CN115964350A - File system management model and system - Google Patents
File system management model and system Download PDFInfo
- Publication number
- CN115964350A CN115964350A CN202211734149.3A CN202211734149A CN115964350A CN 115964350 A CN115964350 A CN 115964350A CN 202211734149 A CN202211734149 A CN 202211734149A CN 115964350 A CN115964350 A CN 115964350A
- Authority
- CN
- China
- Prior art keywords
- file
- management module
- logical
- management
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The file system model divides a file storage system into a first sub-model of a file management module, a second sub-model of object management and a third sub-model of a flash memory medium management module, the file management module gives all attributes of a file to the object, and the file management module converts reading, writing, deleting or rewriting of the file into operation on the object, thereby removing double mapping logic of the file system and the block management module, simplifying IO path, and eliminating double writing amplification in the file system and the block management module. The object storage management module performs storage optimization according to ZNS SSD physical attributes by using inherited file attribute characteristics, such as cold and hot data and different application data which are stored separately, and reduces garbage recycling expense caused by mixed storage of different mode data. Meanwhile, through an optimization algorithm, the calculation acceleration based on the file is directly completed through the object at the storage equipment end, and the stored disposable file is recombined according to the size of the Zone partition space, so that the service efficiency of the Zone space is improved; and the same file is continuously stored to exert the performance advantage when the medium is sequentially accessed, and the overall performance of the storage system is improved.
Description
Technical Field
The invention relates to a file system design method, in particular to a file system based on a flash memory storage medium.
Background
In the background of big data era, the types and the quantity of data information are increased sharply, and new challenges are brought to the design of a storage system. In the current technology, a file system provides storage, query and processing functions of various mixed data, various multimedia data, recorded data, archived data and the like are usually written once during storage, and then files are not modified, management data needs to support read-write, and the current file system is not designed according to the two situations.
With the popularization and application of the SSD, a file system generally adopts a mode of combining random writing and converting into sequential log recording to accelerate the performance of the storage system in order to utilize the performance advantages of the SSD as much as possible, but with the increase of the data mixing degree and the influence of garbage recycling of the file system, the distribution of effective data inside the SSD tends to be dispersed, for the file system, additional overhead is also brought by spatial arrangement and garbage recycling of the file system itself, and the file system constructed on the SSD at present cannot utilize the characteristics of multimedia data, data recording and data archiving, so that the maximum performance of the flash memory medium is exerted.
On the other hand, with the application of a computing storage technology, a storage system can provide more computing functions, but computing requirements are usually related to an application scene, in the existing technical scheme, the computing storage system is built through technologies such as a file system and block management, data storage granularity is a fixed-length logic block, and application scene computing is based on a file level, so that a storage end can only complete computing acceleration based on the data block level, a computing module of the application scene needs to be deployed on a host end in combination with the file system, data storage and computing are separated actually, and the advantage of reducing system overhead by computing storage cannot be fully played.
Disclosure of Invention
The invention provides a technical scheme capable of reducing write amplification of a flash memory storage system, and particularly provides a file system model and a file system which are optimized aiming at application scenes of write-once files and computing storage.
The invention provides a file system model, which at least comprises a first sub-model of a file management module, a second sub-model of object management and a third sub-model of a flash memory medium management module;
the file management module first sub-model is configured to receive a file access request of a user, a file management policy processes the file access request, and the file access request is converted into an object access request of the object management second sub-model;
the second sub-model of object management is configured to provide a continuous logical object area to the first sub-model of file management module and receive the object access request from the first sub-model of file management module, a logical object management policy processes the object access request and converts the object access request into a flash memory read-write-erase operation request to a third sub-model of flash memory medium management module;
the third submodel of the flash memory medium management module provides a read, write and erase interface for the object management second submodel, receives the flash memory read, write or erase operation request from the second submodel, completes the read-write and erase operation of the flash memory medium, and records the running state information and the failure medium information of the flash memory;
the first sub-module of the file management module divides the continuous logic object area provided by the second sub-module of the object management into at least a management area, a readable and writable file area and a disposable file area; the object management second submodel divides the continuous physical address partition into at least a fixed-length logical area and a variable-length logical area.
Further, the fixed-length logical area is configured to store data from the management area of the file system and data of the file-readable-writable area, and the size of the object data in the fixed-length logical area is fixed; the variable-length logical area is configured to store data from the disposable file area of a file system, and the fixed-length logical area and the variable-length logical area determine, according to a user configuration policy, a size of a physical Zone partition actually occupied and whether the physical Zone partition is shared.
Further, the file management policy is configured to: for the disposable file, the occupied logical object ID is recorded in the upper directory file, the directory file records the address mapping of all the disposable files, each mapping table entry comprises three fields of a file name, a logical object ID, size and the like, wherein the file name is used for indexing, the logical object ID points to the actually stored logical object, and the Size represents the Size of the file.
Further, the processing of the object access request by the logical object management policy includes: mapping a logical object to a continuous physical storage space according to a logical object mapping table, wherein the logical object mapping table comprises a plurality of mapping table items; each mapping table item comprises at least two fields of a physical address PBA and a logical object Size, and the logical object mapping table is indexed according to the logical object ID number; the physical address PBA is divided into two parts of Zone encoding and Zone in-Zone migration.
Further, the second object management submodel further comprises an object storage optimization unit, wherein the object storage optimization unit is used for acquiring the ID and Size information of the logical object to form a Size information pool; and combining and summing the Size information in the Size information pool through a preset rule, and comparing and matching the Size information with the Size of the Zone partition.
Further, the Size of the Size information in the Size information pool is A respectively 1 ，A 2 ……，A m ，A N Said Zone partition size is 2 K (ii) a The predetermined rule is configured to include the steps of: 1) Performing a summation operation on any number of the Size of the logical objects, 2) if the result of the summation is less than 2 K Defining the corresponding logic object as a virtual logic object; storing the disposable files corresponding to the objects in the virtual objects in the same Zone partition; if the result of the summation is greater than 2 K And then returning to the step 1); 3) Returning the result to the second sub-model for object management, and updating the corresponding logical object mapping table; wherein k is a positive integer and satisfies 2 K Greater than max (A) 1 ，A 2 ……，A m ，A N )。
Further, the flash memory medium management module organizes the physical addresses of the same operation into continuous Zone partitions according to the physical rule characteristics of flash memory page reading and writing, block erasing and block page sequence writing, wherein the erasing operation is performed according to the Zone partitions, and the writing operation needs to be performed in one Zone partition and the addresses are continuous.
Further, the Zone partitions are sequentially arranged from small to large, and the sizes of adjacent Zone partitions are different.
Further, the user configuration policy comprises at least the following steps: 1) Reading data of a Zone partition corresponding to the readable and writable file into a system memory, and performing reading, writing, deleting or covering operation on the readable and writable file in the memory; 2) And in a preset time interval, carrying out an operation of covering the data of the Zone partition corresponding to the readable and writable file of the fixed-length logic area by the latest readable and writable file in the memory.
Meanwhile, the invention provides a system applying the file system model, which comprises a file management module, a logic object management module and a flash memory medium management module; the system responds to the user request and completes corresponding operation; the operations include:
when a new write-once file request of a user is received, firstly, the file management module carries out variable-length write logical object allocation according to the current available object state, then the logical object management module converts an object ID into a new physical address according to the current write-in point state, then the flash memory medium management module starts file data transmission, writes one-time file data into the converted physical address, feeds back the result to the file management module after the write-in is completed, and the file management module modifies the management area data, records the file name, the allocated corresponding logical object number and the file length into the management area data and completes the file write operation;
when receiving a request for reading a disposable file from a user, firstly, the file management module indexes a logical object number corresponding to the file through management area data, converts the logical object number into an object reading request, sends the corresponding reading request to the logical object management module, then, the logical object management module searches an object address mapping table to obtain a flash memory physical address corresponding to the reading request, initiates a flash memory reading operation, and finally, the flash memory medium management module initiates the flash memory reading operation and transmits the read data back to an application to complete the reading operation;
when a request of rewriting the disposable file is received by a user, firstly, the file management module indexes a corresponding logical object number of the file through management area data, the logical object number is converted into an object writing request, then, the logical object management module converts an object ID into a new physical address according to the current writing point state and makes a waste physical address, then, the flash memory medium management module starts file data transmission, writes the primary file data into the converted physical address, and feeds back the result to the file management module after the writing is finished, the file management module modifies the management area data, records the modified file length and the modified file attribute into the management area data, and finishes the file rewriting operation;
when a request of a user for deleting a disposable file is received, firstly, the file management module indexes a logical object number corresponding to the file through management area data, converts the logical object number into an object recovery request and sends the object recovery request to the logical object management module, the logical object management module invalidates a physical address occupied by an object, background garbage recovery is started when available space is insufficient, the invalidated physical address is recovered for writing operation of a subsequent file, after deletion is completed, a result is fed back to the file management module, the file management module modifies the management area data, the management data corresponding to the file is deleted, and file deletion and writing operation is completed.
Compared with the background art, the invention has the beneficial effects that: the file storage system is divided into three submodels, double mapping logics of the file system and the block management module are removed, IO paths are simplified, double writing amplification inside the file system and inside the block management module is eliminated, all attributes of the files are endowed to the objects, calculation acceleration based on the files can be directly completed through the objects, the function of near data calculation at a storage device end is achieved, meanwhile, the object storage management module can also utilize inherited attribute characteristics of the files to perform storage optimization, for example, cold data and hot data and different application data are stored separately, garbage recycling cost caused by data mixed storage of different modes is reduced, the same file is continuously stored, and performance advantages when media are sequentially accessed are exerted, and the overall performance of the storage system is improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or technical solutions in related arts, the drawings used in the description of the embodiments or prior arts will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
The structures, proportions, and dimensions shown in the drawings and described in the specification are for illustrative purposes only and are not intended to limit the scope of the present disclosure, which is defined by the claims, but rather by the claims, it is understood that these drawings and their equivalents are merely illustrative and not intended to limit the scope of the present disclosure.
FIG. 1 is a diagram of a file system model structure according to the present invention
FIG. 2 is a diagram of the hierarchical architecture of the file storage system of the present invention
FIG. 3 is a diagram of a mapping table of a storage address of a disposable file according to the present invention
FIG. 4 is a logical object memory address map of the present invention
FIG. 5 is a block diagram of the flash memory physical address
FIG. 6 is a flow chart of writing a file according to the present invention
FIG. 7 is a flowchart illustrating a file reading process according to the present invention
FIG. 8 is a flow chart of rewriting files in accordance with the present invention
FIG. 9 is a flowchart illustrating deletion of a file according to the present invention
Detailed Description
The embodiments in this application will be described more fully hereinafter with reference to the accompanying drawings, in which some, but not all embodiments of the invention are shown. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The invention provides a file system model and a file system which can reduce write amplification of a flash memory storage system and are specially optimized for write-once files and application scenes of computing storage.
The invention relates to a file system model and a file system, wherein a file storage system is divided into three submodels, a first submodel of a file management module, a second submodel of object management and a third submodel of flash memory medium management, a concept of a virtual storage object is used for unifying storage granularity and a calculation unit into an object and supporting a variable-length object, the file system is constructed based on the object instead of a logic block, double mapping logic of the file system and the block management module is removed, the size of a file is equal to that of the object, and therefore, all attributes of the file are inherited by the attributes of the object.
The file management module comprises a first sub-model which is responsible for storing files, performing access control and retrieval on the stored files, and organizing and distributing storage space by taking a variable-length object as a management unit, wherein the file system metadata, a directory file and readable and writable files take a fixed-length object as a storage unit, a management method of a traditional file system can be adopted, such as a streaming file mode, a record file mode or a tree file mode or a memory management method, and further preferably a memory management method is adopted. Meanwhile, because the file system data is located in the memory, the access speed is high, and the efficiency of file operation is favorably improved. Aiming at a write-once file, a variable-length object is used as a storage unit, file data is completely stored in an object unit, the file corresponds to the object one by one, a complex logic block mapping table in a file index node is removed, the logical sequence and the physical sequence of data in the file are kept unified, and the file rewrites, deletes and recycles corresponding operations of the corresponding object, so that the first sub-model of the file management module only needs to maintain one current available object bitmap.
The object management second submodel organizes the storage space into a variable-length object, is particularly suitable for a flash memory medium needing ectopic update, realizes data access by mapping the object to a section of physical storage area, allocates a new physical space in sequence when the object is written each time, records the physical address and the storage size corresponding to each object, starts garbage recovery when the storage space is insufficient, intensively stores effective objects, releases a whole continuous physical space for subsequent object writing, and adjusts the size of the continuous physical space based on the characteristics and the actual situation of the flash memory medium module, and the typical value is the size of a plurality of flash memory blocks.
And the flash memory medium management third sub-model is responsible for flash memory failure management, operates the flash memory to perform actual flash memory page reading, flash memory page writing and flash memory block erasing operations, and provides reading, writing and erasing interfaces for the object management second sub-model. The actual physical space of the flash memory medium is organized according to the size of the flash memory block according to the Zone block, the Zone is used as a unit for erasing, random reading is supported, and the operation is continued in the Zone block.
In order to make the aforementioned objects, features and advantages of the present application more comprehensible, the present application is described in further detail with reference to the accompanying drawings and the detailed description.
Example 1
A file system model, as shown in fig. 1, includes at least three submodels, which are a first submodel of a file management module, a second submodel of an object management module, and a third submodel of a flash memory medium management module, respectively;
the file management module comprises a file management module first submodel, a file management module second submodel and a file management module second submodel, wherein the file management module first submodel is configured to receive a file access request of a user, process the file request through a file management strategy and convert the file request into an object operation on the second submodel;
the second submodel of the object management module receives an object access request from the first submodel of the file management module, processes the object request through a logic object management strategy and converts the object request into flash memory read-write operation on a third submodel;
the third submodel of the flash memory medium management module receives the flash memory operation request from the second submodel, completes the read-write operation of the flash memory medium, and is responsible for recording the running state information of the flash memory and managing the information of the failed medium.
As shown in fig. 2, the file management module divides the continuous logical object area provided by the object management module into three regions: a management area, a readable and writable file area and a one-time file area. And the management area or the readable and writable file area is managed by adopting a memory. The management area or the readable and writable file area is read into the memory through the file system, when the data of the management area or the readable and writable file area needs to be modified, the modification operation can be directly carried out at the original position of the data, the whole block does not need to be operated, the repeated erasing and writing of the storage equipment are avoided, the loss of the storage equipment is reduced, and the reduction of the number of writers is facilitated. Meanwhile, because the file system data is located in the memory, the access speed is high, and the efficiency of file operation is favorably improved. The management area is used for storing file system metadata, the readable and writable file area is used for storing directory files and readable and writable files from users, the disposable file area is used for managing disposable files of the users, and the respective logic objects of the three areas are configured according to user requirements.
The logical object management module divides the contiguous physical address partition into two regions: a fixed-length logical area and a variable-length logical area. The size of object data in the fixed-length logical area is fixed and is used for storing management area data and readable and writable file data from the file system, the variable-length logical area stores disposable files of the file system, and the fixed-length logical area and the variable-length logical area determine the size of a physical Zone partition actually occupied and determine whether the physical Zone partition is shared according to a user configuration strategy. The flash memory medium management module organizes a continuous Zone partition according to the physical characteristics of reading and writing by pages of the flash memory, erasing by blocks and writing by page sequence in the blocks.
Further, for a one-time file, as shown in fig. 3, the file management policy is configured to: for the disposable file, the occupied logical object ID is recorded in the upper directory file, the directory file records the address mapping of all the disposable files, each mapping table entry comprises three fields of a file name, a logical object ID, size and the like, wherein the file name is used for indexing, the logical object ID points to the actually stored logical object, and the Size represents the Size of the file.
Further, the processing of the object access request by the logical object management policy includes: mapping a logical object to a continuous physical storage space according to a logical object mapping table, wherein the logical object mapping table comprises a plurality of mapping table items; each mapping table item comprises at least two fields of a physical address PBA and a logical object Size, and the logical object mapping table is indexed according to the logical object ID number; the physical address PBA is divided into two parts, i.e. Zone encoding and Zone in-Zone migration, as shown in fig. 4 and fig. 5.
The second sub-model for object management may further include an object storage optimization unit, where the object storage optimization unit is configured to obtain the ID and the Size information of the logical object, and form a Size information pool; size information Size in the Size information pool is A respectively 1 ，A 2 ……，A m ，A N Said Zone partition size is 2 K (ii) a And combining the Size information in the Size information pool through a preset rule, comparing and matching the Size information with the Size of the Zone partition, and then classifying and storing the Size information. Through the operation, the storage states of the disposable files with different sizes can be optimized, at least one disposable file with the maximum capacity is stored according to the size of the Zone partition, the utilization efficiency of the Zone partition space is improved, the use expense of a garbage recycling function is reduced, and the storage state of the disposable files with different sizes is improvedThe execution efficiency of the system is improved
The predetermined rule comprises the steps of: 1) Performing a summation operation on the Size of any number of the logical objects, 2) if the summation result is less than 2 K Defining the corresponding logic object as a virtual logic object; storing the disposable files corresponding to the objects in the virtual objects in the same Zone partition; if the result of the summation is greater than 2 K And then returning to the step 1); 3) Returning the result to the second sub-model for object management, and updating the corresponding logical object mapping table; wherein k is a positive integer and satisfies 2 K Greater than max (A) 1 ，A 2 ……，A m ，A N )。
When the file system model is used, the system comprises a file management module, a logic object management module and a flash memory medium management module; the system responds to the user request and completes corresponding operation. The operations include: a new write-once file request, a read-once file request, an overwrite-once file request, and a delete-once file request.
As shown in fig. 6, when a new write-once file request from a user is received, the file management module first allocates a variable-length write logical object according to the current available object state, then the logical object management module converts the object ID into a new physical address according to the current write-in point state, then transmits file data, writes the file data into the converted physical address, and finally modifies the management area data, records the file name, the allocated corresponding logical object number, and the file length into the management area data, and completes the file write operation.
As shown in fig. 7, when a request for reading a disposable file is received from a user, the file management module first indexes a logical object number corresponding to the file through management area data, converts the logical object number into an object read request, sends the corresponding read request to the logical object management module, then the logical object management module searches an object address mapping table to obtain a flash memory physical address corresponding to the read request, initiates a flash memory read operation, and finally, the flash memory medium management module initiates a flash memory read operation and transmits the read data back to an application to complete the read operation.
As shown in fig. 8, when receiving a request for rewriting a disposable file from a user, the file management module first indexes a logical object number corresponding to the file through the management area data to convert the logical object number into an object write request, then the logical object management module converts an object ID into a new physical address according to a current write-in point state and invalidates the old physical address, then transmits the file data to write the converted physical address, and finally modifies the management area data, records the modified file length and attribute into the management area data, and completes the operation of rewriting the file.
As shown in fig. 9, when a request for deleting a disposable file is received from a user, the file management module first indexes a logical object number corresponding to the file through management area data, converts the logical object number into an object recovery request, and sends the object recovery request to the logical object management module, the logical object management module invalidates a physical address occupied by an object, background garbage recovery is started when an available space is insufficient, the invalidated physical address is recovered for subsequent file writing operation, and finally, the management area data is modified, the management data corresponding to the file is deleted, and the operation of deleting and writing the file is completed.
The file storage system is divided into three submodels, double mapping logics of the file system and the block management module are removed, IO paths are simplified, double writing amplification inside the file system and inside the block management module is eliminated, all attributes of the files are endowed to the objects, calculation acceleration based on the files is directly completed through the objects, the function of completing near data calculation at a storage device end is achieved, meanwhile, the object storage management module can also utilize inherited file attribute characteristics to perform storage optimization, for example, cold data and hot data and different application data are stored separately, garbage recycling cost caused by data mixed storage of different modes is reduced, the same file is continuously stored, and performance advantages when media are sequentially accessed are exerted, and the overall performance of the storage system is improved.
The embodiments in the present description are described in a progressive manner, or in a parallel manner, or in a combination of a progressive manner and a parallel manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments can be referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
It should be noted that in the description of the present application, it is to be understood that the terms "upper", "lower", "top", "bottom", "inner", "outer", and the like, indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, are only used for convenience in describing the present application and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be configured and operated in a specific orientation, and thus, should not be construed as limiting the present application. When a component is referred to as being "connected" to another component, it can be directly connected to the other component or intervening components may also be present.
It is further noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that an article or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such article or apparatus. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in an article or apparatus that comprises the element.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A file system model at least comprises a first submodel of a file management module, a second submodel of object management and a third submodel of a flash memory medium management module;
the file management module first sub-model is configured to receive a file access request of a user, a file management policy processes the file access request, and the file access request is converted into an object access request of the object management second sub-model;
the second sub-model of object management is configured to provide a continuous logical object area to the first sub-model of file management module and receive the object access request from the first sub-model of file management module, a logical object management policy processes the object access request and converts the object access request into a flash memory read-write operation request to a third sub-model of flash memory medium management module;
the third submodel of the flash memory medium management module provides a read, write and erase interface for the object management second submodel, receives the flash memory read, write or erase operation request from the second submodel, completes the read-write and erase operation of the flash memory medium, and records the running state information and the failure medium information of the flash memory;
the file management module first sub-module divides the continuous logic object area provided by the object management second sub-module into at least a management area, a readable and writable file area and a disposable file area; the object management second submodel divides the continuous physical address partition into at least a fixed-length logical area and a variable-length logical area.
2. The file system model of claim 1, wherein: the fixed-length logical area is configured to store data from the management area and data of the writable and readable file area of a file system, and the size of the object data in the fixed-length logical area is fixed; the variable-length logical area is configured to store data from the one-time file area of the file system, and the fixed-length logical area and the variable-length logical area determine, according to a user configuration policy, whether to actually occupy a physical Zone partition size and to share the physical Zone partition.
3. The file system model of claim 1, wherein: the file management policy is configured to: the logical object ID of the disposable file is recorded in a directory file at the upper layer, the directory file records the address mapping of all the disposable files, each mapping table entry in the address mapping comprises three fields of a file name, a logical object ID, size and the like, wherein the file name is used for indexing, the logical object ID points to the actually stored logical object, and the Size represents the file Size.
4. The file system model of claim 1, wherein: the processing of the object access request by the logical object management policy comprises: mapping a logical object to a segment of continuous physical storage space according to a logical object mapping table, wherein the logical object mapping table comprises a plurality of mapping table items; each mapping table item comprises at least two fields of a physical address PBA and a logical object Size, and the logical object mapping table is indexed according to the logical object ID number; the physical address PBA is divided into two parts, zone encoding and in-Zone migration.
5. The file system model according to claim 3 or 4, characterized in that: the second submodel for object management further comprises an object storage optimization unit, wherein the object storage optimization unit is used for acquiring the ID and the Size information of the logical object to form a Size information pool; and combining and summing the Size information in the Size information pool through a preset rule, and then matching the Size information with the target Zone partition.
6. The file system model of claim 5, wherein: size information Size in the Size information pool is A respectively 1 ，A 2 ……，A m ，A N The size of the target Zone partition is 2 K (ii) a The predetermined rule is configured to include the steps of: 1) Performing a summation operation on the Size of any number of the logical objects, 2) if the summation result is less than 2 K Defining the corresponding logic object as a virtual logic object; storing the disposable files corresponding to the objects in the virtual logic objects in the same Zone partition; if the result of the summation is greater than 2 K And then returning to the step 1); 3) Returning the result to the second sub-model for object management, and updating the corresponding logical object mapping table; wherein k is a positive integer and satisfies 2 K Greater than max (A) 1 ，A 2 ……，A m ，A N )。
7. The file system model of claim 1, wherein: the flash memory medium management module organizes physical addresses of the same operation into continuous Zone partitions according to the physical rule characteristics of flash memory page reading and writing, block erasing and block page sequence writing, wherein the erasing operation is performed according to the Zone partitions, and the writing operation needs to be in one Zone partition and the addresses are continuous.
8. The file system model of claim 7, wherein: the Zone partitions are sequentially arranged from small to large, and the sizes of the adjacent Zone partitions are different.
9. The file system model of claim 1, wherein: the user configuration policy at least comprises the following steps: 1) Reading data of the Zone partition corresponding to the readable and writable file into a system memory, and performing reading, writing, deleting or covering operation on the readable and writable file in the memory; 2) And in a preset fixed time interval, carrying out an operation of covering the data of the Zone partition corresponding to the readable and writable file of the fixed-length logic area by the latest readable and writable file in the memory.
10. A system applying the file system model according to any one of claims 1-9, the system comprising a file management module, a logical object management module, and a flash media management module; the system responds to the user request and completes corresponding operation; the operations include:
when a new write-once file request of a user is received, firstly, the file management module carries out variable-length write logical object allocation according to the current available object state, then the logical object management module converts an object ID into a new physical address according to the current write-in point state, then the flash memory medium management module starts file data transmission, writes one-time file data into the converted physical address, feeds back the result to the file management module after the write-in is completed, and the file management module modifies the management area data, records the file name, the allocated corresponding logical object number and the file length into the management area data and completes the file write operation;
when receiving a request for reading a disposable file from a user, firstly, the file management module indexes a logical object number corresponding to the file through management area data, converts the logical object number into an object reading request, sends the corresponding reading request to the logical object management module, then, the logical object management module searches an object address mapping table to obtain a flash memory physical address corresponding to the reading request, initiates a flash memory reading operation, and finally, the flash memory medium management module initiates the flash memory reading operation and transmits the read data back to an application to complete the reading operation;
when a request of rewriting the disposable file is received by a user, firstly, the file management module indexes a corresponding logical object number of the file through management area data, the logical object number is converted into an object writing request, then, the logical object management module converts an object ID into a new physical address according to the current writing point state and makes a waste physical address, then, the flash memory medium management module starts file data transmission, writes the primary file data into the converted physical address, and feeds back the result to the file management module after the writing is finished, the file management module modifies the management area data, records the modified file length and the modified file attribute into the management area data, and finishes the file rewriting operation;
when a request of deleting a disposable file is received by a user, firstly, the file management module indexes a corresponding logical object number of the file through management area data, converts the logical object number into an object recovery request and sends the object recovery request to the logical object management module, the logical object management module invalidates a physical address occupied by an object, background garbage recovery is started when an available space is insufficient, the invalidated physical address is recovered for writing operation of a subsequent file, a result is fed back to the file management module after deletion is completed, the file management module modifies the management area data, deletes the management data corresponding to the file, and the operation of deleting and writing the file is completed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211734149.3A CN115964350A (en) | 2022-12-30 | 2022-12-30 | File system management model and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211734149.3A CN115964350A (en) | 2022-12-30 | 2022-12-30 | File system management model and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115964350A true CN115964350A (en) | 2023-04-14 |
Family
ID=87354572
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211734149.3A Pending CN115964350A (en) | 2022-12-30 | 2022-12-30 | File system management model and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115964350A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117056245A (en) * | 2023-08-18 | 2023-11-14 | 武汉麓谷科技有限公司 | Data organization method for log record application based on ZNS solid state disk |
-
2022
- 2022-12-30 CN CN202211734149.3A patent/CN115964350A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117056245A (en) * | 2023-08-18 | 2023-11-14 | 武汉麓谷科技有限公司 | Data organization method for log record application based on ZNS solid state disk |
| CN117056245B (en) * | 2023-08-18 | 2024-02-23 | 武汉麓谷科技有限公司 | Data organization method for log record application based on ZNS solid state disk |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110825748B (en) | High-performance and easily-expandable key value storage method by utilizing differentiated indexing mechanism | |
| CN107180092B (en) | File system control method and device and terminal | |
| US9043334B2 (en) | Method and system for accessing files on a storage system | |
| JP6205650B2 (en) | Method and apparatus utilizing non-uniform hash function to place records in non-uniform access memory | |
| CN103838853B (en) | Mixed file system based on different storage media | |
| CN111221776A (en) | Method, system and medium for implementing file system facing nonvolatile memory | |
| TW201935243A (en) | SSD, distributed data storage system and method for leveraging key-value storage | |
| KR20110039417A (en) | Apparatus, system, and method for efficient mapping of virtual and physical addresses | |
| CN107562367B (en) | Method and device for reading and writing data based on software storage system | |
| CN114546296B (en) | ZNS solid state disk-based full flash memory system and address mapping method | |
| US9430492B1 (en) | Efficient scavenging of data and metadata file system blocks | |
| KR20200122994A (en) | Key Value Append | |
| CN111143285A (en) | Small file storage file system and small file processing method | |
| CN110968269A (en) | SCM and SSD-based key value storage system and read-write request processing method | |
| CN113535670B (en) | Virtual resource mirror image storage system and implementation method thereof | |
| CN113568582A (en) | Data management method and device and storage equipment | |
| CN114356246B (en) | Storage management method and device for SSD internal data, storage medium and SSD device | |
| JP2018181202A (en) | Device, method, and program for storage control | |
| WO2022262381A1 (en) | Data compression method and apparatus | |
| CN115964350A (en) | File system management model and system | |
| CN103942301A (en) | Distributed file system oriented to access and application of multiple data types | |
| JP6531574B2 (en) | Storage device, storage device control program and storage device control method | |
| KR20090007926A (en) | Apparatus and method for managing index of data stored in flash memory | |
| WO2024108939A1 (en) | Multi-level mapping framework and data operation request processing method and system | |
| US10185660B2 (en) | System and method for automated data organization in a storage system |
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 |