CN109491927A - Data storage, read method, device and electronic equipment - Google Patents
Data storage, read method, device and electronic equipment Download PDFInfo
- Publication number
- CN109491927A CN109491927A CN201811311003.1A CN201811311003A CN109491927A CN 109491927 A CN109491927 A CN 109491927A CN 201811311003 A CN201811311003 A CN 201811311003A CN 109491927 A CN109491927 A CN 109491927A
- Authority
- CN
- China
- Prior art keywords
- layer
- mapping table
- map unit
- layer mapping
- physical address
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the present invention has invented a kind of storage of data, the method, apparatus read and electronic equipment.Wherein, date storage method, comprising: be multiple first layer map units by logical block addresses Range-partition in response to write operation；Judge whether first layer map unit is continuous and complete；Discontinuous or incomplete first layer map unit is divided into multiple second layer map units；It is physical address by the address of cache of the logical blocks；The corresponding first layer mapping table of discontinuous or incomplete first layer map unit is directed toward second layer mapping table；Second layer mapping table is written into the physical address of logical blocks in first layer map unit.Avoid the problem of writing amplification.
Description
Technical field
The present invention relates to reading and writing data fields more particularly to a kind of storage of data, the method, apparatus of reading and electronics to set
It is standby.
Background technique
Solid state hard disk system generally includes host interface protocol layer, flash translation layer (FTL) (FTL, Flash Translation
) and flash interface protocol layer Layer.One major function of flash translation layer (FTL) is to carry out in host interface protocol logically
The mapping between the physical address of flash memory is arrived in location (LBA, Logical Block Address).Logical address arrives physical address
Mapping table (Logical to Physical Address Mapping Table, or L2P table) is flash translation layer (FTL)
Key data structure.With the increase of the capacity of solid state hard disk, the memory space that L2P mapping table needs is also increasing, to solid
The system design of state hard disk causes difficulty.The memory space of mapping table can be reduced by simply increasing mapping granule, but can be made
The problem of at amplification is write.
Summary of the invention
In view of this, the present invention provides a kind of storage of data, the method, apparatus read and electronic equipments, at least partly
Solution problems of the prior art.
In a first aspect, the present invention provides a kind of date storage methods, comprising:
It is multiple first layer map units by logical block addresses Range-partition in response to write operation；
Judge whether first layer map unit is continuous and complete；
Discontinuous or incomplete first layer map unit is divided into multiple second layer map units；
It is physical address by the address of cache of the logical blocks；
The corresponding first layer mapping table of discontinuous or incomplete first layer map unit is directed toward second layer mapping
Table；
The second layer is written into the physical address of logical blocks in discontinuous or incomplete first layer map unit
Mapping table.
As a specific embodiment of the present invention, the address of cache by the logical blocks is physical address
After step, comprising:
By the physical address write-in first layer mapping of the logical blocks in continuous and complete first layer map unit
Table.
It is described to judge whether first layer map unit is continuous and complete as a specific embodiment of the present invention
The step of after, further includes:
Such as first layer map unit is continuous and complete, then judges whether first layer map unit has correspondence
Second layer mapping table；
If first layer map unit is that have corresponding second layer mapping table, then the second layer mapping table is discharged；
First layer map unit is mapped directly into physical address, and updates the first layer mapping table.
It is described by discontinuous or incomplete first layer map unit as a specific embodiment of the present invention
Corresponding first layer mapping table is directed toward second layer mapping table, comprising:
Judge whether the corresponding first layer mapping table of discontinuous or incomplete first layer map unit has been directed to
Second layer mapping table,
The corresponding first layer mapping table of first layer map unit as described in discontinuous or incomplete is not directed toward the second layer
Mapping table then distributes corresponding second layer mapping table for first layer map unit.
As a specific embodiment of the present invention, the second layer mapping table is stored in second layer mapping table resource pool
It is interior.
As a specific embodiment of the present invention, when needing second layer mapping table, from the second layer mapping table
Second layer mapping table is distributed in resource pool；
When the second layer mapping table is not needed, then the second layer mapping table is discharged into the second layer mapping table
Resource pool.
Second layer mapping table as a specific embodiment of the present invention, in the second layer mapping table resource pool
When being reduced to threshold value, map unit union operation is carried out.
As a specific embodiment of the present invention, the map unit union operation, comprising:
Data in all logical blocks for being under the jurisdiction of same first layer map unit are read out,
The data of reading are reconsolidated；
The data reconsolidated are put into first layer map unit, and first layer map unit is passed through first
Layer mapping table is mapped as physical address, reflects so that the second layer mapping table that the first layer mapping table is directed toward is discharged into the second layer
In firing table resource pool.
As a specific embodiment of the present invention, the triggering of the map unit union operation in not storing data
Threshold value is less than the activation threshold value of map unit union operation when storing data.
As a specific embodiment of the present invention, the size of the map unit of the first layer is logical blocks size
Integral multiple, the size of second layer map unit is identical with the size of logical blocks.
As a specific embodiment of the present invention, the first layer mapping table is array, the first layer mapping table
Array in element be physical address or second layer mapping table address, the quantity of element in the array of the first layer mapping table
It is the quantity of first layer map unit of logical block addresses Range-partition.
As a specific embodiment of the present invention, the second layer mapping table is array, the second layer mapping table
Array in element be physical address, the quantity of element is that first layer mapping is single in the array of the second layer mapping table
The quantity for the logical blocks that member is included.
As a specific embodiment of the present invention, the first layer mapping of second layer map unit can be divided into
The quantity of unit is equipped with upper limit value.
As a specific embodiment of the present invention, comprising:
It is multiple first layer map units by logical block addresses Range-partition in response to write operation；
First layer map unit is divided into multiple second layer map units；
Second layer map unit is divided into multiple third layer map units；
Until
The n-th layer map unit is divided into multiple (n+1)th layer map units.
As a specific embodiment of the present invention, the n-th layer map unit size is (n+1)th layer of map unit
The integral multiple of size, and described (n+1)th layer of map unit size is identical as logical blocks size.
Second aspect, the present invention also provides a kind of method for reading data, comprising:
It is the first layer map unit by logical block addresses Range-partition in response to read operation；
Inquire the corresponding first layer mapping table of first layer map unit；
The corresponding physical address of logical blocks is such as inquired in the first layer mapping table, then directly from the first layer
Physical address is read in mapping table, and data are then read by the physical address in the first layer mapping table；
The corresponding physical address of logical blocks is not inquired in the first layer mapping table such as, then from the first layer
The address of second layer mapping table is read in mapping table, then reads physical address from second layer mapping table；Then pass through described the
The physical address in two layers of mapping table reads data.
The third aspect, the present invention also provides a kind of data storage devices, use the date storage method of first aspect；
With or
The method for reading data of second aspect.
Fourth aspect, the present invention also provides a kind of electronic equipment, use the data storage device of the third aspect.
Data provided by the invention storage, the method, apparatus read and electronic equipment, wherein the date storage method passes through
Second layer map unit is divided into the first layer map unit, and the corresponding first layer mapping table of the first layer map unit is directed toward
Second layer mapping table, so that second layer mapping table is written in the physical address of the logical blocks in first layer map unit.
When only writing data into a logical blocks, by being further split to the first layer map unit, first layer mapping table
First layer map unit is directed toward second layer mapping table, to only need the logical blocks being newly written second layer mapping table is written
Interior physical address.It reads, avoids out of flash memory without the logical blocks in the first layer map unit that will be saved originally
When only writing data into a logical blocks, the logical blocks in same first layer map unit are read out of flash memory
Out, after the data being then written by the data of reading and newly merge, it is mapped as physical address again, writes data into new
Physical address, thus caused by the problem of writing amplification.
The above description is only an overview of the technical scheme of the present invention, in order to better understand technological means of the invention, and
It can be implemented in accordance with the contents of the specification, and to allow above and other objects, features and advantages of the invention can be brighter
Show understandable, it is special below to lift preferred embodiment, and cooperate attached drawing, detailed description are as follows.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this field
For those of ordinary skill, without creative efforts, it can also be obtained according to these attached drawings other attached drawings.
Fig. 1 is a kind of flow chart of date storage method provided in an embodiment of the present invention；
Fig. 2 is the flow chart of release second layer mapping table provided in an embodiment of the present invention；
Fig. 3 is the flow chart of date storage method concrete application provided in an embodiment of the present invention；
Fig. 4 is the structural block diagram of single layer mapping table provided in an embodiment of the present invention and the double-deck mapping table；
Fig. 5 is the flow chart of mapping union operation provided in an embodiment of the present invention.
Specific embodiment
The embodiment of the present invention is described in detail with reference to the accompanying drawing.
It will be appreciated that illustrating embodiments of the present invention below by way of specific specific example, those skilled in the art can
By content disclosed by this specification, other advantages and efficacy of the present invention can be easily understood.Obviously, described embodiment is only
It is only a part of the embodiment of the present invention, instead of all the embodiments.The present invention can also pass through in addition different specific implementations
Mode is embodied or practiced, and the various details in this specification can also be based on different viewpoints and application, without departing from this
Various modifications or alterations are carried out under the spirit of invention.It should be noted that in the absence of conflict, following embodiment and embodiment
In feature can be combined with each other.Based on the embodiments of the present invention, those of ordinary skill in the art are not making creativeness
Every other embodiment obtained, shall fall within the protection scope of the present invention under the premise of labour.
It should be noted that the various aspects of embodiment within the scope of the appended claims are described below.Ying Xian
And be clear to, aspect described herein can be embodied in extensive diversified forms, and any specific structure described herein
And/or function is only illustrative.Based on the present invention, it will be understood by one of ordinary skill in the art that one described herein
Aspect can be independently implemented with any other aspect, and can combine the two or both in these aspects or more in various ways.
For example, carry out facilities and equipments in terms of any number set forth herein can be used and/or practice method.In addition, can make
With other than one or more of aspect set forth herein other structures and/or it is functional implement this equipment and/or
Practice the method.
It should also be noted that, illustrating the basic structure that only the invention is illustrated in a schematic way provided in following embodiment
Think, only shown in schema with it is of the invention in related component rather than component count, shape and size when according to actual implementation draw
System, when actual implementation kenel, quantity and the ratio of each component can arbitrarily change for one kind, and its assembly layout kenel can also
It can be increasingly complex.
In addition, in the following description, specific details are provided for a thorough understanding of the examples.However, fields
The skilled person will understand that the aspect can be practiced without these specific details.
Mapping granule: i.e. the size of map unit, refer to logical address to physical address map minimum unit.
Solid state hard disk does storage medium using flash memory, and flash memory must be wiped first before writing, and wipe granularity (block,
Block) much larger than the granularity (page, page) of write-in.Due to the characteristic of flash memory, host logically address write-in data when, FTL
Algorithm logical address is mapped as to the physical address of flash memory, flash memory then is written into according to physical address in data again；Work as host
When repeating that data are written to identical logical address, it can be mapped as different physical address every time.Solid state hard disk L2P
Mapping table come record logical address to physical address mapping relations.
Logical address to physical address mapping granularity (namely size of map unit) be usually fixed (such as
4KB).If ranges of logical addresses (i.e. the capacity of solid state hard disk) is L, and (i.e. solid state hard disk storage is situated between the range of physical address
The actual capacity of matter) it is P, the size of map unit is U, then the size of mapping table can be indicated by big O labelling method are as follows:
S=O (L/U*log (P/U)),
It can be seen that mapping table is smaller if mapping granule is bigger.
Mapping table is commonly stored in the SDRAM of solid state hard disk.Due to the limitation of SDRAM memory space, mapping table it is big
It is small cannot be too big, therefore the capacity of solid state hard disk is also restrained.If solid state hard disk capacity is too big so that SDRAM is stored not
Map unit is usually expanded as the size that mapping table is reduced comprising multiple logical blocks by lower mapping table, FTL algorithm.At this
In the case of kind, if host only writes data into one of logical blocks, FTL algorithm usually require by data from other
It is read out positioned at having been written into the logical blocks of flash memory for same mapping unit, merges with the data being newly written, remap
Then new physical address is written in data after merging by new physical address.The write-in of one logical blocks of host computer side in this way,
It needs to read multiple logical blocks in flash memory side, and multiple logical blocks is written, cause to write amplification.If map unit includes N
A logical blocks, then writing amplification factor caused by this mapping algorithm is also N.
FTL natively exists when handling the write-in of logic random address writes amplification, this to write amplification and solid state hard disk
Excessive accomodation coefficient (Over-Provisioning Ratio) it is related.Map unit comprising N number of logical blocks can make originally
It is existing to write N times of further expansion of amplification, to cause very big influence to the performance of random address write-in.
Embodiment one:
As shown in Figure 1, a kind of date storage method, comprising:
Step S101: being multiple first layer map units by logical block addresses Range-partition in response to write operation；
Logical block addresses range includes multiple logical blocks, and ranges of logical addresses is the capacity of solid state hard disk.For dividing
Being segmented into multiple in multiple first layer map units includes one or more, is divided into one if inadequate one.
Step S102: judge whether first layer map unit is continuous and complete；Continuously refer to the logic of write-in data
Whether address is continuous, when data are written in host, it may be possible to random address write-in, it is also possible to multiple logical blocks are continuously written into,
In random address write-in, it is discontinuous to will cause the first layer map unit, for being completely, if the data of host write-in cannot
Gather together enough first layer map unit, then the first layer map unit be exactly it is imperfect, instead then, if write-in data can gather together enough
One the first layer map unit, then first layer map unit is complete.
Step S103: discontinuous or incomplete first layer map unit is divided into multiple second layers and maps list
Member；
Second layer map unit can be smaller than the first layer map unit, therefore imperfect or discontinuous first layer is mapped list
Member is divided into lesser second layer map unit.
In a preferred embodiment, the size of the map unit of first layer is the integral multiple of logical blocks size, the
The size of two layer map units and the size of logical blocks are identical.Because of the size of the second layer map unit and the size of logical blocks
Identical, then discontinuous for the map unit of first layer or incomplete write-in data can be divided into complete and continuous
Two layer map units.
Step S104: being physical address by the address of cache of the logical blocks；
Address of cache by the logical blocks of host write-in is the physical address of flash memory in solid state hard disk, that is, is first had to write
The address allocated physical address of the logical blocks entered.
Step S105: the corresponding first layer mapping table of discontinuous or incomplete first layer map unit is directed toward the
Two layers of mapping table；
If the logical blocks of host write-in are continuous and completely by the first layer map unit i.e. by logical blocks
Address is mapped as physical address by first layer mapping table, but if be it is incomplete, first layer mapping table be directed toward will not be
Physical address, but second layer mapping table.
Step S106: the physical address of the logical blocks in discontinuous or incomplete first layer map unit is write
Enter second layer mapping table, after first layer mapping table is directed toward second layer mapping table, logical blocks are final or to store flash memory
It is interior, therefore storage is the physical address of flash memory in second layer mapping table, the address of such logical blocks passes through the first mapping table
It is directed toward second layer mapping table, then physical address is directed toward by second layer mapping table, and logical blocks storage is mapped to the second layer
In the physical address that table is directed toward.
As a specific embodiment of the present invention, it is by the address of cache of the logical blocks described in the step S104
After the step of physical address, comprising:
By the physical address write-in first layer mapping of the logical blocks in continuous and complete first layer map unit
Table.
Physical address is directly just mapped as by first layer mapping table if it is continuous and complete first layer map unit.
The present invention directly will be just mapped as physical address by first layer mapping table and be known as single layer mapping, and will pass through first
Mapping table is directed toward second layer mapping table, then the referred to as double-deck mapping of physical address is directed toward by second layer mapping table.
As a specific embodiment of the present invention, step S102 judge first layer map unit it is whether continuous and
After complete step, further includes:
Step S201: such as first layer map unit is continuous and complete, then judges first layer map unit
Whether corresponding second layer mapping table is had；
Step S202: if first layer map unit is that have corresponding second layer mapping table, then the second layer is reflected
Firing table release；
Step S203: being mapped directly into physical address for first layer map unit, and updates the first layer mapping
Table.
As a specific embodiment of the present invention, step S105 maps the discontinuous or incomplete first layer
The corresponding first layer mapping table of unit is directed toward second layer mapping table, comprising:
Judge whether the corresponding first layer mapping table of discontinuous or incomplete first layer map unit has been directed to
Second layer mapping table,
The corresponding first layer mapping table of first layer map unit as described in discontinuous or incomplete is not directed toward the second layer
Mapping table then distributes corresponding second layer mapping table for first layer map unit.
As shown in figure 3, host or FTL write operation are responded, by logical block addresses model in a specific application scenarios
It encloses and is divided into multiple first layer map units；
Judge whether the first layer map unit is continuous and complete；
If it is judged that be yes, then judge whether this first layer map unit has done the double-deck mapping, if
The double-deck mapping has been done, then has been discharged the corresponding second layer mapping table of this first layer map unit, can preferably be discharged into the
In two layers of mapping table resource pool.
If it is determined that whether the first layer map unit continuous and complete result be it is no, then judge the mapping of this first layer singly
Whether member has done the double-deck mapping, if not doing the double-deck mapping, distributes second layer mapping for this first layer map unit
Table is preferably chosen in second layer mapping table resource pool and is allocated.
Then judge whether this first layer map unit has done single layer mapping, such as done single layer mapping, then by
Physical address in one layer map unit reads from first layer mapping table and is converted into the physical address of each logical blocks.
Then the logical block addresses being newly written are mapped as physical address, this first layer map unit is judged if
After result is the step of being in the step of whether having done the double-deck mapping, or whether to judge this first layer map unit
Through done single layer mapping result be it is no after step after, the logical block addresses for being required to newly to be written are mapped as physical address.
The first layer map unit is directed toward second layer mapping table in first layer mapping table.
And second layer mapping table is written into the physical address of each logical blocks in the first layer map unit, to will count
According in the flash memory for being written to solid state hard disk.
In a specific application scenarios when operation of host or FTL itself triggering write-in flash memory, FTL first will even
Continuous logical block addresses are cut into the first layer map unit one by one, and these map units are mapped as physical address,
Then mapping relations are recorded in the mapping table of first layer；If the logical blocks of write-in are big less than the first layer map unit
Small, this first layer map unit is just first directed toward a second layer mapping table in first layer mapping table by FTL, then again
In two layers of mapping table the logical blocks in the first layer map unit are mapped as physical address one by one, then record mapping relations
In the mapping table of the second layer.
In a specific application scenarios, first layer mapping table only one.Therefore it is described in technical solution of the present invention
The corresponding first layer mapping table of the first layer map unit is directed toward second layer mapping table, refer to by first layer mapping table with the
The corresponding part of one layer map unit.
As a specific embodiment of the present invention, first layer mapping table can be an array when implementing, and
Element in array is physical address or second layer mapping table address, if the element in single layer mapping array is exactly physical address,
Such as using the double-deck mapping, then the element in array is exactly second layer mapping table address, the quantity of element is exactly logic in array
The quantity of the first layer map unit of address.Such as A, that is, logical address, B, that is, physical address, mapping relations are: A0-B4, A1- > B9,
A2- > B3, A3- > B5 just only need [B4, B9, B3, B5] in that mapping table.The index of mapping table is exactly logical address, content
It is exactly physical address.
If logical block addresses range is L, the size of the first layer map unit is U1, then the quantity of element is in array
L/U1。
If the data of write-in include one or more complete first layer map units, each first layer mapping is single
Member can disposably be mapped as physical address, and element corresponding with these first layer map units is just in first layer mapping table
It is corresponding physical address.
If range of physical addresses is P, physical address can be divided into P/U1 the first layer map units, therefore the
The physical address of one layer map unit needs log2 (P/U1)/8 byte to indicate.
The storage size that first layer mapping table needs is at least are as follows:
S1=L/U1*log2 (P/U1)/8.
As soon as if the data of host write-in cannot gather together enough first layer map unit, then this first layer map unit
It cannot be mapped according to single layer；In this case, in first layer mapping table, this first layer map unit institute is right
The element answered just is not directed to physical address, and is directed to second layer mapping table.Each second layer mapping table is also a number
Group, the element in array is physical address, and the quantity of element is exactly the logic area that the first layer map unit is included in array
The quantity of block.
It is specifically as shown in Figure 4 for the relationship of the first layer map unit, first layer mapping table and second layer mapping table.
As a specific embodiment of the present invention, the first layer mapping of second layer map unit can be divided into
The quantity of unit is equipped with upper limit value.
I.e. each first layer logical mappings unit that cannot do single layer mapping has a corresponding second layer mapping table.
For save mapping table memory space, to cannot do single layer mapping first layer logical mappings unit limited amount system.If
At most allow there are the J first layer logical mappings units that cannot do single layer mapping, then being up to J second layer mapping in system
Table.
If the first layer map unit includes M the second layer map units, in that second layer mapping table, the quantity of element is just
It is M.
If range of physical addresses is P, the second layer map unit size is U2, then each element needs in second layer mapping table
Log2 (P/U2)/8 byte is wanted to indicate.
The storage size that second layer mapping table needs in system is at least are as follows:
S2=J*M*log2 (P/U2)/8.
As a specific embodiment of the present invention, the second layer mapping table is stored in second layer mapping table resource pool
It is interior.
As a specific embodiment of the present invention, when needing second layer mapping table, from the second layer mapping table
Second layer mapping table is distributed in resource pool；
When the second layer mapping table is not needed, then the second layer mapping table is discharged into the second layer mapping table
Resource pool.
All second layer mapping tables are placed in a resource pool by FTL, from money when needing to carry out the double-deck mapping
Second layer mapping table is distributed in the pond of source, is released into resource pool if the second layer mapping table no longer needs.
It, may be to script single layer when host carries out random address write-in in a specific application scenarios
Data (the corresponding data of logical address) in first layer map unit of mapping are modified, this map unit just needs in this way
It to be mapped from single layer and be changed to the double-deck mapping.In this case, a second layer is first distributed from second layer mapping table resource pool
Mapping table, original physical address of the second layer map unit (i.e. logical blocks) that data are not changed is from first layer mapping table
It copies and is transformed into second layer mapping table, then there is the second layer map unit of change to reflect again data in second layer mapping
It penetrates as new physical address and is recorded in second layer mapping table, the first layer map unit is directed toward this in first layer mapping table
Second layer mapping table.
When host carries out sequence address write-in, the multiple second layers mapping for needing the double-deck mapping originally may be made
Unit can be with Continuous Mappings for first layer map unit, that is to say, that change into single layer mapping by the original double-deck mapping.
In this case, first original second layer mapping table is discharged into second layer mapping table resource pool, then by corresponding first
Map unit is once mapped as physical address and updates into first layer mapping table.
Second layer mapping table as a specific embodiment of the present invention, in the second layer mapping table resource pool
When being reduced to threshold value, map unit union operation is carried out.
As a specific embodiment of the present invention, the map unit union operation, comprising:
Data in all logical blocks for being under the jurisdiction of same first layer map unit are read out,
The data of reading are reconsolidated；
The data reconsolidated are put into first layer map unit, and first layer map unit is passed through first
Layer mapping table is mapped as physical address, reflects so that the second layer mapping table that the first layer mapping table is directed toward is discharged into the second layer
In firing table resource pool.
As a specific embodiment of the present invention, the triggering of the map unit union operation in not storing data
Threshold value is less than the activation threshold value of map unit union operation when storing data.
In a specific application scenarios, as shown in Figure 5.When host carries out random address write-in, FTL is used
The quantity of second layer mapping table will increase, the resource quantity in second layer mapping table resource pool can be reduced in other words；Work as host
When being sequentially written in, the quantity for the second layer mapping table that FTL is used can be reduced, in other words second layer mapping table resource pool
In resource quantity will increase.When the resource quantity in second layer mapping table resource pool be reduced to certain thresholding (threshold value) when
It waits, FTL will start map unit union operation, that is, the institute for being under the jurisdiction of same first layer map unit that bilayer is mapped
There are the data in logical blocks to read out from flash memory, reconsolidates and carry out single layer mapping, reflected to discharge the original second layer
Firing table is into second layer mapping table resource pool.And judge whether the second layer mapping table resource in second layer mapping table resource pool is high
In terminate thresholding, as be higher than end thresholding if terminate map unit union operation, be such as not above end thresholding, can such as continue into
Row map unit union operation.
FTL usage threshold when idle state is different from the thresholding that working condition uses, and makes it in idle state ratio
Map unit merging is more energetically carried out in working condition.Idle state is in no progress write operation, and working condition is
When carrying out write operation.Host will not trigger FTL when doing random writing whithin a period of time and do map unit merging；And it is writing
After entering stopping, it is possible to triggering map unit merging.If host continuously does random writing, FTL eventually because
Second layer mapping table reduces too many and starts map unit union operation.
The scheme of storing data as a preferred option can be carried out further on the basis of the above-mentioned double-deck mapping
Multi-level mapping specifically includes:
It is multiple first layer map units by logical block addresses Range-partition in response to write operation；
First layer map unit is divided into multiple second layer map units；
Second layer map unit is divided into multiple third layer map units；
Etc. until,
The n-th layer map unit is divided into multiple (n+1)th layer map units.
In the mapping scheme of multilayer: n-th layer map unit size is the integral multiple of (n+1)th layer of map unit size,
And the last layer map unit size is identical as logical blocks size；N-th layer map unit is in n-th layer mapping table or is directed toward
Physical address (if the logical blocks that the map unit includes are continuous and complete) or (n+1)th layer of mapping table of direction be not (if
The map unit can be mapped directly into physical address)；First layer mapping table can be a fixed array, and other layers reflect
Firing table then dynamically can distribute and merge.Multi-level mapping is that the double-deck extension mapped, principle and method are mapped with double-deck,
This is repeated no more.
Embodiment two:
Corresponding with date storage method in embodiment one, the present invention also provides a kind of method for reading data, comprising:
It is the first layer map unit by logical block addresses Range-partition in response to read operation；
Inquire the corresponding first layer mapping table of first layer map unit；
The corresponding physical address of logical blocks is such as inquired in the first layer mapping table, then directly from the first layer
Physical address is read in mapping table, and data are then read by the physical address in the first layer mapping table；
The corresponding physical address of logical blocks is not inquired in the first layer mapping table such as, then from the first layer
The address of second layer mapping table is read in mapping table, then reads physical address from second layer mapping table；Then pass through described the
The physical address in two layers of mapping table reads data.
When host carries out read operation, logical block addresses are first divided into the first layer map unit by FTL, are then looked into
First layer mapping table；If it find that being single layer mapping, physical address is directly read from mapping table；It is mapped if it is bilayer, first
The address of second layer mapping table is read from first layer mapping table, then reads physical address from second layer mapping table；Then lead to
It crosses physical address and reads data.
Embodiment three:
The present invention also provides a kind of data storage devices, use the date storage method for implementing one；
With or
Implement two method for reading data.
Such as solid state hard disk or other data storage devices based on solid state hard disk, or the storage principle with solid state hard disk
The same data storage device, such as data storage of existing mobile phone or tablet computer.
Example IV:
The present invention also provides a kind of electronic equipment, use the data storage device of embodiment three.
Such as mobile phone, computer, VR equipment.
In specific application, host whithin a period of time random writing when will not trigger map unit merging, from
Amplification is write without additionally increasing, will not influence the performance of host random writing；When host computer side stops being written, FTL can
Map unit merging can be triggered, more second layer mapping tables is discharged, writes amplification although will cause this when, still
Due to occurring during idle time, the performance that host is write is nor affected on.
If host continuously carries out random writing, that FTL is eventually because remaining second layer mapping table is very little
And map unit merging is carried out while being forced in host write-in.But this write mode is the extreme situation of comparison, very much
It will not occur in actual operation flow.
Basic principle of the invention is described in conjunction with specific embodiments above, however, it is desirable to, it is noted that in the present invention
The advantages of referring to, advantage, effect etc. are only exemplary rather than limitation, must not believe that these advantages, advantage, effect etc. are of the invention
Each embodiment is prerequisite.In addition, the detail of foregoing invention is merely to exemplary effect and the work being easy to understand
With, rather than limit, above-mentioned details is not intended to limit the present invention as that must realize using above-mentioned concrete details.
In the present invention, relational terms such as first and second and the like be used merely to by an entity or operation with
Another entity or operation distinguish, and without necessarily requiring or implying between these entities or operation, there are any this realities
The relationship or sequence on border, the present invention involved in device, device, equipment, system block diagram be only used as illustrative example
And it is not intended to require or imply to be attached in such a way that box illustrates, arrange, configure.Such as those skilled in the art
What member will be recognized, it can be connected by any way, arrange, configure these devices, device, equipment, system.Such as " comprising ",
The word of "comprising", " having " etc. is open vocabulary, is referred to " including but not limited to ", and can be used interchangeably with it.Here institute
The vocabulary "or" and "and" used refers to vocabulary "and/or", and can be used interchangeably with it, unless context be explicitly indicated be not as
This.Vocabulary " such as " used herein above refers to phrase " such as, but not limited to ", and can be used interchangeably with it.
In addition, as used herein, the "or" instruction separation used in the enumerating of the item started with "at least one"
It enumerates, so that enumerating for such as " at least one of A, B or C " means A or B or C or AB or AC or BC or ABC (i.e. A and B
And C).In addition, wording " exemplary " does not mean that the example of description is preferred or more preferable than other examples.
It may also be noted that in the system and method for the present invention, each component or each step are can to decompose and/or again
Combination nova.These, which decompose and/or reconfigure, should be regarded as equivalent scheme of the invention.
The technology instructed defined by the appended claims can not departed from and carried out to the various of technology described herein
Change, replace and changes.In addition, the scope of the claims of the invention is not limited to process described above, machine, manufacture, thing
Composition, means, method and the specific aspect of movement of part.Can use carried out to corresponding aspect described herein it is essentially identical
Function or realize essentially identical result there is currently or later to be developed processing, machine, manufacture, event group
At, means, method or movement.Thus, appended claims include such processing, machine, manufacture, event within its scope
Composition, means, method or movement.
The above description of invented aspect is provided so that any person skilled in the art can make or use this
Invention.Various modifications in terms of these are readily apparent to those skilled in the art, and are defined herein
General Principle can be applied to other aspect without departing from the scope of the present invention.Therefore, the present invention is not intended to be limited to
Aspect shown in this, but according to the principle and the consistent widest range of novel feature in the invention.
In order to which purpose of illustration and description has been presented for above description.In addition, this description is not intended to reality of the invention
It applies example and is restricted to form in the invention.Although already discussed above multiple exemplary aspects and embodiment, this field skill
Its certain modifications, modification, change, addition and sub-portfolio will be recognized in art personnel.
Claims (18)
1. a kind of date storage method characterized by comprising
It is multiple first layer map units by logical block addresses Range-partition in response to write operation；
Judge whether first layer map unit is continuous and complete；
Discontinuous or incomplete first layer map unit is divided into multiple second layer map units；
It is physical address by the address of cache of the logical blocks；
The corresponding first layer mapping table of discontinuous or incomplete first layer map unit is directed toward second layer mapping table；
By the physical address write-in second layer mapping of the logical blocks in discontinuous or incomplete first layer map unit
Table.
2. date storage method according to claim 1, which is characterized in that the address of cache by the logical blocks
After the step of for physical address, comprising:
First layer mapping table is written into the physical address of logical blocks in continuous and complete first layer map unit.
3. date storage method according to claim 1, which is characterized in that described to judge that first layer map unit is
After no continuous and complete step, further includes:
As first layer map unit be it is continuous and complete, then judge whether first layer map unit has corresponding the
Two layers of mapping table；
If first layer map unit is that have corresponding second layer mapping table, then the second layer mapping table is discharged；
First layer map unit is mapped directly into physical address, and updates the first layer mapping table.
4. date storage method according to claim 1, it is characterised in that: it is described will be discontinuous or incomplete described
The corresponding first layer mapping table of one layer map unit is directed toward second layer mapping table, comprising:
Judge whether the corresponding first layer mapping table of discontinuous or incomplete first layer map unit has been directed to second
Layer mapping table,
The corresponding first layer mapping table of first layer map unit as described in discontinuous or incomplete is not directed toward second layer mapping
Table then distributes corresponding second layer mapping table for first layer map unit.
5. date storage method according to claim 1, which is characterized in that the second layer mapping table is stored in the second layer
In mapping table resource pool.
6. date storage method according to claim 5, it is characterised in that:
When needing second layer mapping table, second layer mapping table is distributed from the second layer mapping table resource pool；
When the second layer mapping table is not needed, then the second layer mapping table is discharged into the second layer mapping table resource
Pond.
7. date storage method according to claim 6, it is characterised in that:
When the second layer mapping table in the second layer mapping table resource pool is reduced to threshold value, carries out map unit and merge behaviour
Make.
8. date storage method according to claim 7, which is characterized in that the map unit union operation, comprising:
Data in all logical blocks for being under the jurisdiction of same first layer map unit are read out,
The data of reading are reconsolidated；
The data reconsolidated are put into first layer map unit, and first layer map unit is reflected by first layer
Firing table is mapped as physical address, so that the second layer mapping table that the first layer mapping table is directed toward is discharged into second layer mapping table
In resource pool.
9. date storage method according to claim 7, it is characterised in that:
In not storing data, the activation threshold value of the map unit union operation is less than map unit when storing data and closes
And the activation threshold value operated.
10. date storage method according to claim 1, it is characterised in that:
The size of the map unit of the first layer is the integral multiple of logical blocks size, the size of second layer map unit
It is identical with the size of logical blocks.
11. date storage method according to claim 1, it is characterised in that:
The first layer mapping table is array, and the element in the array of the first layer mapping table is that physical address or the second layer reflect
Firing table address, the quantity of element is the first layer of logical block addresses Range-partition in the array of the first layer mapping table
The quantity of map unit.
12. date storage method according to claim 11, it is characterised in that:
The second layer mapping table is array, and the element in the array of the second layer mapping table is physical address, described second
The quantity of element is the quantity for the logical blocks that first layer map unit is included in the array of layer mapping table.
13. date storage method according to claim 1, it is characterised in that:
The quantity that the first layer map unit of second layer map unit can be divided into is equipped with upper limit value.
14. date storage method according to claim 1, it is characterised in that: include:
It is multiple first layer map units by logical block addresses Range-partition in response to write operation；
First layer map unit is divided into multiple second layer map units；
Second layer map unit is divided into multiple third layer map units；
Until
The n-th layer map unit is divided into multiple (n+1)th layer map units.
15. date storage method according to claim 14, it is characterised in that:
The n-th layer map unit size is the integral multiple of (n+1)th layer of map unit size, and (n+1)th layer of the mapping
Cell size is identical as logical blocks size.
16. a kind of method for reading data characterized by comprising
It is the first layer map unit by logical block addresses Range-partition in response to read operation；
Inquire the corresponding first layer mapping table of first layer map unit；
The corresponding physical address of logical blocks is such as inquired in the first layer mapping table, then is directly mapped from the first layer
Physical address is read in table, and data are then read by the physical address in the first layer mapping table；
The corresponding physical address of logical blocks is not inquired in the first layer mapping table such as, then is mapped from the first layer
The address of second layer mapping table is read in table, then reads physical address from second layer mapping table；Then pass through the second layer
The physical address in mapping table reads data.
17. a kind of data storage device, which is characterized in that use data storage side described in claim 1 to claim 15
Method；
With or
Method for reading data described in claim 16.
18. a kind of electronic equipment, which is characterized in that use the data storage device described in claim 17.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811311003.1A CN109491927B (en) | 2018-11-06 | 2018-11-06 | Data storage method, data reading method, data storage device, data reading device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811311003.1A CN109491927B (en) | 2018-11-06 | 2018-11-06 | Data storage method, data reading method, data storage device, data reading device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109491927A true CN109491927A (en) | 2019-03-19 |
| CN109491927B CN109491927B (en) | 2023-02-03 |
Family
ID=65693914
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811311003.1A Active CN109491927B (en) | 2018-11-06 | 2018-11-06 | Data storage method, data reading method, data storage device, data reading device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109491927B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110347614A (en) * | 2019-07-08 | 2019-10-18 | 青岛镕铭半导体有限公司 | Memory space mapping algorithm, buffer status machine, storage device, storage medium |
| CN111158584A (en) * | 2019-11-26 | 2020-05-15 | 国网江苏省电力有限公司盐城供电分公司 | Distribution network construction safety quality control platform big data storage method and system based on ubiquitous power Internet of things |
| CN111897745A (en) * | 2019-05-05 | 2020-11-06 | 北京兆易创新科技股份有限公司 | Data storage method and device, electronic equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110283052A1 (en) * | 2010-05-14 | 2011-11-17 | Tdk Corporation | Memory controller, flash memory system with memory controller, and method of controlling flash memory |
| CN102819496A (en) * | 2012-08-16 | 2012-12-12 | 无锡紫芯集成电路系统有限公司 | Address translation method of flash FTL (Flash Translation Layer) |
| CN104133779A (en) * | 2013-05-02 | 2014-11-05 | 擎泰科技股份有限公司 | Non-volatile memory and method of dynamically adjusting mapping manner of same |
| US20160048448A1 (en) * | 2013-03-25 | 2016-02-18 | Ajou University Industry-Academic Cooperation Foundation | Method for mapping page address based on flash memory and system therefor |
| CN106547703A (en) * | 2016-10-08 | 2017-03-29 | 华中科技大学 | A kind of FTL optimization methods based on block group structure |
| US20170235488A1 (en) * | 2016-02-11 | 2017-08-17 | SK Hynix Inc. | Window based mapping |
-
2018
- 2018-11-06 CN CN201811311003.1A patent/CN109491927B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110283052A1 (en) * | 2010-05-14 | 2011-11-17 | Tdk Corporation | Memory controller, flash memory system with memory controller, and method of controlling flash memory |
| CN102819496A (en) * | 2012-08-16 | 2012-12-12 | 无锡紫芯集成电路系统有限公司 | Address translation method of flash FTL (Flash Translation Layer) |
| US20160048448A1 (en) * | 2013-03-25 | 2016-02-18 | Ajou University Industry-Academic Cooperation Foundation | Method for mapping page address based on flash memory and system therefor |
| CN104133779A (en) * | 2013-05-02 | 2014-11-05 | 擎泰科技股份有限公司 | Non-volatile memory and method of dynamically adjusting mapping manner of same |
| US20170235488A1 (en) * | 2016-02-11 | 2017-08-17 | SK Hynix Inc. | Window based mapping |
| CN106547703A (en) * | 2016-10-08 | 2017-03-29 | 华中科技大学 | A kind of FTL optimization methods based on block group structure |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111897745A (en) * | 2019-05-05 | 2020-11-06 | 北京兆易创新科技股份有限公司 | Data storage method and device, electronic equipment and storage medium |
| CN111897745B (en) * | 2019-05-05 | 2024-04-05 | 兆易创新科技集团股份有限公司 | Data storage method and device, electronic equipment and storage medium |
| CN110347614A (en) * | 2019-07-08 | 2019-10-18 | 青岛镕铭半导体有限公司 | Memory space mapping algorithm, buffer status machine, storage device, storage medium |
| CN110347614B (en) * | 2019-07-08 | 2022-08-12 | 镕铭微电子(济南)有限公司 | Storage space mapping algorithm, cache state machine, storage device, and storage medium |
| CN111158584A (en) * | 2019-11-26 | 2020-05-15 | 国网江苏省电力有限公司盐城供电分公司 | Distribution network construction safety quality control platform big data storage method and system based on ubiquitous power Internet of things |
| CN111158584B (en) * | 2019-11-26 | 2022-09-30 | 国网江苏省电力有限公司盐城供电分公司 | Distribution network construction safety quality control platform big data storage method and system based on ubiquitous power Internet of things |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109491927B (en) | 2023-02-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7962687B2 (en) | Flash memory allocation for improved performance and endurance | |
| US8010770B2 (en) | Caching device for NAND flash translation layer | |
| CN108021510A (en) | The method for operating the storage device being managed to multiple name space | |
| US7761648B2 (en) | Caching method for NAND flash translation layer | |
| TWI494761B (en) | Method of partitioning physical block and memory system thereof | |
| US20140089564A1 (en) | Method of data collection in a non-volatile memory | |
| CN109491927A (en) | Data storage, read method, device and electronic equipment | |
| US20090037648A1 (en) | Input/output control method and apparatus optimized for flash memory | |
| CN104866428A (en) | Data access method and data access device | |
| US20180129602A1 (en) | Garbage collection method for data storage device | |
| US20190391756A1 (en) | Data storage device and cache-diversion method thereof | |
| CN111240601B (en) | Method, device, equipment and storage medium for determining superblock of partitioned space | |
| CN102541755B (en) | The method of flash memories and reception data thereof | |
| CA2758235A1 (en) | Device and method for storage, retrieval, relocation, insertion or removal of data in storage units | |
| CN109582219A (en) | Storage system, computing system and its method | |
| JP2010515162A5 (en) | ||
| CN101673246A (en) | High-efficient first-in first-out (FIFO) data pool reading and writing method | |
| US20120131264A1 (en) | Storage device | |
| WO2020192710A1 (en) | Method for processing garbage based on lsm database, solid state hard disk, and storage apparatus | |
| CN107329909B (en) | Data management method and device | |
| CN105630410B (en) | The method for managing file storage area using non-volatile random read-write storage chip dynamic | |
| CN113918087B (en) | Storage device and method for managing namespaces in the storage device | |
| CN105630697B (en) | A kind of storage device using MRAM storage small documents | |
| CN114003169B (en) | Data compression method for SSD | |
| CN117616377A (en) | Superblock allocation across superdevices in ZNS SSDs |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20211209 Address after: 250100 room 2-302-131, building 7, North District, Shandong design creative industrial park, No. 868, Tangye West Road, Jinan area, free trade pilot zone, Jinan, Shandong Province Applicant after: Rongming Microelectronics (Jinan) Co.,Ltd. Address before: 266100 blue Silicon Valley entrepreneurship center, No. 1, Keji Road, Jimo District, Qingdao, Shandong Applicant before: QINGDAO RONGMING SEMICONDUCTOR Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |