CA2550871A1

CA2550871A1 - File system having inverted hierarchical structure

Info

Publication number: CA2550871A1
Application number: CA002550871A
Authority: CA
Inventors: Dan Dodge
Original assignee: Individual
Current assignee: BlackBerry Ltd
Priority date: 2005-07-01
Filing date: 2006-06-23
Publication date: 2007-01-01
Anticipated expiration: 2026-06-23
Also published as: CN1983265A; JP2007012060A; US8959125B2; KR20070003577A; US20070005615A1; EP1739575A2; CA2550871C; EP1739575A3; EP1739575B1

Abstract

A computer system that improves upon the traditional file system hierarchy is disclosed. The computer system includes a processor and a persistent data storage device that is accessible by the processor. The system also may include file system software that is executed by the processor. The file system software may operate to organize the files of the persistent data storage device into a plurality of records that are contained in a metafile. within the metafile, the files and directories of the file system may be organized in an inverted hierarchical structure. In the inverted hierarchical structure, records corresponding to child files and child directories of a parent directory may directly point to a record in the metafile that corresponds to the parent directory. In one implementation of the file system, individual records of the metafile may be each addressable by a file identifier. Each of the individual records, in turn, may include a parent identifier that corresponds to the file identifier of its parent directory record.

Description

FILE SYSTEM HAVING INVERTED HIERARCHICAL STRUCTURE
IN V ENTOR[S]
Dan Dodge BACKGROUND OF THE INVENTION
1. Technical Field.
[0001] This invention is generally directed to a file system for use in a computer, embedded controller, or the like. More particularly, this invention is directed to a file system to having an inverted hierarchical structure.

2 . Related Art.
[0002] Computers, embedded controllers, and other microprocessor based systems are typically constructed from a variety of different hardware components. The hardware components may include a processor, I/O devices, human interface devices, and the like.
~ 5 Additionally, such systems use memory storage units to maintain the data used in the system.
The memory storage units may take on a variety of different forms including, but not limited to, hard disk drives, floppy disk drives, random access memory, flash memory, and the like.

[0003) High-level application programs that are executed in such systems must often interact seamlessly with these hardware components, including the memory storage units. To 2o this end, many systems run an operating system that acts as an interface between the application programs and the system hardware. File system software may be included as part of the operating system, or it may be provided as an ancillary software component that interacts with the operating system. In either instance, the file system software organizes the data within the memory storage units for ready access by the processor and the high-level 2s application programs that the processor executes.

[0004] In a traditional file system, the file system software organizes the data within the memory storage units in a hierarchical structure that includes data files and corresponding containers, called directories. The traditional hierarchical structure is in the form of an upside-down tree having a starting directory that is called the root directory. Additional directories and files may depend from the root directory. The additional directories, in turn, may include further subdirectories and files.

[0005] The file system software maintains one or more metafiles that include records used to keep track of the name and attributes of each file/directory entry in the file system.
These attributes include, for example, whether the entry is a data file or directory, read only file, hidden file, system file, and the like. The information needed to physically access file data that is stored in the memory storage unit also may be contained in or pointed to by the records in the metafiles.

[0006] When the file system is organized using the traditional file system hierarchy, to the processor must update a substantial number of different records in the metafiles when any change is made to a data file or directory. Updating of the metafile records must occur, for example, whenever a file or directory is created, renamed, or moved from one directory to another. The substantial number of different records in the metafiles that must be updated under these circumstances increases the likelihood that an update to a given record will not t 5 occur if, for example, the system experiences a power loss or other hardware or software failure. Further, reconstruction of the file system after such a failure may be complicated and time-consuming when the files are organized in the traditional manner.
Accordingly, an alternative to the traditional file system hierarchy is needed.

SUMMARY

[0007] A computer system that improves upon the traditional file system hierarchy is disclosed. The computer system includes a processor and a persistent data storage device that is accessible by the processor. The system also may include file system software that is executed by the processor. The file system software may operate to organize the files of the persistent data storage device into a plurality of records that are contained in a metafile.
Within the metafile, the files and directories of the file system may be organized in an inverted hierarchical structure. In the inverted hierarchical structure, records corresponding to child files and child directories of a parent directory point directly to a record in the to metafile that corresponds to the parent directory. In one implementation of the file system, individual records of the metafi(e may be each addressable by a file identifier. Each of the individual records, in turn, may include a parent identifier that corresponds to the file identifier of its parent directory record.

(0008] Transactions relating to a file/directory represented by a record in the metafile may be executed atomically, if desired, without a corresponding modification of other related parent or child records. In an optimized implementation, only a single metafile record may be updated wherein a file or directory is created, renamed, or moved from one directory to another.

[0009] The inverse hierarchical structure of the file system also may be readily expanded as needed to increase the amount of information associated with a file or directory record in the metafile. For example, each metafile record may include an extended file identifier that points to a corresponding extended file record that may be stored in the metafile. The extended record may include additional information corresponding to the file or directory that cannot fit within the memory space allocated to a single file or directory record in the metafile.

(0010] The inverted hierarchical structure of the file system is also well-suited for use in a transaction based system, particularly one in which the transaction file for the system is stored on flash media. In one such transaction based system, a reconstructed version of the file system may be generated in random access memory during system startup.
The 3o reconstructed file system may be generated, at least in part, from the file and directory records stored in the metafile. The records in the metafile are organized in an inverted hierarchical structure, which may be used to facilitate efficient startup operations and optimize the generation of the reconstructed file system.

[0011] Other systems, methods, features and advantages of the invention will be, or will become, apparent to one with skill in the art upon examination of the following figures and detailed description. It is intended that all such additional systems, methods, features and advantages be included within this description, be within the scope of the invention, and be protected by the following claims.

BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The invention can be better understood with reference to the following drawings and description. The components in the figures are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the invention. Moreover, in the figures, like referenced numerals designate corresponding parts throughout the different mews.

[0013] Figure 1 is a block diagram of a computer system that may implement a file system having an inverted hierarchical structure.

[0014] Figure 2 is a tree diagram showing one example of an arrangement of files and to directories within file system having an inverted hierarchical structure.

[0015] Figure 3 is a block diagram illustrating one manner in which records of a metafile may be arranged to implement a file system having an inverted hierarchical structure.

[0016] Figure 4 illustrates one manner of logically arranging a transaction record in a t 5 transaction file of a transaction based computer system.

[0017] Figure 5 shows the physical arrangement of memory in one type of flash media device.

[0018] Figures 6 and 7 illustrate various manners in which transaction records may be arranged in flash media devices.
20 [0019] Figure 8 illustrates a number of interrelated processing steps that may be used to generate an extents pool that, in turn, is employed in a reconstructed file system that is created by the computer system during startup.
[0020] Figures 9 through l l are directed to exemplary formats for various record types used in the processing steps shown in Figure 8.
25 [0021] Figure 12 is directed to an exemplary format for a directory node record of the regenerated file hierarchy used in the reconstructed file system.
[0022] Figure 13 is directed to an exemplary format for a file node record of the regenerated file hierarchy used in the reconstructed file system.
[0023] Figure 14, illustrates a number of interrelated processing steps that may be 3o used to construct the regenerated file hierarchy used in the reconstructed file system.

[0024] Figure 15 is a logical representation of a reconstructed file system that has been generated in the manner set forth in connection with Figures 8 through 14 as applied to the exemplary file and directory arrangement shown in Figure 2.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0025] Figure I illustrates the components that may be employed in an exemplary computer system 10. As shown, the exemplary system 10 includes a processor I5, read only memory 20, and a persistent storage unit 30. Computer system 10 also may include random access memory 35, an I/O interface 40, and a user interface 45. The specific components that are used in computer system 10 may be tailored to the particular functions) that are to be executed by the computer system 10. Accordingly, the presence or absence of a component, other than processor 15, may be specific to the design criterion imposed on the computer system 10. For example, user interface 45 may be omitted when the computer system 10 o takes the form of an embedded controller or the like.
[0026] Read only memory 20 may include operating system code 43 that controls the interaction between high-level application programs executed by the processor 15 and the various hardware components, including memory devices 20 and 35, the persistent storage unit 30, and the interface devices 40 and 45. The operating system code 43 may include file system software for organizing files stored on the persistent storage unit 30.
Alternatively, the file system software may be provided as a separate software component that merely interacts with the operating system code 43. In the latter case, the code corresponding to the file system software may be stored in read only memory 20, persistent storage unit 30 or the like.
When computer system 10 is networked with other computers and/or storage devices through 2o I/O interface 40, the file system software may be stored remotely and downloaded to computer system 10 as needed. Figure 1, however, illustrates storage of the file system software 47 in read only memory 20.
[0027] The persistent storage unit 30 may take on any number of different forms. For example, the persistent storage unit 30 may take the form of a hard disc drive, floppy disk drive, etc. It also may be in the form of a non-rotating media device, such as non-volatile memory implemented in an integrated circuit format (e.g., flash memory, and the like). Still further, persistent storage unit 30 need not be limited to a single memory structure. Rather, the persistent storage unit 30 may include a number of separate storage devices of the same type (e.g., all flash memory) and/or separate storage devices of different types (e.g., one or 3o more flash memory units and one or more hard disk drives).

[0028] The files stored in the persistent storage unit 30 include data that is interpreted in accordance with a predetermined format used by an application program or by the operating system code 43. For example, the data stored within a file may constitute the software code of an executable program, the ASCII text of a database record, data corresponding to transactions executed (or not executed) by computer system 10, or the like.
[0029] The file system software 47 organizes the files stored on the persistent storage unit 30 using an inverted hierarchical structure. Figure 2 is a diagram showing one manner in which the inverted hierarchical structure, shown generally at 50, may be implemented. In the traditional hierarchical structures used by many file systems, the top level of the file structure t0 begins with the root directory and each directory points downward to the files and subdirectories contained within the directory. In the exemplary inverted hierarchical structure 50, however, the child files and child directories contained within a parent directory point upward to the parent directory. Depending on where the file system begins its organization, the root directory may constitute the lowest level of the file system structure.
[0030] The exemplary inverted hierarchical structure 50 includes five files 55, 60, 65, 70 and 75 at the highest level of the file system structure. Files 55, 60 and 65 are contained within directory 80 while files 70 and 75 are contained within directory 85.
Accordingly, the file system software 47 organizes the file system so that the file system records representing child files 55, 60 and 65 point to the record for their parent directory 80.
Similarly, file 2o system records representing child files 70 and 75 point to the record for their parent directory 85.
[0031] At the next level of the exemplary inverted hierarchical structure 50, files 90 and 95 as well as directory 80 are contained within directory 100, while directory 85 may be contained within directory 105. Accordingly, the file system software 47 organizes the file system so that file system records representing child directory 80 and child files 90 and 95 point to the record for their parent directory 100. Similarly, the file system record representing child directory 85 points to the record for its parent directory 105.
[0032] The root directory 110 may form the trunk of the inverted hierarchical structure 50. In this example, directories 100 and 105 and file 115 are contained within the root directory 110. Accordingly, the file system software 47 organizes the file system so that file system records representing child directories 100 and 105 and child file 1 15 point to the record for their parent directory 105.

[0033] One manner in which the file system software 47 may organize the records of the file system to implement an inverted hierarchical structure is shown in Figure 3. In this implementation of the file system, the file system software 47 may generate one or more metafiles that include records corresponding to each file and directory used in the file system.
Figure 3 shows a single metafile 120 and an exemplary manner in which the records within the metafile 120 may be arranged and formatted. In this example, metafile 120 may be arranged as a table that includes a plurality of equal length record entries 125. Each record entry 125 corresponds to a single file or directory used in the file system. A
unique file identifier, such as the one shown at 130, may be used by the file system software 47 to o address a corresponding record 125 of the metafile 120. If each record entry 125 has the same record length, the format for the file identifier 130 may be chosen so that it may be used, either directly or indirectly, as an index to the desired record in metafile 120. For example, file identifier 130 may constitute an offset value that is used along with the memory address location of the first record of metafile 120 to calculate the memory address location of the ~s first byte ofthe metafile record having the desired directory/file information.
[0034] In the example of Figure 3, the file identifier 130 is pointing to record 135 (Entry 7) in metafile 120. Record 135 is shown in Figure 3 in an expanded form adjacent to the metafile 120. The expanded form of record 135 also illustrates a basic record format that may be used for each record entry 125. In this example, record 135 includes a number of 2o different fields containing information relating to the file or directory represented by the record. This information, among other things, corresponds to the logical location of the file or directory within the structure of the file system.
[0035] The inverted hierarchical structure of the file system may be implemented by employing a metafile record format in which each metafile record includes a pointer to the 2s metafile record representing its parent directory. Figure 3 shows a metafile record format in which each metafile record includes a parent identifier field 140 that stores the file identifier of its parent directory. In this example, the parent record identifier 140 of metafile record 135 corresponds to the file identifier used to address record 145 (Entry 9).
Record 145, in turn, includes information pertaining to the directory containing the file or directory 3o represented by record 135.
[0036] Each metafile record also may include other information pertaining to the directory or file that the record represents. In the exemplary record format of record 135, a number of different information fields are employed. The information fields include a mode field 150, user identification field 155, group identification field 160, access time field 165, modified time field 170, created time field 175, file size field 180 and short name field 185.
The mode field 150 may be used to determine whether the file or directory represented by the record is a system file/directory, a hidden file/directory, a read only file/directory, etc. The user identification field 155 and group identification field 160 contain information relating to user and group ownership of the represented file or directory. The access time field 165, modified time field 170, and created time field 175 contain information relating to the time at which the represented file or directory was last accessed, the time at which the represented to file or directory was last modified and the time at which the represented file or directory was created, respectively. The size field 185 contains information on the size of the file represented by the record and is zero for directory records. Finally, the short name field 185 contains ASCII characters representing the short text name of the corresponding file or directory. The length of the short name field 185 may be chosen, for example, to conform to is the POS1X standard. Additionally, each record may include hash values and/or name sums that correspond to the short name. Such hash values and/or name sums may be used by the file system software 47 to quickly search for a particular directory and/or file record.
[0037] Each record in metafile 120 also may include a field for an extended record identifier 190. The extended record identifier 190 may be used as a file identifier that points 2o to an extended record in the metafile 120. The extended record may contain further information for the file or directory represented by the record and may be particularly useful in instances in which all of the information pertaining to a particular file or directory does not fit within the memory space allocated for a single metafile record.
[0038] Figure 3 illustrates one manner in which an extended record identifier 25 may be used. In this example, the extended record identifier 190 of record 135 corresponds to the file identifier (fid) used to access record 195 (Entry 11) in metafile 120. An exploded view of record 195 is shown adjacent the exploded view of record 135 in Figure 3. This exploded view illustrates one record format that may be used for the extended record. As shown, each extended record may include its own parent identifier field 200.
The parent 3o identifier field 200 of an extended record, however, corresponds to the file identifier of the record which points to the extended record. In the example shown in Figure 3, the contents ofthe parent identifier field 200 may be used to point back to record 135 (Entry 7).

[0039] In those instances in which the memory space allocated for two record entries is insufficient to hold all of the information pertaining to a file or directory, the extended record 195 may point to yet a further extended record using its own extended record identifier, such as the one included in field 205 of record 195. Although the format for the s further extended record pointed to by extended file identifier 125 is not shown, the further extended record may likewise include a parent record identifier that points back to record 195.
[0040] The type of information included in an extended record may vary between file systems. In Figure 3, the extended record 195 includes a long name field 210 that contains to ASCII characters corresponding to the text of the long name of the file or directory represented by the record 135. Further fields may be reserved in an expansion area 215 of each extended record, such as record 195, to store additional information relating to the corresponding file or directory.
[0041] In the previous example, the extended records used by the file system are is stored in metafile 120. However, the extended records and any further extended records may alternatively be stored in a separate metafile, multiple metafiles, and the like. The separate metafile(s) need not share the same storage medium with metafile 120 nor with each other.
Rather, the metafiles may be stored in different storage media accessible to processor 15.
Even the basic metafile records (directory and file records that do not have corresponding 2o extended records) may be distributed among multiple files and/or multiple storage media. As such, although the metafile records of the exemplary system are stored in a single metafile, the metafile may alternatively be in the form of many individual files on the same or different storage media.
[0042] By organizing the files and directories of computer system 10 in an inverted 25 hierarchical structure, it becomes possible to realize one or more file system advantages. For example, the file system is capable of being implemented in any manner in which typical file and directory transactions (i.e., moving a file/directory, deleting a file/directory, creating a file/directory, copying a file/directory) are accomplished atomically as a change, addition or deletion of a single metafile record. In this implementation, for example, the file/directory 3o represented by record 135 may be moved to another directory in the hierarchy merely by changing the parent identifier 140 so that it points to the metafile record for the new parent directory. This may be accomplished with a single write operation to record 135 in the metafi le 120.
[0043] The inverted hierarchical structure may be employed to optimize a transactional or log-based system. An exemplary transactional or log-based system may be constructed from the components shown in Figure 1. In this example, a transaction file 220 may be maintained in the persistent storage unit 30 and may be used to keep records of the transactions associated with each file and directory of the file system.
Updates to the file system are committed atomically based on the transaction records contained in transaction file 220. In one of its simplest form, every transaction record may be stored as a single logical to page that may be mapped to a physical block or sector of the persistent storage unit 30.
[0044] One manner in which a transaction record 225 may be formatted for use in computer system 10 is shown in Figure 4. Generally stated, each transaction record 225 of the transaction file 220 includes a header field 230 and a corresponding data field 235. The header field 230 may include a number of different sub-fields. The sub-fields shown in Figure 4 include a transaction sequence field 240, a file identification field 245, a transaction status field 250, a cluster high field 255, a cluster low field 260 and number of clusters field 265. Additionally, further sub-fields may be included in header 230 to verify the integrity of the transaction and for error correction. These further sub-fields include a cluster sum field 247, a transaction sum field, an error correction code field 257 to check and correct header 2o 230, an error correction code field 259 to check and correct data 235, and a further status field 262 indicative of the condition of the memory locations in which the transaction record may be stored.
[0045] Each of the sub-fields of header field 230 has a meaning to the file system software 47. In this example, the transaction sequence field 240 may be a monotonically increasing transaction identifier that is assigned by the file system software 47. When a new transaction record is added to the transaction file 220, the value stored in the transaction sequence field 240 of the new record may be increased by a predetermined amount over the value of the transaction sequence field of the chronologically preceding transaction record.
Consequently, transaction records having larger transaction identifier values are considered to 3o have been added to the transaction file 220 later in time than transaction records having lower transaction identifier values. This chronological sequencing of the transactions, as represented by the value of the transaction sequence field 240 (and, in certain circumstances, the position of the transaction record within a block of the transaction file 220), allows the file system software 47 to apply (i.e., commit) the transactions in the proper order to maintain the integrity of the file system contents. Other ways of keeping track of the chronological sequencing of the transactions also may be used.
[0046] File system software 47 uses the transaction status field 250 to determine whether the transaction of a transaction record 225 has been committed. Once a transaction has been committed, further alteration of the committed transaction record 225 may be inhibited by the file system software 47. This ensures consistency of the file system and also allows the file system to store the transaction file 220 in, for example, write-once media, flash to media, or the like.
[0047] The file identification field 245 of header 230 identifies the file that is affected by the transaction record 225. The format for the file identification field 245 may be selected so that it is the same as the file identifiers used in the metafile records.
The cluster high field 255 and cluster low field 260 may be used by the file system software 47 to determine the starting address (or offset) at which the data 235 is to be written into the identified file while the number of clusters field 265 may be used to determine how many clusters of the identified file are to be overwritten by the data 235.
[0048] As noted above, persistent storage unit 30 may include one or more flash memory devices. Flash memory devices store information in logic gates, called "memory 2o cells," each of which typically stores one bit of information. More recent advances in flash memory technology have also enabled such devices to store more than 1 bit per cell, sometimes referred to as mufti-level cell devices. Additionally, flash memory is non-volatile, which means that the contents of memory cells are not lost when power is withdrawn from the device.
[0049] Although flash device technology is continuously evolving, dominant technologies include NAND flash memory and NOR flash memory. NOR flash devices and HAND flash devices generally differ in the type of logic gate used for each storage cell. An exemplary logical architecture 270 of one type of NAND flash memory device 275 is shown in Figure 5. As illustrated, the available memory on the device 275 may be organized into 3o contiguous physical blocks 280 each having an equal number of memory cells (i.e., 16K
bytes). NAND flash memory device 275 further divides each of the contiguous blocks 280 into a specific number of physical sectors or pages 290. Each physical page 290, in turn, may be further divided into a data area 295 and spare area 300. The data area 295 is normally reserved for storage of data, while the spare area 300 is typically reserved for maintenance of meta-information about the data stored in data area 295. The meta-information may include, for example, error-correcting codes used for verification and correction of sector contents, cyclic redundancy check data, and the like.
[0050] NOR flash devices have an architecture similar to that shown in Figure 5, except that the spare areas of each page are located on opposite sides of the data area. NOR
flash devices also offer random access read and programming operations, allowing individual memory locations to be read on or read. However, once a memory location in a block has to been written, NOR flash devices do not allow the block to be rewritten a smaller granularity than a block. Likewise, NOR flash devices do not allow erase operations at a smaller granularity than a block.
[0051] The data area 295 and spare area 300 are typically set to specific sizes in both NOR and NAND flash devices. For example, each page 290 of the exemplary NAND
flash ~s device 275 of Figure 5 includes a data area 295 of 512 bytes and a spare area 300 of 16 bytes for a total page size of 528 bytes. The NAND flash device 275 also employs 32 pages 290 per block 280. Other page sizes may be used in computer system 10 and are commercially available. For example, many NAND devices include blocks having 64 pages where each page stores 21 12 bytes so that the total data area per page is 2048 bytes and the spare area per 2o page is 64 bytes.
[0052] Flash memory devices, such as NAND flash device 275, typically perform erase operations on an entire block 280 of memory at a time. An erase operation sets all bits within the block 280 to a consistent state, normally to a binary "1" value.
Programming operations on an erased block 280 of flash device 275 can only change the contents of an 25 entire page 290 (although NOR flash devices may be programmed in a slightly different manner). Once a page 290 of a NAND flash device is programmed, its state cannot be changed further until the entire block 280 may be erased again. Reading of the contents of flash device 275 also occurs at the page level.
[0053] Figure 6 illustrates one manner in which transaction records may be organized 3o in a flash memory device, such as NAND flash device 275. In this example, each transaction record 305 may be comprised of two or more contiguous logical pages 315. Each logical page 315, in turn, may be comprised of two or more contiguous physical pages 290 of a block 280 of device 275. Meta-data information for the transaction record 310 may be stored in spare area 300, and may include some of the fields described in connection with header 230 of Figure 4. Depending on the size of the spare area 300 of each page 290, the meta-data information may be divided among multiple spare areas 300 of the transaction record 310. A
division of the meta-data information between the spare areas 300 of two consecutive physical pages 290 is shown in Figure 6. The transaction records shown in Figure 6 also may be organized so that each transaction 310 corresponds to a single logical page 315 that, in turn, may be comprised of, for example, two contiguous physical pages 290.
[0054] An alternative arrangement, in which there may be a one-to-one o correspondence between each logical page 315 and a physical page 290 of flash device 275, is shown in Figure 7. A difference between this arrangement and the one shown in Figure 6 is that all of the meta-data information 320 may be stored in a single spare area 300 of the first physical page 290 of the transaction 310. Arrangements of this type may be particularly suitable when large capacity flash devices are employed. However, the meta-data information t 5 320 also may be divided between the spare areas 300 of the two contiguous physical pages 290 of the transaction record.
[0055] The sequence identifiers for the transaction records 310 stored in the same device block 290 may have the same values. In such instances, the sequence identifier provides chronological information that may be used to compare the time relationship 2o between the transaction records of different device blocks. Chronological information on the transaction records 310 stored in the same block can be derived from the offset location of the transaction record 310 within the block 290, with later occurring transaction records 310 occurring at larger offsets.
[0056] After the computer system 10 has been started or powered on, the integrity of 25 the file system may be verified by generating a reconstructed version of the file system in random access memory 35. The reconstructed file system, shown generally at 330 of Figure 1, may be generated using the valid, committed transactions stored in the transaction file 220 and from the file/directory information stored in metafile 120. In Figure 1, the reconstructed file system 330 includes a regenerated file hierarchy 335 and an extents table 340.
30 [0057] One manner of generating the extents table 340 is shown in Figures 8 through 11. Figure 8 illustrates a number of interrelated processing steps that may be used to generate the extents pool 340 while Figures 9 through 11 illustrate the logical organization of various tables and arrays generated and used in these operations.
[0058] Generation of the extents table 340 may commence at step 345 of Figure 8 by scanning the blocks of the transaction file 220 to find all of the transaction records. The blocks may be scanned in sequence from the lowest ordered block to the highest ordered block in which a committed transaction record is found. As transactions are found within the blocks, an array of block records identifying each device block having a transaction record may be generated at step 350.
[0059] As the file system software 47 scans the blocks of the transaction file 220 for o transactions, the file system software may encounter a block that has been erased as a result of transactions that have been retired, or because the blocks have not yet been assigned for use in the file system. The transaction header may be structured so that there are no valid transactions that will have all of the bits of the header set to the erased value, typically a binary "1 ". As the file system software 47 scans the blocks of the transaction file 220, any ~5 transaction in which the header indicates an erased block may be skipped.
This header invariant may be enforced by using a single bit as a flag to indicate the transaction is in use by the file system when it is the inverse of the erase value. Upon finding such an erase signature value in a transaction header, scanning of the remaining pages in the block may be skipped thereby saving the time that would otherwise be used to access the erased pages. The 20 overall system startup time may be correspondingly decreased.
[0060] The organization of an exemplary block array 355 is shown in Figure 9.
Each block array record 360 includes a sequence field 365, a begin transaction field 370 and a number of transactions field 375. The sequence field 365 may be used to store the transaction identifier value for the transaction records stored in the block.
The begin 25 transaction field 370 may be used to store an index to the first transaction in the block and the number of transactions field 375 may be used to store the number of transactions found in the block.
[0061] At step 380 of Figure 8, the file system software 47 populates a transaction list table for each record entry in the block array 355. Figure 9 illustrates one manner in which 3o the transaction list table 385 may be organized. In this example, each record 360 of the block array 355 may point to at least one transaction list record 390 of the transaction list table 385.
More particularly, a transaction list record 390 may be generated for each transaction found in the block represented by a given block array record 360. The value stored in the number of transactions field 375 of the given block array record 360 may correspond to the number of transactions in the given block and designates how many records 390 for the given block will be added to transaction list table 385.
[0062] Each transaction list record 390 of the transaction list table 385 may have the same record length and include the same record fields. The exemplary fields used in records 390 of Figure 9 include a file cluster offset field 395, a device cluster index field 400, a number of clusters field 405 and a file identifier/idx field 410. The file cluster offset field 395 may be used to identify the physical location of the transaction within the block. The to device cluster index field 400 may be used to identify where the data for the transaction begins. The number of clusters field 405 may be used to identify how many clusters of data are present within the transaction. Finally, the file identifier/idx field 410, as will be set forth below, is multipurpose. Initially, however, the value stored in the file identifier/idx field 410 may be used to identify the file to which the transaction applies. The file identifier value t5 stored in field 410 may directly correspond to the file identifier used to reference the record in metafile 120. Upon the completion of step 380, the records 360 of block array 355 will be arranged, for example, in increasing block order, while the records 390 for each block array record 360 will be arranged in increasing page order.
[0063] At step 415, the records 360 of block array 355 are sorted based on the values 2o stored in the sequence fields 365. This operation may be performed to place the records 390 of the transaction list table 385 in chronological order (i.e., the order in which the corresponding transactions are to be applied to the files of the file system).
[0064) A temporary file 440 storing file node information corresponding to the transaction records of the file system may then be generated in RAM 35 using the sorted 25 records of block array 355 and transaction list table 385. To this end, a basic record corresponding to the root directory of the file system may be first added to temporary file 440. The information used to generate the root directory node in temporary file 440 may be obtained from the record corresponding to the root directory file stored in metafile 120.
[0065] A logical representation of one manner of arranging the file node records in 3o temporary file 440 is shown generally at 445 of Figure 10. In this example, each file node record 450 includes a file node field 455 and a start field 460. The contents of the file node field 455 may be used to identify the file node to which various transaction records 390 of the transaction list table 385 may be linked. For the sake of simplicity, the contents of the file node field 455 may have the same format as the file identifiers used to access the corresponding record entries 125 of metafile 120. The contents of the start field 460 may be used to identify the location of the first transaction record 390 in transaction list table 385 that s corresponds to the file identified in the file node field 455. As such, each file node record 450 identifies a file within the file system as well as the location of the first transaction relating to the identified file.
[0066] At step 420, each of the sorted records 360 and 390 of the block array 355 and transaction list table 385 are traversed to determine whether or not the temporary file 440 to includes a file node record 450 corresponding to the file identifier stored in file identifier/idx field 410. If a file node record 450 with the same file identifier as the transaction record 390 is not found in the temporary file 440, a new file node record 450 may be created at step 430.
Once a file node record 450 corresponding to the transaction list record 390 exists in temporary file 440, the transaction list record 390 may be linked into a list of transactions for t5 the file node record 450. In this example, the transaction list record 390 may be linked into the fist of transactions for the file node record 450 at step 435 of Figure 8.
The manner in which a transaction list record 390 may be linked into the list of transactions for the file node may depend on whether the transaction list record 390 is the first transaction list record of the file node or a subsequent transaction list record for the file node. If it is the first transaction 20 list record of the file node, the start field 460 of the file node record 450 may be updated to identify the starting location of this first transaction list record 390. As such, the contents of the start field 460 of the file node record 450 may be used to point to a location in the transaction list table 385 that, in turn, contains extent information for the first transaction applied to the file. The function of the file identifier/idx field 410 changes when the 25 transaction list record 390 may be to be appended to existing transaction list records for the file node (i.e., when it is not the first transaction list record for the file node). More particularly, the value and the function of the field 410 may be changed so that it points to the last transaction record 390 associated with the file node. This is illustrated in Figure 10, where the start field 460 of file node record 450 points to the beginning of transaction list 3o record 390. The file identifier/idx field 410 of record 390, in turn, points to the beginning of transaction list record 465, which contains the information on the location of the second transaction for the file represented by the file node record 450. Similarly, the start field 460 of file node record 470 points to the beginning of transaction list record 475. The file identifier/idx field 410 of transaction list record 475 points to the beginning of transaction list record 480, which contains the information on the location of the second transaction for the tile represented by the tile node record 470.
[0067] Once all of the transaction list records of the transaction list table 385 have been linked in the proper manner with the corresponding file node records, the transaction list records for each file node are traversed at step 485 to remove any transaction list records that reference uncommitted and/or bad file transactions. Removal of such transaction list records may be accomplished in a variety of different manners. For example, the file system o software 47 may check the status field of the last occurring transaction to determine whether or not it was committed. If the transaction has been committed, the corresponding record in the transaction list table 385 may be left undisturbed. If the transaction has not been committed, however, the corresponding record in the transaction list table 385 may be removed or otherwise ignored.
~5 [0068] To expedite this type of transaction commitment checking, the file system software 47 only needs to ensure that the last occurring transaction has been committed.
Commitment checking of all other records may be skipped since only the last occurring transaction is impacted by a power failure, improper system shutdown, or the like. By skipping commitment checking of all other records, the time required for system startup may 2o be substantially reduced.
[0069 Although it is shown as part of a linear sequence, step 485 may be executed as each transaction list record may be processed for incorporation in the corresponding file node. For example, file system software 47 may check the status information included in the header of each transaction record to determine whether the transaction has been committed.
25 This check may occur as each transaction record may be used to populate the corresponding transaction list record. Once the file system software 47 finds a transaction that has not been committed, no further processing of the transaction list table 385 in steps 420 through 485 of Figure 8 is necessary.
[0070] At step 490, entries are generated in extents pool 340 for each of the file 3o nodes. One manner in which this may be accomplished is shown in Figure I1.
In this example, the content of the start field 460 of each file node may be changed so that it now operates as an extents index field 487. The extents index field 487 points to the first location

19 in the extents pool 340 containing information on the location of the transaction data for the first transaction for the file. Each extents record 490 may include a number of clusters field 495, a start cluster field 500, and a next extent field 505. The start cluster field 500 identifies the starting location in device 270 where the first file transaction for the file corresponding to the file node may be stored. The number of clusters field 495 identifies how many contiguous clusters of device 270 are used to store the file transaction. The next extents field 505 identifies the extents index of the next extents record for the file represented by the file node. In this example, extents index 487 points to extents record 510 while the next extents field 505 of extents record 510 points to extents record 515.
to [0071] The data used to populate the records of the extents pool 340 may be derived, at least in part, from the data stored in the transaction list table 385. In the example shown here, the extents pool 340 is a more compact form of the transaction list table 385. To this end, file system software 47 may combine transaction list records having contiguous data into a single extents record entry if the transaction list records are part of the same file node.
~5 Similarly, there is no further need to maintain the block array 355 in RAM
35. Therefore, block array 355 may be discarded from RAM 35.
[0072] The integrity of the transactions in the transaction file 220 may be checked during the execution of the various steps used to generate extents pool 340.
For example, integrity checking of the transaction records may be executed during either steps 350 or 380

20 of Figure 8. Common data checks include CRC and ECC techniques.
[0073] To decrease the startup time of the computer system 10, error checking techniques may be limited to the information included in the header for certain transactions.
As transactions are found during the startup process shown in Figure 8, the file system software 47 may identify whether the transaction impacts file data or metadata, such as 25 directory structure information in metafile 120. This distinction may be based on the file identifier associated with the transaction. Normally, metadata will be represented by file identifiers that are well-known and hard coded into the file system software 37 (e.g., they will identify the metafile 120 as the file that is the subject of the transaction).
Since only the metadata is required to ensure that the files system is in a consistent state after startup, data 3o checking techniques on the data portion of the transaction are only performed when the transaction relates to such metadata. If the transaction does not relate to a change of the metadata, data checking techniques may be initially limited solely to the checking of the header information. In the transaction record format shown in Figure 6, the principal header information that must be verified on system startup may be stored in the first spare area 300 of each transaction record 310. This allows the file system software 47 to skip verification of the header information included in the second spare area of each transaction record 310 thereby further optimizing the startup sequence. As will be explained in further detail below, error checking of the data portion of each transaction may be deferred until the time that the corresponding file is first accessed by the file system software 37 after completion of the startup sequence.
[0074] Any startup verification of the transaction records may be further optimized by o limiting error checking solely to the first transaction header of a series of sequential transactions. During startup scanning of the transaction file 220, when a transaction header is found that indicates that a number of sequential transaction records for the same file follow, verification of the headers of the trailing transactions in the sequence may be skipped once the header for the first transaction record of the sequence has been verified.
Scanning and t 5 verification of header information may then resume with the next block following the last of the trailing transactions.
[0075] The next broad step in generating the reconstructed file system 330 in RAM
35 includes the construction of the regenerated file hierarchy 335. In this example, the regenerated file hierarchy 335 may be comprised of both file and directory node records. An 2o exemplary format for a directory node record is shown generally at 520 of Figure 12 while a corresponding exemplary format for a file node record is shown generally at 525 of Figure 13.
[0076] Directory node record 520 includes a number of different fields that are used by the file system software 37. More particularly, directory node record 520 may include a z5 sibling field 530, a file identifier field 535, a parent identifier field 540, a child field 545 and a directory named field 550. Similarly, file node record of Figure 13 includes a number of different fields that are used by the file system software 37. The file node record fields may include a sibling field 555, a file identifier field 560, an extents index field 565 and a name sum field 570.
30 [0077] Since the data contained in the records of metafile 120 may be used in the construction of the regenerated file hierarchy 335, the manner in which the metafile records are arranged in the metafile 120 will have an impact on the system startup performance. To

21 this end, the records of metafile 120 are arranged in a single metafile as contiguous records having the same length and are all stored in the same storage media. This arrangement enhances the speed with which the file system software 47 may access the metafile data and reduces the amount of processing that may be required for such access.
[0078] One sequence of steps that may be used to populate the fields for each file node record 525 and directory node record 520 of the regenerated file hierarchy 335 is shown in Figure 14. The illustrated sequence may be executed for each record in metafile 120 and may start at step 575. At step 575, a file identifier may be generated based on the offset of the first record entry within the metafile 120. A check of the regenerated file hierarchy 335 1o may be made at step 580 to determine whether a file node record 525 or directory node record 520 corresponding to the file identifier is already present. If a corresponding record 520 or 525 is not present, a new record file may be created in the regenerated file hierarchy 335. The format of the newly created record depends on whether the file identifier corresponds to a file entry or directory entry in metafile 120. The file system software 37 will t5 make this determination and apply the proper record format 520 or 525.
[0079] At step 585, the fields for the newly created record are populated using the attributes for the file/directory that are found in the metafile 120. If the newly created record corresponds to a directory node, the parent identifier field 540 and directory name field 550 are populated using the data in the parent file identifier and short name fields of the 2o corresponding record in metafile 120. If the newly created record corresponds to a file node, the name sum field 570 may be populated using data that is directly stored or derived from the file name data of the corresponding record in metafile 120. The extents index field 565 may be populated using the data found in the extents index field 487 of the corresponding file node record 450 (see Figure 11 ).
25 [0080] If the newly created file corresponds to a directory node, a search through the regenerated file hierarchy 335 may be undertaken at step 590 to determine whether the parent node exists. if the parent node does not exist, a directory record corresponding to the parent node may be added to the regenerated file hierarchy 335.
[0081] At step 595, the newly generated file/directory record may be linked into the 3o tree structure for the parent directory node. If the child field 545 of the newly generated file/directory record indicates that the parent directory has no children, the value of the child field 545 of the parent directory record may be reset to point to the newly generated

22 file/directory record and the sibling field 555 or 530 of the newly generated fileldirectory record may be set to indicate that the newly generated file/directory record does not have any siblings. If the child field 545 of the parent node record indicates that the parent directory node has children, the sibling field 565 or 530 of the newly generated file/directory record may be set to point to the existing child of the parent directory and the child field 545 of the parent directory may be set to point to the newly generated file/directory record. If the newly generated file/directory record corresponds to a directory node, the parent identifier field 540 of the newly generated directory record may be set to point to the parent directory node.
[0082) At step 600, the file system software 47 recursively ascends the parent nodes, t0 beginning with the parent directory of the newly generated file/directory record, and executes a series of processing steps until the root node is reached. At this point, the parent directory node of the newly generated file/directory record may be referred to as the current directory node. In the exemplary process shown in Figure 14, the file system software 47 checks the regenerated file hierarchy 335 to determine whether a directory node record corresponding to the parent node of the current directory exists. This process may be executed at steps 605 and 610. If such a directory record does not exist in the regenerated file hierarchy 335, a new directory record may be generated at step 615. The child field 545 of the newly generated directory record may be then set to point to the current directory node record as the only child of the new directory record. At step 620, the parent identifier field 540 of the current directory node record may be set to point to the newly generated directory record. The sibling field 530 of the current directory node record may be set to indicate that there are no siblings for the current directory node record at step 625.
[0083) If the check executed at steps 605 and 610 indicate that there is a directory record in the regenerated file hierarchy 335 that corresponds to parent node of the current directory, then the current directory node may be linked into the generalized tree structure of the parent directory node at step 630. To this end, the parent identifier field 540 of the current node may be set to point to the location of the parent node record in the regenerated file hierarchy 335. The sibling field 530 of the current directory node may be set to point to the same record as pointed to by the child field 545 of the parent node record. Finally, the 3o child field 545 of the parent directory node may be set to point to the location of the current directory node.

23 [0084] At step 635, the file system software 37 may check to determine whether the recursive directory processing is completed. In this example, the recursive directory processing is completed when the processing a sends to the root node, which has a unique and recognizable file identifier. If the root node has been reached at step 635, processing of the next file record entry in metafile 120 may begin at step 640, which returns control of the processing back to step 575. If the root node has not been reached at step 635, then processing of the next parent node in the ascending file/directory hierarchy may be repeated beginning at step 605.
[0085] Figure 15 is a logical representation of the reconstructed file system 330 and 1o corresponds to the application of the processing steps of Figures 8 and 14 to a file system having the file hierarchy shown in Figure 2. In this exemplary representation, lines 665, 670, 675, and 680 represent pointers that correspond to the content of the parent identifier fields 540 for the directory node records representing directories 105, 100, 80 and 85, respectively.
Lines 645, 650, 660, 655 and 652 represent pointers that correspond to the content of the is child identifier fields 545 for the directory node records representing directories 110, 100, 105, 80 and 85, respectively. Lines 685, 690, 695 and 705 represent pointers that correspond to the content of the sibling identifier fields 530 for the directory node records corresponding directories 100, 105 and 80, respectively. Lines 700, 705, 710 and 715 represent pointers that correspond to the content of the sibling identifier fields 555 for the file node records 2o corresponding to files 90, 55, 60 and 70, respectively.
[0086] One manner of accessing data in the transaction file 220 of persistent storage unit 30 using the reconstructed file system 330 is also illustrated in Figure 15. As shown, the file system software 37 provides a file identifier 730 for the file node record that the software is to access. In this example, the file identifier 730 points to the file node record representing 25 file 55. The file system software 37 then uses the contents of the extents index 565 of the file node record as an index into extents pool 340 to locate the data for the file in the transaction file 220. It will be recognized, however, that the file system software 37 may use the contents of the reconstructed file system 330 in a variety of different manners other than the one illustrated in Figure 15.
30 [0087] As noted above, complete verification of the integrity of a file is not performed during startup so that startup processing may be expedited. Instead, the file system software 47 may defer complete verification of the file until the first time that the file

24 may be accessed. To this end, the file system software 47 may maintain a table indicating whether or not the integrity of each file has been completely verified.
Alternatively, the file system software 47 may use one or more bits of each file node record in the regenerated file hierarchy 335 to indicate whether the integrity of the file has been completely verified. This indicator may be checked by the file system software 47 at least the first time that a file may, be accessed after startup. If the indicator shows that the file has not been completely verified, a complete verification of the file may be executed at that time.
Alternatively, since the headers of the transactions for the file have already been checked, the file system software need only verify the integrity of the data portions of each transaction for the file. The to verification processes may include one or more CRC processes, one or more ECC processes, and the like.
[0088] As shown in Figures S, 6 and 7, a number of different fields in each of the transaction record headers may be dedicated to verifying the integrity of the entire transaction record. Ifthe integrity checks fail and an application using the relevant error-correcting codes ~5 cannot correct the error, then a program error may be reported back to the application or system that made the request to access the file contents.
[0089] While various embodiments of the invention have been described, it will be apparent to those of ordinary skill in the art that many more embodiments and implementations are possible within the scope of the invention. Accordingly, the invention is 2o not to be restricted except in light of the attached claims and their equivalents.

Claims

1. ~A computer system comprising:
a processor;
a persistent data storage device accessible by the processor;
at least one metafile accessible by the processor, the metafile having a plurality of records that correspond to files on the persistent data storage device and to directory containers for the files, where the individual records of the plurality of records are each addressable by a file identifier, the individual records include a parent identifier, and the parent identifier of a given record corresponds to the file identifier of a parent directory record for the given record.

2. ~A computer system as claimed in claim 1 in which each of the plurality of records has a fixed length.

3. ~A computer system as claimed in claim 1 in which each record of the plurality of records further comprises an extended record identifier, where the extended record identifier for a given record corresponds to a file identifier of an extended record in the metafile, and the extended record contains further information pertaining to the file or directory represented by the given record.

4. ~A computer system as claimed in claim 3 in which the extended record further comprises a plurality of ASCII characters that correspond to a long name of a file or directory represented by the extended record.

5. ~A computer system as claimed in claim 3 in which the parent identifier of a given extended record corresponds to the file identifier of a record having an extended record identifier addressing the given record.

6. ~A computer system as claimed in claim 1 in which each of the plurality of record entries further comprises a plurality of ASCII characters that correspond to a short name of a file or directory represented by the record.

7. ~A computer system as claimed in claim 1 in which each of the plurality of record entries further comprises:
a mode field containing information on the type of file corresponding to the record;
an access time field containing information on the time that the file corresponding to the record was last accessed; and a modified time field containing information on the time that the file corresponding to the record was last modified.

8.~A computer system comprising:
a processor;
a persistent data storage device accessible by the processor;
file system software executed by the processor, where the file system software operates to organize the files of the persistent data storage device into a plurality of records contained in at least one metafile, and the file system software organizes the files and directories in the metafile in an inverted hierarchical structure in which records corresponding to child files and child directories of a parent directory point directly to a record in the metafile that corresponds to the parent directory.

9. ~A computer system as claimed in claim 8 in which each of the plurality of records has a fixed length.

10. ~A computer system as claimed in claim 8 in which each record of the plurality of records further comprises:
an extended record identifier, where the extended record identifier for a given record corresponds to a file identifier of an extended record in the metafile, and the extended record contains further information that pertains to the file or directory represented by the given record.

11. A computer system as claimed in claim 10 in which the extended record further comprises a plurality of ASCII characters corresponding to a long name of a file or directory represented by the extended record.

12. A computer system as claimed in claim 10 where the parent identifier of a given extended record corresponds to the file identifier of a record having an extended record identifier addressing the given record.

13. A computer system as claimed in claim 8 where each of the plurality of record entries further comprises a plurality of ASCII characters corresponding to a short name of a file or directory represented by the record.

14. A computer system as claimed in claim 8 where each of the plurality of record entries further comprises:
a mode field containing information on the type of file corresponding to the record;
an access time field containing information on the time that the file corresponding to the record was last accessed; and a modified time field containing information on the time that the file corresponding to the record was last modified.