such as RAID data protection schemes can be provided.For example,
In container 0, bulk 0-0 and bulk 1-1 is used to store user data, and bulk 2-0 is used for
The storage verification data corresponding with bulk 0-0, bulk 1-1.In container 0, verification data
Storage location is not limited to bulk 2-0, can also be stored in bulk 0-0 or bulk 1-1.As another
Individual example, the bulk 0-2 and bulk 1-2 of container 1 are used to store user data.Still conduct
One example, the multiple bulks (bulk 0-0, bulk 1-1 and bulk 2-0) for constituting container 0 are come
From different disks, so that the access to container 0 will be distributed in disk 0, disk 1 and disk 2.With this side
Formula, adds the concurrency of the access to disk, storage performance is lifted.Art technology
Personnel would also recognize that the disk in storage system can have different capacity, and disk may be replaced,
So as to which the bulk in container may be from identical disk.
According to one embodiment of present invention, the logical address of mapping set (referring to Fig. 3)
For indexing the deviant in bulk and/or bulk in container.In a reservoir, go back and number of users
According to accordingly storing metadata.The volume identifier for the volume that metadata belongs to including corresponding data,
Version information and/or logical address., can also be because bulk has been logical layer storage object
Logical address is not stored in the data or metadata of bulk.
Referring back to Fig. 5 A and Fig. 5 B, in step 530, merging snapped volume T1520 and source book
During 510 mapping set, for snapped volume T1Each element in 520 mapping set,
If an element is not present in the mapping set of source book 510, the element is inserted into source book 510
Mapping set.In the case, the volume belonging to the data of the storage location indicated by the element changes.
In Fig. 5 A example, volume belonging to the data of the storage location indicated by the element from belong to originally in
Snapped volume T1520, it is changed to belong to source book 510.Thus in Fig. 7 B example, also modification should
Volume identifier in the corresponding metadata of the bulk of storage location indicated by element.Further
Embodiment in, the modification of the metadata in Fig. 7 B container will be recorded in NVDIMM,
And polymerize in NVDIMM after multiple renewal operations to metadata, it is then written to NVM
In.
In Fig. 7 B example, the example that container is constituted with bulk is illustrated.Can be by other modes
To constitute bulk.For example in another example, bulk is not provided, container corresponds to one section of disk
Memory space, such as one section continuous logical address space.
In an embodiment according to the present invention, it is lifting performance of storage system, data is write to container
When, write or order WriteMode using addition.In other words, from the beginning container can only write tail, and hold
The optional position of device can not be capped write-in.
, it is necessary to rebuild snapshot chain and its mapping set after storage system is re-powered.Fig. 8
For the flow chart of the storage system power up according to the embodiment of the present invention.In step S810, from
In NVDIMM and NVM read volume Operation Log, by the volume Operation Log in NVDIMM with
Volume Operation Log in NVM merges, so that being constructed in internal memory in storage system to be present
One or more volumes, and all snapshot chains as shown in Figure 2 are constructed, to represent source book and one
Relation between individual or multiple snapped volumes.In volume Operation Log, the establishment of volume/snapshot have recorded
With deleting and creating the sequencing with deletion action.In another example, daily record is only existed
In NVM or NVDIMM, when storage system is re-powered, only from NVM or
Volume Operation Log is read in NVDIMM.
In step S820, data and metadata are read from NVM, container and/or NVDIMM,
Therefrom obtain the corresponding relation of data and metadata, the mapping set as shown in Figure 3 for rebuilding,
Wherein each data are with metadata to an element as mapping set to be reconstructed.Metadata bag
Include logical address, volume identifier and/or timestamp.If the volume identifier obtained from metadata is deposited
In being one of one or more volumes for rebuilding in step S810, then by the data and metadata pair
It is added in mapping set.If the volume identifier obtained from metadata is not present in step S810
In multiple volumes of middle reconstruction, then show that corresponding volume has been deleted, corresponding data are invalid datas.
Also record indicates the information of the data invalid in the case.In one example, reflected with data block
Penetrate the validity of the data of set (for example, bitmap) record storage position.In another example,
Garbage reclamation part is given by the information transmission for indicating the data invalid, by garbage reclamation part appropriate
In the case of, reclaim invalid data.In still another example, garbage reclamation part is according to note
The data block mapping set of the validity of data is recorded, the storage location for storing invalid data is returned
Receive, and/or effective data are moved.
In a further embodiment, if the volume identifier obtained from metadata is present in step
In one of one or more volumes rebuild in S810, and presence belongs to being reflected corresponding to identical for the volume
The multiple data and metadata of set entry are penetrated to (for example, multiple numbers with same logical address
According to metadata to), according to the version information in metadata, select newest data and metadata
To being added in mapping set.
By scanning NVM, container and/or NVDIMM, one in storage system is reconstructed
Or the mapping set of multiple volumes, asked so that storage system can be accessed according to mapping set response data
Ask.
As it was previously stated, according to one embodiment of present invention, in the snapped volume and source for completing to be deleted
After the merging of the mapping set of volume, garbage reclamation module is notified, is returned by garbage reclamation module in rubbish
During receipts, the modification to the volume information of data block is handled together.By garbage reclamation part according to note
The data block mapping set of the validity of data is recorded, the storage location for storing invalid data is returned
Receive, and/or effective data are moved.
Fig. 9 A show the signal of the execution garbage collection operation according to one embodiment of the present invention
Figure.As shown in Figure 9 A, in storage region 1, Data1 is valid data, and Data2 is nothing
Imitate data.Metadata1 is metadata corresponding with Data1, and metadata2 is corresponding with data2
Metadata.In garbage collection procedure, valid data Data1 is moved into storage region 2,
Without migration invalid data Data2.Referring back in Fig. 5 A and Fig. 5 B, when deletion snapped volume
T1When 520, snapped volume T1520 partial data will turn into invalid data, and another part data
Though effectively need to be subordinated to snapped volume T1520 are changed to belong to source book 510.For this purpose,
Need data1 metadata metadata1 being revised as metadata1 '.And due to NVM spy
Point, the modification to metadata metadata1 will be implemented as data1 and metadata1 ' it is written to and deposits
Storage area domain 2.
Garbage collection operation is used to reclaim the storage region occupied by invalid data.Garbage collection operation
It is additionally operable to during valid data are migrated, update the data the volume identifier corresponding to block.In root
According in embodiments of the invention, cause occur invalid number in system because deleting snapped volume or other reasonses
According to rear, or when volume identifier corresponding to data block and inconsistent volume belonging to the data block, and
Non-immediate performs garbage collection operation.The opportunity of garbage reclamation can be performed according to a variety of strategy decisions.
For example, when the invalid data in storage region 1 exceedes certain proportion, performing garbage reclamation；Or
Person when an insufficient storage space exists, performs garbage reclamation；In another example, in response to rubbish
The request of reclaimer operation and be immediately performed garbage reclamation.In still another example, periodically
Garbage reclamation is performed, or garbage reclamation is performed when storage system task load is relatively low.
Fig. 9 B are the flow charts of the method for garbage reclamation (GC) according to embodiments of the present invention.
The method that garbage reclamation (GC) is described with reference to Fig. 9 A.
As shown in Figure 9 B, the method for garbage reclamation (GC) includes：
Step S910：Choose the most storage region 1 of invalid data；
Step S920：The data block mapping set of scanning area 1, obtains having in region 1
Imitate data data1 and/or invalid data data2；
Step S930：Judge that the corresponding volumes of data1 whether there is；
Step S940：If the corresponding volumes of data1 are not present, by the member corresponding with data1
The volume identifier that data modification is rolled up for the upper neighbour of the volume corresponding to record data1；
Step S950：If the corresponding volumes of data1 are present, relative with data1 without changing
The volume identifier recorded in metadata answered；
Step S960：The metadata write-in NVDIMM that will be updated, and in NVDIMM
After the middle data aggregate by renewal is bulk, then NVM is write.
In step S910, object of the most region 1 of invalid data as garbage reclamation is chosen.
One of ordinary skill in the art will realize various ways triggering garbage collection operation or determination
The object of garbage collection operation.
In step S920, scanning obtains storage region 1 corresponding to the bitmap of storage region 1
In valid data (such as data1) and/or invalid data.In another example, use
The validity for the data that chained list, array or other data structures are come in record storage region.
In step S930, judge whether the volume corresponding to valid data data1 is deleted.Its
In, metadata includes volume identifier.By with the metadata corresponding to valid data data1
Metadata1 obtains the volume identifier where valid data data1, judges the volume identifier institute
Corresponding volume (calling volume 1 in the following text) whether there is within the storage system.If being not present, show the volume
1 has been deleted, and in step S940, is found in current snapshot chain (referring to Fig. 2) with being somebody's turn to do
Volume 1 is in time close to and creation time is later than the upper of the volume 1 and adjacent rolls up (calling volume 2 in the following text).
Volume 2 can be snapped volume or source book.In one example, the upper of volume 1 is obtained from daily record
Neighbour's volume (volume 2)., will metadata corresponding with valid data data1 in step S940
Volume identifier in metadata1 is revised as the volume identifier of volume 2.
In step S930, if judging, the volume corresponding to valid data data1 is not deleted,
Without changing the volume identifier (step in the metadata corresponding with valid data data1
S950)。
In step S960, metadata metadata1 ' the write-in NVDIMM that will be updated, and
After being bulk by the data aggregate of renewal in NVDIMM, then NVM is write.So as to keep away
Exempt from the problem of amplification is write in generation.
One of ordinary skill in the art will realize writes NVDIMM's by the metadata of renewal
Step is not essential, and the metadata to be updated can be written directly into NVM.In basis
In the further embodiment of the present invention, when carrying out Data Migration, by valid data data1
Directly write in NVM, and the metadata corresponding with valid data data1 is write
NVDIMM, and it is aggregated into after large-size data, then write NVM.
According to another aspect of the present invention, the present invention also provides one kind and deleted within the storage system
The device of snapped volume, the storage system includes the logical address of snapshot chain and snapped volume with depositing
The mapping set for the mapping relations that storage space is put, as shown in Figure 10, including：For receiving snapshot
The device 1000 of the delete command of specified snapped volume in chain；Mapping set for specifying snapped volume
In each element, for judge its logical address whether the device in the mapping set of upper adjacent volume
2000, wherein, if the logical address of the specified element in the specified snapped volume is not present in institute
In the mapping set for stating adjacent volume, then the specified element in the specified snapped volume is added to described
In the mapping set of upper adjacent volume；The device that the operation of the specified snapped volume is completed is deleted for feeding back
3000；Wherein, the upper adjacent volume creation time is later than and close to the specified snapped volume, the member
Element includes the storage location of data, the logical address of the data and the data.
According to another aspect of the present invention, the present invention also provides a kind of comprising computer program code
Computer program, when being loaded into computer system and performing on the computer systems, the meter
Calculation machine program code makes the computer system perform methods described above.
According to another aspect of the present invention, a kind of program including program code is also provided, works as quilt
When being loaded into storage device and performing on a storage device, the meter program code makes the storage device
Perform methods described above.
The present invention obtains beneficial effect：What is introduced during solution deletion snapshot writes scale-up problem；Reduce
The erasing times of solid state hard disc, and then improve the service life of solid state hard disc medium；Delete fast
During according to volume, shorten the time for performing snapshot deletion action.
It should be understood that the combination of the frame of each frame and block diagram and flow chart of block diagram and flow chart can
To be implemented respectively by the various devices including computer program instructions.These computer program instructions
It can be loaded on all-purpose computer, special-purpose computer or other programmable data control devices to produce
Life device, so that the instruction performed on computer or other programmable data control devices is created
Device for realizing the function of being specified in one or more flow chart box.
These computer program instructions, which can also be stored in, can guide computer or other are programmable
So as to work in a specific way in the computer-readable memory of data control equipment, so as to
Manufactured using the instruction being stored in computer-readable memory including for realizing one or many
The product of the computer-readable instruction of specified function in individual flow chart box.Computer program instructions are also
Can be loaded on computer or other programmable data control devices with cause computer or its
A series of operating procedure is performed on his programmable data control device, so as to produce computer realization
Process, and then the instruction performed on computer or other programmable data control devices provides
The step of for realizing specified function in one or more flow chart box.
Thus, the frame of block diagram and flow chart is supported the combination of the device for performing specified function, used
In the combination for performing the step of specifying function and program instruction means for performing specified function
Combination.It should also be understood that the group of the frame of each frame and block diagram and flow chart of block diagram and flow chart
Function or step, hardware based dedicated computer system realization can be specified by performing by closing, or
Realized by the combination of specialized hardware and computer instruction.
At least a portion of above-mentioned different masses, operation and technology can be performed, by using
Hardware, control device performs firmware instructions, and control device performs software instruction, or and its any
Combination.When being performed using the control device for performing firmware and software instruction, software or firmware refer to
Order can be stored in any computer-readable recording medium, such as disk, CD or other
Storage medium, in RAM or ROM or flash memory, control device, hard disk, light
Disk, disk etc..Similarly, software and firmware instructions can be transferred to user or system,
Included by any known or desired transmission means, for example, in computer readable diskette or
Other portable computer memory mechanisms pass through telecommunication media.Telecommunication media typically embodies
Computer-readable instruction, data structure, sequence module or other numbers in modulated message signal
According to such as carrier wave or other transmission mechanisms.By example, not limit, communication media includes
Line medium such as cable network or single line connection, and wireless medium, such as sound, wireless frequency,
Infrared and other wireless mediums.So as to, software and firmware instructions can be transferred to user or
System, by communication channel, such as telephone wire, DSL lines, cable TV line, optical fiber cable,
Wireless channel, internet, etc. (passes through portable storage media and provides such software, its quilt
Regard identical as or interchangeable).Software or firmware instructions can include machine readable refer to
Order, these readable instruction cause control device to perform different actions when being performed by control device.
When performing within hardware, hardware can include one or more discrete components, integrated circuit,
Integrated circuit (ASIC) of application, etc..
It is to be appreciated that the present invention can be with pure software, pure hardware, firmware and above-mentioned each
Combination is planted to realize.Hardware for example can be control device, application specific integrated circuit, large-scale integrated
Circuit etc..
Although the example of present invention reference is described, it is intended merely to the purpose explained rather than right
The limitation of the present invention, the change to embodiment, increase and/or deletion can be made without de-
From the scope of the present invention.
Teaching involved by these embodiments, presenting from the description above and in associated accompanying drawing
Technical staff in the field of benefit will be recognized that many modifications of the invention and its recorded here
His embodiment.It should therefore be understood that the invention is not restricted to disclosed embodiment, purport
Include within the scope of the appended claims with other embodiment by modification.Although herein
Employ specific term, but only on general significance and describing significance using they and not
Used for the purpose of limitation.
Claims (10)
1. a kind of method for deleting snapped volume within the storage system, the storage system includes record
The upper adjacent volume of mapping set and record of the logical address of snapped volume and the mapping relations of storage location
Logical address and storage location mapping relations mapping set, methods described includes：
Receive the delete command for specifying snapped volume；
By the element in the mapping set of specified snapped volume, the mapping set of the upper adjacent volume is merged into
In；
The operation that feedback deletes the specified snapped volume is completed；
Wherein, the upper adjacent volume creation time is later than and close to the creation time of the specified snapped volume.
2. according to the method described in claim 1, wherein by the mapping set of specified snapped volume
Element, in the mapping set for being merged into the upper adjacent volume, including：
For specify snapped volume mapping set in each element, judge its logical address whether
In the mapping set of upper adjacent volume,
If the logical address of the element in the mapping set of the specified snapped volume is not present in described
In the mapping set of upper adjacent volume, then the element in the mapping set of the specified snapped volume is added to
In the mapping set of the upper adjacent volume.
3. the method according to claim 1-2, further comprises：
Before the operation that feedback deletes the specified snapped volume is completed, generation is recorded in NVDIMM
Table deletes the daily record for specifying snapped volume.
4. the method according to one of claim 1-3, the storage system includes data block
Mapping set, the whether effective information of data block of data block mapping set record storage position, institute
The method of stating further comprises：
If the logical address of the first element in the mapping set of the specified snapped volume is present in described
In the mapping set of upper adjacent volume, then in data block mapping set the element of mark first storage location
Data block it is invalid.
5. the method according to one of claim 1-4, further comprises：
Delete the mapping set of the specified snapped volume.
6. method according to claim 4, further comprises：
Garbage collection operation is performed according to the data block mapping set.
7. method according to claim 6, wherein, performed according to data block mapping set
Garbage collection operation includes：
It is the first data block to obtain mark in data block mapping set；
Judge whether the snapped volume indicated by the volume identification data corresponding with the first data block is deposited
, wherein,
If the snapped volume indicated by the volume identification data corresponding with first data block is not deposited
The volume identification number corresponding with the first data block is being updated with the volume identification data of the upper adjacent volume
According to, and the volume identification data corresponding with the first data block after renewal are write into NVDIMM；
If the snapped volume indicated by the volume identification data corresponding with first data block is present,
The volume identification data corresponding with the first data block are write into NVDIMM.
8. the method according to claim any one of 6-7, further comprises, it will write
NVDIMM data block polymerize and writes nonvolatile memory.
9. the method according to one of claim 1-8, wherein, the storage system enters one
Step includes the mapping set of the mapping relations for the logical address and storage location for recording adjacent volume, institute
The method of stating further comprises, the access pointer of the adjacent mapping set rolled up is pointed to the mapping of lower adjacent volume
Set, wherein, the lower adjacent volume creation time earlier than and close to the specified snapped volume establishment when
Between.
10. a kind of device for deleting snapped volume within the storage system, the storage system includes note
Record the upper neighbour of the mapping set and record of the logical address of snapped volume and the mapping relations of storage location
The mapping set of the logical address of volume and the mapping relations of storage location, described device includes：
For the module for the delete command for receiving specified snapped volume；
For by the element in the mapping set of specified snapped volume, being merged into the mapping of the upper adjacent volume
Module in set；
The module that the operation of the specified snapped volume is completed is deleted for feeding back；
Wherein, the upper adjacent volume creation time is later than and close to the creation time of the specified snapped volume.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2016100097805 | 2016-01-06 | ||
| CN201610009780 | 2016-01-06 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106951375A true CN106951375A (en) | 2017-07-14 |
| CN106951375B CN106951375B (en) | 2021-11-30 |
Family
ID=59466114
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610048363.1A Active CN106951375B (en) | 2016-01-06 | 2016-01-25 | Method and device for deleting snapshot volume in storage system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106951375B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107491529A (en) * | 2017-08-18 | 2017-12-19 | 华为技术有限公司 | A kind of snapshot delet method and node |
| CN109445681A (en) * | 2018-08-27 | 2019-03-08 | 华为技术有限公司 | Storage method, device and the storage system of data |
| CN109902034A (en) * | 2019-02-25 | 2019-06-18 | 杭州宏杉科技股份有限公司 | Snapshot creation method, device, electronic equipment and machine readable storage medium |
| CN109933460A (en) * | 2017-12-25 | 2019-06-25 | 新华三大数据技术有限公司 | Data snapshot method and device |
| CN110018988A (en) * | 2017-11-08 | 2019-07-16 | 阿里巴巴集团控股有限公司 | Snapshot delet method, processing method, apparatus and system |
| CN110399333A (en) * | 2018-04-20 | 2019-11-01 | 伊姆西Ip控股有限责任公司 | Delete method, equipment and the computer program product of snapshot |
| CN111125033A (en) * | 2018-10-31 | 2020-05-08 | 深信服科技股份有限公司 | Space recovery method and system based on full flash memory array |
| CN111124747A (en) * | 2018-10-31 | 2020-05-08 | 伊姆西Ip控股有限责任公司 | Method, apparatus and computer program product for deleting snapshots |
| CN111338850A (en) * | 2020-02-25 | 2020-06-26 | 上海英方软件股份有限公司 | Method and system for improving backup efficiency based on COW mode multi-snapshot |
| CN111857593A (en) * | 2020-07-21 | 2020-10-30 | 星辰天合（北京）数据科技有限公司 | Method and device for improving read-write speed of snapshot data and clone data |
| CN112347037A (en) * | 2020-10-13 | 2021-02-09 | 北京天融信网络安全技术有限公司 | Snapshot processing method, device, medium and electronic equipment for storage volume |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6957362B2 (en) * | 2002-08-06 | 2005-10-18 | Emc Corporation | Instantaneous restoration of a production copy from a snapshot copy in a data storage system |
| CN101140536A (en) * | 2007-05-21 | 2008-03-12 | 中兴通讯股份有限公司 | Snapshot system supporting cascades snapshot and snapshot processing method thereof |
| CN101183383A (en) * | 2007-12-17 | 2008-05-21 | 中国科学院计算技术研究所 | Snapshot system and method of use thereof |
| CN101419564A (en) * | 2008-12-11 | 2009-04-29 | 杭州华三通信技术有限公司 | Method and device for recovering data by employing snapshot |
| US7593973B2 (en) * | 2006-11-15 | 2009-09-22 | Dot Hill Systems Corp. | Method and apparatus for transferring snapshot data |
| CN101788934A (en) * | 2009-01-23 | 2010-07-28 | 英业达集团(天津)电子技术有限公司 | Data manipulation method of logical volume manager |
| CN102402471A (en) * | 2011-12-23 | 2012-04-04 | 创新科软件技术（深圳）有限公司 | Method and system for backing up data in real time based on snapshot function of memory array |
| CN102591790A (en) * | 2011-12-30 | 2012-07-18 | 记忆科技(深圳)有限公司 | Method for implementing data storage snapshot based on solid state disk, and solid state disk |
| CN102779080A (en) * | 2012-06-18 | 2012-11-14 | 上海爱数软件有限公司 | Method for generating snapshot, method and device for data recovery by using snapshot |
| US20130305002A1 (en) * | 2012-05-13 | 2013-11-14 | Xtremio Ltd. | Snapshot mechanism |
| CN103761159A (en) * | 2014-01-23 | 2014-04-30 | 天津中科蓝鲸信息技术有限公司 | Method and system for processing incremental snapshot |
-
2016
- 2016-01-25 CN CN201610048363.1A patent/CN106951375B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6957362B2 (en) * | 2002-08-06 | 2005-10-18 | Emc Corporation | Instantaneous restoration of a production copy from a snapshot copy in a data storage system |
| US7593973B2 (en) * | 2006-11-15 | 2009-09-22 | Dot Hill Systems Corp. | Method and apparatus for transferring snapshot data |
| CN101140536A (en) * | 2007-05-21 | 2008-03-12 | 中兴通讯股份有限公司 | Snapshot system supporting cascades snapshot and snapshot processing method thereof |
| CN101183383A (en) * | 2007-12-17 | 2008-05-21 | 中国科学院计算技术研究所 | Snapshot system and method of use thereof |
| CN101419564A (en) * | 2008-12-11 | 2009-04-29 | 杭州华三通信技术有限公司 | Method and device for recovering data by employing snapshot |
| CN101788934A (en) * | 2009-01-23 | 2010-07-28 | 英业达集团(天津)电子技术有限公司 | Data manipulation method of logical volume manager |
| CN102402471A (en) * | 2011-12-23 | 2012-04-04 | 创新科软件技术（深圳）有限公司 | Method and system for backing up data in real time based on snapshot function of memory array |
| CN102591790A (en) * | 2011-12-30 | 2012-07-18 | 记忆科技(深圳)有限公司 | Method for implementing data storage snapshot based on solid state disk, and solid state disk |
| US20130305002A1 (en) * | 2012-05-13 | 2013-11-14 | Xtremio Ltd. | Snapshot mechanism |
| CN102779080A (en) * | 2012-06-18 | 2012-11-14 | 上海爱数软件有限公司 | Method for generating snapshot, method and device for data recovery by using snapshot |
| CN103761159A (en) * | 2014-01-23 | 2014-04-30 | 天津中科蓝鲸信息技术有限公司 | Method and system for processing incremental snapshot |
Non-Patent Citations (1)
| Title |
|---|
| 邓换方 等: "一种高效的虚拟机快照管理系统", 《计算机系统应用》 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107491529B (en) * | 2017-08-18 | 2020-05-08 | 华为技术有限公司 | Snapshot deleting method and node |
| CN107491529A (en) * | 2017-08-18 | 2017-12-19 | 华为技术有限公司 | A kind of snapshot delet method and node |
| CN110018988B (en) * | 2017-11-08 | 2023-04-04 | 阿里巴巴集团控股有限公司 | Snapshot deleting method, processing method, device and system |
| CN110018988A (en) * | 2017-11-08 | 2019-07-16 | 阿里巴巴集团控股有限公司 | Snapshot delet method, processing method, apparatus and system |
| CN109933460A (en) * | 2017-12-25 | 2019-06-25 | 新华三大数据技术有限公司 | Data snapshot method and device |
| CN110399333B (en) * | 2018-04-20 | 2023-07-18 | 伊姆西Ip控股有限责任公司 | Method, apparatus and computer program product for deleting snapshots |
| CN110399333A (en) * | 2018-04-20 | 2019-11-01 | 伊姆西Ip控股有限责任公司 | Delete method, equipment and the computer program product of snapshot |
| US12008263B2 (en) | 2018-08-27 | 2024-06-11 | Huawei Technologies Co., Ltd. | Garbage collection and data storage method and apparatus, and storage system |
| CN109445681A (en) * | 2018-08-27 | 2019-03-08 | 华为技术有限公司 | Storage method, device and the storage system of data |
| CN111125033A (en) * | 2018-10-31 | 2020-05-08 | 深信服科技股份有限公司 | Space recovery method and system based on full flash memory array |
| CN111124747A (en) * | 2018-10-31 | 2020-05-08 | 伊姆西Ip控股有限责任公司 | Method, apparatus and computer program product for deleting snapshots |
| CN111125033B (en) * | 2018-10-31 | 2024-04-09 | 深信服科技股份有限公司 | Space recycling method and system based on full flash memory array |
| CN111124747B (en) * | 2018-10-31 | 2023-09-01 | 伊姆西Ip控股有限责任公司 | Method, apparatus and computer readable medium for deleting snapshots |
| CN109902034A (en) * | 2019-02-25 | 2019-06-18 | 杭州宏杉科技股份有限公司 | Snapshot creation method, device, electronic equipment and machine readable storage medium |
| CN111338850A (en) * | 2020-02-25 | 2020-06-26 | 上海英方软件股份有限公司 | Method and system for improving backup efficiency based on COW mode multi-snapshot |
| CN111857593A (en) * | 2020-07-21 | 2020-10-30 | 星辰天合（北京）数据科技有限公司 | Method and device for improving read-write speed of snapshot data and clone data |
| CN112347037A (en) * | 2020-10-13 | 2021-02-09 | 北京天融信网络安全技术有限公司 | Snapshot processing method, device, medium and electronic equipment for storage volume |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106951375B (en) | 2021-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106951375A (en) | The method and device of snapped volume is deleted within the storage system | |
| CN106708427B (en) | A kind of storage method suitable for key-value pair data | |
| CN103473251B (en) | The method and system of the file system snapshot using selectivity tuple versioned is provided in the computing environment using processing apparatus | |
| CN105453021B (en) | System and method for atom storage operation | |
| CN103514249B (en) | A kind of automatic data reduction and system and storage device | |
| CN103080910B (en) | Storage system | |
| US8225029B2 (en) | Data storage processing method, data searching method and devices thereof | |
| CN106662981A (en) | Storage device, program, and information processing method | |
| CN104331453B (en) | A kind of construction method of distributed file system and distributed file system | |
| CN106708425A (en) | Distributed multimode storage management | |
| CN102567427B (en) | Method and device for processing object data | |
| CN104239443B (en) | A kind of storage method of serialized data operation log | |
| CN107870970B (en) | A kind of data store query method and system | |
| CN106021031B (en) | A kind of the deletion data reconstruction method and device of BTRFS file system | |
| CN107045531A (en) | A kind of system and method for optimization HDFS small documents access | |
| CN107180092A (en) | A kind of control method of file system, device and terminal | |
| CN103544045A (en) | HDFS-based virtual machine image storage system and construction method thereof | |
| CN110196847A (en) | Data processing method and device, storage medium and electronic device | |
| CN103226965B (en) | Based on the audio/video data access method of time bitmap | |
| CN104281717B (en) | A kind of method for setting up magnanimity ID mapping relations | |
| CN108604165A (en) | Storage device | |
| CN107239569A (en) | A kind of distributed file system subtree storage method and device | |
| CN110196818A (en) | Data cached method, buffer memory device and storage system | |
| CN107704203A (en) | It polymerize delet method, device, equipment and the computer-readable storage medium of big file | |
| CN110377529A (en) | A kind of method, apparatus and equipment of full flash memory storage system data management |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 100192 room A302, building B-2, Dongsheng Science Park, Zhongguancun, 66 xixiaokou Road, Haidian District, Beijing Applicant after: Beijing yihengchuangyuan Technology Co.,Ltd. Address before: 100192 room A302, building B-2, Dongsheng Science Park, Zhongguancun, 66 xixiaokou Road, Haidian District, Beijing Applicant before: BEIJING MEMBLAZE TECHNOLOGY Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |