TWI288870B - Method and system of data pool allocation and management using one or more data storage drives - Google Patents

Abstract

Aspects of the present invention allow the implementation of one or more data pools using portions or sectors of one or more hard disk drives. Aspects of the invention incorporate at least a method and system of implementing the one or more data pools. The method of implementing a data pool using one or more data storage drives may comprise first generating a first partition table and a second partition tale, wherein the first partition table is a mirror image of said second partition table. The method further comprise second generating a first pool information block and a second pool information block, wherein the first pool information block is a mirror image of said second pool information block. The system of implementing a data pool using one or more data storage drives may comprise a memory, a software resident in the memory, and a processor capable of executing the software.

Description

Ϊ288870 can be used to create #库库. For example, use the first hard drive bay. The name of the new driver 11 - a combination of partitions _ into a data cry to argue ϊ: ΐ: a representative embodiment, can collect all the hard disk drive in the real two columns 睹ί Li and / or Provide mirroring or data segmentation space. In a representative one, a plurality of hard disk drives that are connected or connected may be physically included in the storage device. The data storage device can also be accessed by the device' providing storage for any number of data processing or computing devices. The computing device can include one or more electronic computers. This second-night =111 feature provides a shared access by one or more households to the storage. The data storage devices mentioned below can be called fields. Additional storage device (NAS). A database can be accessed by creating one or more ping = zones by one or more users. In a representative embodiment, one: the stock library may include one or more common areas. Each shared area occupies one partition of the database. 1 is a typical system diagram of a network storage device 100 in an embodiment of the present invention. The network storage device 1 (8) provides data storage for one or more data storage devices. As shown in FIG. 1, a switching device is coupled to the network storage device 1 and one or more data processing devices. The switching device can provide a wireless or wired connection. For example, a wireless router utilizes any of the following wireless or wired data communication protocols: 10/100 Ethernet protocol, one billion Ethernet protocol, 8〇2.11 χ 疋, Bluetooth agreement, etc. The one or more data storage devices include a digital camera $, a digital camera, an MP3 player, a PDA, and one or more personal video recorders. As shown in Fig. 1, the personal video recorder can be configured with a hard disk drive or a hard disk drive. In a representative embodiment, the personal video recorder may be a set-top-box having a personal video recorder function, and the personal video recorder mentioned below also includes a set-top box having a video recorder function. As shown in Figure 1, the personal camera is connected to a television or display capable of displaying multimedia content. The use of the network storage device_ provides a centralized storage device for multimedia content received by one or more personal video recorders, and any personal video recorder without a storage device such as a hard disk drive can receive 1288870. In addition, it includes personal video recorders for internal visits and visits. : Any other information can be hard-wired (4) M not more disc personal video recorders can be accessed by this configuration with κGall sigh (10): Butterfly device can access one or more storage devices 100 can be designed more Easy access to additional data to store hard drives. You can use a matching cable and / or connector to connect the hard drive and network storage device. Therefore, the network storage device (10) is an easy-to-size and flexible mechanism that conforms to the growth trend of future data storage. In addition, the network storage device 100 can also implement data mirroring and capitalization.

2 is a block diagram showing the structure of a network storage device 2 in an embodiment of the present invention. The network storage device 200 includes a printed circuit board (NAS pcB) go?, including one or more components. The one or more components are electrically connected to each other by the printed circuit board (pcB) 202. The one or more components include a network storage device chip 2 (M, random access memory, flash memory 212, AC power interface 216, power supply 220, interface module 224, wireless transceiver/antenna The module 228, the one or more hard disk drives 232, and the controller 236. The interface module 224 may include one or more of the following interfaces: IEEE1394, USB, 10/100 Ethernet, and a billion bit Ethernet Net, PCI, SATA, ΑΤΑ, IDE, SCSI, GPIO, etc. The wireless transceiver/day module 228 can be an add-on module or a mini-pci card that can be connected to the printed circuit board 202 of the NAS or can be assembled. On the printed circuit board 2〇2, the one or more hard disk drives 232 include any number of hard disk drives depending on the design of the MS 200. The printed circuit board 202 can be designed to accept an appropriate number of The number of hard disk drives required depends on the type of mirroring or data segmentation (ie RAID) provided by 11 1288870 NAS 200. The present invention also proposes to assign one or more hard disk drives. Or multiple partial shapes A method of a database. For example, to create a database, a portion of a certain disc drive is linked to a portion of another hard drive. Further features of the present invention are to add an additional hard drive to the NAS. Expanding the amount of storage. Another feature of the present invention includes the use of one or more data storage drives of different sizes or velocities to implement data mirroring and data segmentation of one or more databases, i.e., various raid layers. Function execution). In one embodiment, controller 236 provides control for any device (e.g., a hard disk drive) connected to NFOC 204, which may be an IDE controller or a SATA controller. The NASoC 204 can be an integrated circuit chip containing a processor or a central processing unit (CPU) 204. Figure 3 is a block diagram showing the structure of a wafer (NAS〇c) 3〇〇 according to an embodiment of the present invention. The NASoC 300 is an integrated circuit mounted on the aforementioned ms PCB. The NASoC 300 provides one or more features that allow the NAS to function properly. The NASoC 300 includes: a central processing unit (CPU) 304, an on-chip random access memory 308, an Ethernet/MAC controller 312, an encryption accelerator 316, a security/authentication, a key exchange, and a data rights management (匪) circuit. 32〇 and a set of interfaces 324, 328, 332, 336, 340. The interface 324, 328, 332, 336, 340 includes the following types of interfaces, such as a USB device interface 324, a PCI host interface 332, a GPI0/LCD/flash memory medium interface 328, a UI interface 336, and a USB host interface. 340. The NAS wafer 300 can be in communication and/or connection with one or more of the components described in FIG. As shown in Figure 2, the NAS can be connected to a different number of hard disk drives, depending on its data storage capacity and RAID (data mirroring and/or data segmentation) requirements. The base of the NAS 200 can be configured with two, four or more hard disk drives depending on the type of use. For example, the legs can utilize four hard disk drives to implement legs j) 1+0 (both data mirroring and data segmentation), suitable for use in small office or commercial environments. A feature of the present invention provides a hard disk drive that can use different sizes, types or speeds when performing the MID function. In a home user environment, the money can be 12!28887 碟 disc drive, _ the required storage capacity is usually less than &=== demand. Similarly, the NAS __ memory component is also two = __ not _ alone. As the data storage demand increases, the storage frequency of the bead material also increases, and it is necessary to increase the NAS.

The performance, to the installation, /Mr ΛΑ rain, NAS capacity and change the I MS age, can increase the _ bell or _ to be adjusted accordingly. Nahe, 匕 都 都 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在 在The software or firmware will go to a single user == utilize one or more of the one or more hard drives. The user interface can be further configured with one or more _ layers. In an embodiment, the execution of the software causes the http server in the user station = station to display a pre-set user interface. = In the example, the software executed by the processor 204 includes the operating system =

Windows operating system) can identify the configuration rights, so it can be passed, people, browsing = program execution and browsing. When the user completes the initialization of the hall, the document is set to be accessible. It will be generated during the initialization process - use the verification code of the existing program. Saki Microsoft made f':iS ITrf;2000^ caught the file name of the configuration file that was not displayed, and executed the user's data processing device towel to display a user interface. The user can then either provide or configure multiple inputs to initialize or configure the NAS. The input includes the following 1, the name, the management S name, the administrator password, one or more alternate security, the day, the time zone, the network time server, the internet protocol address: the raid indicator, the data Library share name and shared area access password. The second parameter, the RAID indicator, the database shared area name, and the shared area access port are the main parameters of the internal hard disk drive management. In one embodiment, the aforementioned disk drive 13 1288870 manages participation: stored in the flash memory of the NAS, as shown in FIG. The flash memory may be a non-volatile random access memory (NVRAM). 4 is a flow diagram of database allocation and management of one or more hard disk drives within a network storage device in accordance with one embodiment of the present invention. In step 4〇4, the user accesses the corresponding disk management user interface. In one embodiment, the user is required to enter an authorization password into the user interface to generate the disk management user interface. In a representative embodiment, in order to generate a ^- or multi-user interface including the disk management user interface, it is necessary to perform _ tiling. Figure 5 shows a screenshot of the browser, such as the Wind〇ws browser or Netscape, which provides an exemplary disk management interface that allows the user to enter one or more disk library parameters. In step 408, the user determines the number and type of repositories to be created. As shown in Figure 5, the new (four) library name, the magnetic f provided to one or more disk drives: ® 5 shows two drives, named Surya a anasi', each with an available capacity of i3.5Gb. The user interface is in the size of the padding. It also allows the user to select the branch to be right = or cut. In step 412, the user needs to determine the space layer of the one of the drives used to form the library: the eraser 1, and the face. The sharing ‘is based on one or more materials, 4f(10), and the storage space is included in one or the other. When the administrator or multiple users access the database share access permissions. When the code is as shown in Figure 6, the _ household gets the material occupied by each leg = when there is a shared area in the morning, the two rooms are cumulatively applied with respect to the total space provided by the database 14 1288870. In a representative embodiment, the shared area or its shared directory will occupy a portion of the space of the repository, and the space occupied by one or more of the shared areas is not heavy: £. Figure 7 is a screenshot of a browser provided in accordance with one embodiment of the present invention, showing one or more shared areas and their corresponding one or more cookie banks, which also provide one or more shared areas. Create or delete. The foregoing one or more users f are merely examples, and can be healthy, and the actual remuneration of other user interfaces can be applied to the present invention. The present invention also provides a mechanism for storing files for each portion of one or more data storage drives. The mechanism includes dynamic two-times, recovering the data storage drive: the time-invalid is pure. The change includes modifying the size and configuration of the remainder. The material interface of the _ can be versatile, and has a sturdy and sturdy interface; the f-standard system interface is similar to the L-, Unix, and Windows file systems. Including the description system can be a tree-like directory containing files at the time of the actual time, 贝;, 4°^:° poor material including broadcast or directory name, recent-time access capture and write to one or more directories operating. This ship removes = and meta data. The file storage mechanism is in the plane array of the two system elements of the file system data in the two main squad=scale reading = ”. In the case of one generation, the 21 thousand scales are called the middle layer data, shame, ext3fsp=fsm== The prospective mother system 'such as = the above slot system and the middle layer data array or disk partition change. 彳隹 用 吩 吩 吩 或 或 或 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 The underlying data portion can access one or more original magnetic slices for storing data. Each of the initial magnetic disks can be a data storage drive, which is completely used by the underlying negative portion of the slot storage mechanism. In an embodiment, only one or more of the original magnetic sheets can be read and written by the underlying poor portion of the memory storage mechanism. Each of the one or more original magnetic or data storage drives can be used for one or A plurality of shell libraries provide fine storage, and there is no magnetic disk that is used only by a single database. Several databases can share a magnetic disk through the underlying data portion of the file storage mechanism. The underlying data part is characterized by two state types. The first state is temporary 0 input n, there is a miscellaneous (4), including the memorandum of memory, the temporary state may not appear in the original magnetic # In the material storage drive, in the embodiment of the embodiment, the one or more original magnetic sheets will be in the state of multi-side data. When using the magnetic sheet data processing or the poor material calculation When the device is restarted, the temporary state will be destroyed. This situation will only occur when there is a major error. If the data processing or data computing device is shut down or disconnected, the stability of the data state can be further divided into Two types of materials, and meta-data blocks. The meta-data says how the 0-biliary original data is linked together with eight, and 隹* layers of shellfish. Each raw data block should be representative of her neighbors in the middle data array. In the embodiment, there is a value of S in the original data block, which can be used as "standby, block use" can be used to form a layered mirror in the mirror data. The data may include mirror information and segmentation. Although = or multiple floppy disks are not available, it is preferable to have at least some of the metadata to be a valid library for mirroring, which is typically established to enable at least one backup storage of any of the media's intermediate data arrays. In the remaining or eve of the disk. For example, when one or more magnetic sheets are lost, the structure of the 更ί is more advantageous for the effectiveness of the entire array. In the case of the 1st generation i...Bei Shi, the *demagnetized piece indicates that not all the middle layer data array data ^ 1288870 error condition, the bottom part will provide enough diagnostic information to issue the original (four) block of the possible disk _ Butterfly. The division of metadata allows the system to make information available when one or more pieces of magnetic disk are lost. For example, it is important to say that the next day (4) is important. For example, the data is executed with the other size = / sister, or the pair, such as the file name of the poor database. Driven from the storage machine - the -m dorm storage will not produce a backup of the backup, in the case of loss of the situation - in a disk or a material b problem. If all the metadata is easy to understand. The S-storage drive is on, and the problems caused by the power-off will be compared. If the disk containing the metadata is lost, the file storage mechanism will have no data: lit, a disk or block using the file storage mechanism of the present invention in one embodiment is called a disk header. _ φ slice 5 start 』 疋 a fixed size 矣-二# In Figure 8, block _, 812, 816, 820, 824 ^ = t __ loser, the nth partition 824 represents 8G4. The remaining part of the disk is the starting point and size of the ^^ area. In the description of the two headers 804, each partition is given to the disk header 804 or partition_n2 = there will be some space unassigned to create a new partition (create a new database or expand multiple partition tables to determine that many partition table formats are in different systems) Solid, LLWindows and DOS systems. Unlike the existing system, 'the invention can also be used to format the partition table used by the underlying part of the butterfly system. One or more used in the storage system of the existing system. The partition table is located at the starting end of the magnetic disk or the storage drive H. The present invention can also use a dual partition table in the magnetic disk of the data converter. The partition tables are mirrors (four). The dual partition table is used to perform integrity of data in one or more data storage drives when one or more operations are performed. The second side of the operation may be an update related to the database and the disk name. The information of the magnetic head includes the woven material layer _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Example, axis The area contains three components _, delete, two ί, t 别 is the first and second database information block (thorac) 904, and the solid database information block includes 512 bytes. The first database Information Π: Contains poor stock metadata. * The three components, the last component, are any number of bytes (or blocks), as defined by the split table of the disk header. The partition table may also define partitions - one or more blocks, including 512 bit π groups. In other embodiments of the invention, the aforementioned database sense blocks and partition payload blocks may also use unused byte groups. Figure 1A is a structural block diagram of a disk header file in the embodiment of the present invention. The broadcast includes 2560 bytes, which can be composed of the following fields: ·, - - -: 0 bits from the offset At the beginning of the tuple, the "magic arm is used to occupy JM bytes. According to various features of the invention, the second brain contains special information" and the corresponding magnetic ticket is recognized as the bottom part of the magnetic card mechanism. The database-part of the database. The 34-bit meta-information piece that does not contain the above-mentioned information will not be translated. It is expected that from the 34th byte of the offset, the NAS identification word = 〇 8 is used. The 6 兀 group is used for this field 1 〇〇 8. The 6 Bytes are identified. The format is described and the magnetic head lion device is generated. In the specific, the 6-byte identification word represents a network storage device. 歹18 1288870 The third field is from the 40th position of the offset. At the beginning of the tuple, the 41-bit tuple disk name field 1012 is used. The 41-bit tuple disk name field 1〇12 can store a user-readable ASCII or Unicode string as the distinguished name of the disk. In a specific embodiment, the field 1012 may end with 0, and each byte after the first unit is also 〇, so the last byte must always be 〇. The fourth field, starting from the 81st byte of the offset, is used for the zero-filler field 1016 with a zero-bit filler of 3 bytes. Fifth field: Starting from the 84th byte of the offset, the 16-byte identification word is used to identify the disk by the disk identification field 1020. The Μ byte identification word is randomly or semi-randomly generated when the floppy header is created. The sixth block: starting from the offset 1 byte, the bit indicating the bit of the block 1024 is used, and the bit is used to refer to the two partition tables in the slot. Which one is in the startup state. For example, the value "〇" can indicate that the first partition table is in the startup state, and any other value indicates that the second partition table is in the startup state. Seventh Margin: Starting from the 101st byte of the offset, no 411 unused bytes are used for the byte field 1028. The eighth field: starting from the 512th byte of the offset, the first partition table is stored in the first partition field 1032, and the first partition field 1032 includes 1 to 24 bytes. Ninth Intercept · Starting from the 1536th bit of the offset, the second partition table is stored in the second partition block 1036, which also includes 1024 bytes. The second, third, and fifth stops of the head file (such as the NAS identification word, the disk name, and the disk unique identification word) are used to define a unique magnetic piece to avoid any two different magnetic pieces. Confusion. In a representative embodiment, in the case of any segment, only one of the two zone tables is marked by the phrase in the field 1 by using a one-tuple. The ^ bit is activated. When the partition table changes, all new information is written to the partition table that was not started. When the update is made, all the partition tables in the t-storage storage drive are updated. Because the startup distinguishing table has not been modified, when a write operation g] is powered off, it will not appear __ state. In a 19 1288870 type of the present invention, a partition table includes a smart tuple and is divided into each individual σ. Each of the 64 personal ports of the partition table has a second_. When the 16-member tuple of the bribe towel is all 0, the storage unit is regarded as an empty storage unit. If the storage unit is not empty, the size of the poor block of the remaining 8 bits is specified in the front block number. The starting block number and the size are expressed in terms of a data block structure. The starting block number is stored efficiently with the purpose of the disk. The size can specify the partition of the partition: yes = what size. In this embodiment of the New Zealand, the size does not include two database information blocks implied before the payload of the parent is the zone. If the partition table shows that the starting block is block 81 and the size is 13 data blocks, the first database of the partition is block 809 'the second database information block is block 81〇, and the partition payload is block 811 to Block 82j: Since the head file of the disk occupies the first 8 blocks, the minimum effective value of the starting block number of a memory cell in the area table is 1 〇. Each partition corresponds to two database information blocks for determining one or more databases. Each partition has a _database t-block so that when one data block is modified, it can access the other. The two database information blocks (p(6) can respectively correspond to the poor storage library A and the miscellaneous library B. The process of the more bile library information block includes the following methods. When changing all the databases of all the partitions in the database In the information block A, the database stomach block b does not change, so that the database information block b will always be in a stable state when the database information block A changes. Similarly, when the database information block B is updated, All the data banks #讯块8 are not changed, so that the data library b block B is unstable when the data bank # block A is always in a stable state. The aforementioned disk head file block offset, bit The group length and the field size are different, and other implementations may be employed in accordance with various features of the present invention. Less Figure 11 is a block diagram of the structure of a repository information block in accordance with one embodiment of the present invention. In the embodiment, each database information block includes 512 bytes, and may include the following exemplary fields: First field: starting from the offset of the byte, 81 bytes are used On the data 20 1288870 library name block 1104, in one In a representative embodiment, the database name block 1104 is composed of a user-readable SACII or Unicode string for indicating the name of a database corresponding to the database information block. Ending with the 〇 byte, each byte after the first 〇 byte is also 〇, so the last byte of the block in this embodiment must always be 〇. The 81st byte of the offset begins, and the 3 bytes of the zero filler are provided by the first zero filler field 1108. The second field: starting from the 84th byte of the offset 16 bytes are used to define the database identifier in the database identifier field 1112. The ι6 bits

Groups can be generated randomly or semi-randomly when the database is created. Fourth Block: Starting with the offset 1st byte, 6 bytes are used to identify the field (id) 1116 for identifying the network storage device that created the database. This special value corresponds to the special value in the hall identification word block 1008 from the 34th byte shown in Fig. 1A. In this way, a unique database identifier, nas recognition f, and creation time stamp can uniquely identify a database. In essence, the combination of these three identifications is no longer available to any other database. If two different materials rely on the same data reliance, identification, and creation time, it will confuse the partition that forms the database or databases. Any two databases created in the same network storage device should have different timestamps. In two, _ storage devices, _ any two libraries should have different NAS identification = the only database The recognition word is generated randomly, thereby reducing the likelihood of using the phase. In some cases, the actual identification of the leg identification ^ _ _ _ storage device is not the same as - one or more data storage ί people axis. If the data is “returned to her” and then the database is subsequently created or generated by another storage device, then the data, NAS identification word. Since the (10) identifier in the header file should correspond to the network storage device that was built for the first time at == 1288870, the NAS file of the header file will not change when the partition table is updated. In this way, the NAS identification words in the header file can be distinguished from each other by different magnetic disks or poor storage drivers. Similarly, the database information block MNAS identification words also make different databases different from each other. The fifth block: starting from the offset of 106 bytes, the 2 bit group of the zero-bit filler is added to the second zero-filler field 1120. Sixth Block: Starting from the 108th byte of the offset, in the Create Time/Date Stamp Block 1124, 9 bytes are used to record the creation time/date stamp of a database. The recorded time and date in conjunction with the NAS Identification Field 1216 uniquely identifies a _, specific database and distinguishes it from other databases. The first 4 bytes of the 9 bytes are the year, the next byte is the month (1-12), then one byte is the hour (0-23), and then the next one The byte is a minute (〇-59), and the last byte is a second (0-59). In a specific embodiment, the time/date stamp is expressed in accordance with world standard time. 'Seventh Block: Starting from the 117th bit of the offset, the 3 bytes of the zero-bit filler are added to a third zero-filler field 1128. Eighth Block: Starting from the 120th byte of the offset, there are 4 bytes used to indicate the number of data segments or segmented partitions in the database. The 4 byte bits are located in the segmentation block 1132. In a representative embodiment, the values "丨,, indicate, use, segmentation, and value "〇" are invalid values. For a single segment, the data knife #4 is the same as the unsegmented, and there must be at least two In the case of segmentation, it makes sense to segment the data. A ninth field · From the 124th byte of the offset, there are 4 bytes that represent the data in the database or are shot by the I brother. The number of partitions. There are 4 bytes in the mirrored shelf 1136. In a representative embodiment, the value "丨," indicates unused and the value "0" indicates invalid. The tenth field: starting from the 128th byte of the offset, the 4 bytes represent the number of zones in the database used in the spare zone 1140. In a representative embodiment, the value "〇" indicates that no spare area is available. 22 1288870 shows the beginning of the tit ===132 byte, there are 4 bytes in the table U44. The number of spoons in the window. 4 bytes are used for the pane to form a whole _ίί, 鞠卩 and a fine ^ 'Teng created or NST ί from g 19, the total number of panes of the library includes (NST> KNPNSP)), Quantity, the number of data bars or the number of divided panes starting from the address

Refers to the number of mirrors or mirrored panes in the spare area or spare window. The NSP recognizes the number of the mirrored data strip of the younger brother, and so on. The 136-bit tuple of the occupancy offset begins, and the number of bytes is sliced (10). The magnetic field of the byte slice in the pane indicated by the slot 1144 is less than or equal to five or one suffix of the group #, and one pane can be distributed within a slice of Dole == partitions. In a plurality of magnets, how many bits of the heart are mailed. In the specific embodiment, a one-bit slice identification word is included to represent its generation 23=bit slice. In a particular embodiment, the =, special partition view in the pane forms the pane. The byte slice size is the size of the ^/^^^. The _ ΐ some of the listening library letter deletion bit group slice size is the same byte to the ^ 3 & ^,," different from the above described from the (four) 23 1288870 fifteenth stop: from the offset Starting with the 148th byte, the pane partition detail field 1160 occupies a 108-byte pane partition list, and the ^ partition is used as the lower-slot of the database. 1〇8 兀 group is: 81 her tuple silk partition _# name, 位 one byte, f: bit it filler, 6 bytes for · identification word, 16 bytes for The disk-only identification word of the disk containing the specified partition, 4 bytes is the number of designated areas, a total of 81 + 1 bottle 6 + 4 = 108 bytes. If partitioned with the pane, fine The phase_partition is just right after the silk-strip, the mis-partition detail table field points to the first g-cell of the next-database, so the pane partition list field records the dependencies between the partitions. The Longku towel only exists—(four), and the partition is the pane (four) only--one partition, the brain of the job partition will be pointed back to the pane. ^ Sixteenth column: From The 256th bit of the shift, the byte slice partition = the table subsection 1164 occupies a byte of a byte of a 108-bit partition, used to determine which partition is below the pane - a bit slice. If the partition is the next one of the pane, the block refers to the first byte slice of the corresponding pane, so if the pane consists of only one partition, the block The bit will indicate the pane itself. 苐17 Arbitration · Starting from the offset of the 364th byte, the size adjustment block • Bit 1168 occupies 4 bytes, indicating whether the database is currently undergoing resizing operations. In a specific embodiment, the value "〇," indicates that it is in a non-adjusted state, and the value 1 indicates that it is in a processing resizing state, and other values are considered invalid. During the sizing operation, the network storage device stores the corresponding information. When the size f operation is interrupted, the network storage device can use the information to self-recover to the state of the sizing, so that the network storage device is in the network storage device. The information will not be lost or destroyed. The present invention may also retain the target fields in the database resource block after the sizing operation of the corresponding database is performed. In a specific embodiment, the validity flag for the last field of the database information block is not used during the resizing operation. The eighteenth column · From the offset of the 368th byte, the reverse variation field 1Π2 occupies 8 bytes, indicating the degree of reverse operation of the resizing operation. Adjust 24 1288870, when the size of the dump, the size of each partition will increase or decrease (including - partition to shrink to zero, or because the size of a partition is increased from zero to ^), so the data needs to be in the pane Move back and forth. In order to avoid data being overwritten, the order in which materials are moved is very important. The migration or conversion of the data is carried out according to the following steps: · Moving forward through the pane, then reversing the (four) material, and then moving backwards and then moving backwards. If the size of the crane is used as a cap, the reverse change amount is restored to the correct position. Nineteenth stop: starting from the 376th byte of the offset, the positive change amount is 11 = 8 bytes are used to indicate the forward progress of the resizing operation, • the reverse change with the foregoing The quantity barrier corresponds to the number of bytes in the positive direction. Twentyth stop: Starting from the offset of the 384th byte, the pre-adjusted size 1180 occupies 8 bytes to show the size of the partition before the sizing operation (in kilobytes) ). It should be noted that if the corresponding partition is newly added, the block is zero; 曰 the 21st block: starting from the 392th bit of the offset, the adjusted size block occupies 8 A byte that shows the size of the corresponding partition (in kilobytes) after the sizing operation. The block can be zero, indicating that the partition is being removed by a resizing operation. _ The twelfth field · From the offset of the first byte, the fourth zero of the filler 1188 occupies the zero-bit filler of 96 bytes. The '12th place' position. Starting from the offset of the 496th byte, the effective database information block 1192 occupies 16 bytes, indicating that the block is a valid database information block. In a representative embodiment, the field 1192 matches a binary information value as a valid database information block. If this database information block does not contain valid database information block data, it will not be explained as such. For example, if the field 1192 is missing, the file management system will ignore the database information block. The offset, byte length, and field size in the aforementioned database information block fields can be changed in different embodiments. 12 is an operational flow diagram of a launcher that allows a NAS to implement one or more repositories using one or more 25 1288870 poor storage drives, in accordance with one embodiment of the present invention. Of course, in a representative embodiment, the one or more data storage drives can be one or more hard disk drives. As shown in FIG. 12, the network storage device links or concatenates one or more partitions in the at least one data storage drive through a startup program to form one or more databases. The network storage device reads a disk header and a database information block of the data storage drive such that the network storage device implements one or more databases. In step 12, the network storage device is booted, identifying and detecting its data storage drive. In step 1208, the firmware solidified in the memory in the network storage device is executed (refer to the description of FIG. 2 above), and the data storage drive is scanned and analyzed. In the launcher, one or more additional data storage drives will be discovered and identified by the network storage device. In step 1212, the software identifies a partition within the data storage drive by reading a disk header associated with the data storage drive. Then, in step 1216, the software further reads the database information block of each partition in the one or more data storage drives. In step 1220, the database information block that is considered invalid is deleted. If necessary, one or more of the deleted repository information blocks can be recovered by utilizing their corresponding (or duplicate) database information blocks. Then in step 1224, the network storage device determines if a partition is lost. For example, if a chunk or pane partition list indicates that a partition cannot be located, the network storage device will determine that the partition has been lost, and proceeds to step 1228. In step 1228, the network storage device will prompt the user to install one or more data storage drives, which may have been previously removed. In step 1236, the user can correct the aforementioned problem by inserting the one or more data storage drives containing the lost partition. Then in step 124, one or more segments (than the welcome) and the pane in the one or more data storage drives are sorted according to a chunk or pane partition list, and then The finishing process ends. If no partition is lost, go to step 1224 to sort one or more segments and boards within the one or more data storage drives to form one or more suitable databases, and then the process ends. 26 1288870 FIG. 13 is a flow chart showing the operation of partition size adjustment of the data storage drive in the data storage device in an embodiment of the present invention. The data storage device can be a network storage device. In a representative embodiment, the resizing operation further includes adding or deleting a repository to the network storage device. In step 13〇4, the user gives the user an input to the network storage device, such as a request, to initiate a resizing operation. The user input can be generated through a user interface as described for the description of FIG. The user can enter a new database name", indicate mirroring or segmentation, and determine the capacity available after adding a new database. In step 13-8, the execution is performed by control of a processor (refer to the description of FIG. 2) The in-vivo firmware in the network storage device. The software will scan and analyze one or more data storage headers provided by one or more data storage devices of the network storage device. Step 1312 The software reads and analyzes the headstock of the one or more data storage drives. The two partition tables within the disk header relate to one or more partitions within the one or more data storage drives. In step 1316, the size adjustment block of at least one information block in each pair of database resources is searched and read. During the scanning process, the software can find one of the database information blocks used for the size adjustment operation. Or multiple blocking positions. As described in the previous _11, the scanning program will look for the size adjustment tune t reverse change t-shirt, positive change amount, pre-break size block and adjust L position In step 1320, 'according to the pre-adjustment size block and the adjusted size bar: week, and converting the size of the data storage drive partition. The reverse change pulse monitoring A small vertical or backward = 4 step.丨 324 'Continue the conversion step until the partition reaches the new size. ^ If 1 point has been checked for deletion, increase the fineness to reach the fixed size, the program is staffed, and the method of the spoon is set to describe the field. The scope of the present invention is not limited to the specific embodiments disclosed above. All of the embodiments of the present invention fall within the scope of the present invention. The main display _ domain 榊 please "touch the method and system of the method and system", (agent case number 15675_) priority, = number is 60/562847, the application date is April 15, 2004, 2 Its complete theme is here. No./, Wang Ben applied for reference and advocated beauty _ when the special secretary please "data deposit 15675US02" to complete its main main link. ^ Do it all and simple description] equipment (10)) typical department II; 3 & Port A> I / iL I.. . [Simple description of the drawing] Fig. 1 is a structural diagram of the system, ▼ /, "Structure block diagram of the network attached storage device of the month embodiment; ^ _ The structure of the chip (10) (6) is a database allocation and management flow of the network storage adder according to an embodiment of the present invention; the internal memory to the hard disk drive is shown in FIG. 13. A schematic diagram of a two-disc machine that touches and Manasi, 4 Surya = is a view of the establishment of the invention - an embodiment of the ___ _ Figure 7 is an embodiment of the invention created or deleted and at least: Screen capture of a co-cognitive browser; associated with a library, 8 is a block diagram of a file storage drive using the present invention in accordance with an embodiment of the present invention; 1 scoop or data = is the present invention - implementation Example of a magnetic storage drive-partition structure FIG. 10 is an embodiment of the present invention FIG. 11 is a structural block diagram of a database information block according to an embodiment of the present invention; FIG. 12 is a diagram of an embodiment of the present invention for allowing a network storage device to utilize at least one data storage. The driver implements a startup program flow chart of at least one database; FIG. 13 is a flow chart showing a partition size adjustment operation of a data storage drive in the data storage device according to an embodiment of the present invention. [Main component symbol description] Network attached memory 1〇〇, 200 One or more components of printed circuit board (NASPCB) 202 NAS chip (NASoC) 204 Random access memory 208 Flash memory 212 AC power interface 216 Power 220 interface block 224 wireless transceiver/antenna module 228 - one or more hard disk drives 232 controller 236 central processing unit (CPU) 240 NAS chip (NASoC) 300 central processing unit (CPU) 304 in-chip random memory Memory 308 Ethernet/MAC Controller 312 Streaming AES Encryption Accelerator 316 A Security/Authentication, Key Exchange, DRM Chip 320 USB Device I/F 324 PCI Host I/F 332 GPIO/LCD/Flash Memory Body Media i/f 328 ΑΤΑ Interface 336 USB Host I/F 340 Header File 804 First Partition 808 Second Part 812 Third Partition 816 29 1288870 Fourth Partition 820 Part N Part 824 Network Attached Memory 900 First PIB 904 Second PIB908 Partition payload (in blocks) 912 "Magic" header file field 1004 NAS identification word field 1〇〇8 Disk name block 1012 Zero bit filler field 1016 Disk identification word block Bit 1020 marked with fields 1024

Unused byte field 1028 First partition field 1032 Second partition block 1036 Database name field 1104 First zero place filler field 1108 Unique data pool identification word block 1112 NAS identification block (Π3 1116 second zero filler field 1120 time/date stamp field 1124 third zero filler block 1128 data block 1132 mirror field 1136 spare area block 1140 pane field 1144 bit Group block 1148 byte slice recognition word block 1152 RAID byte slice field 1156 pane partition schedule field 1160 byte slice partition table field 1164 size adjustment mark 1168 reverse change amount Bit 1172 Forward change amount 1176 Adjust before size block 1180 Adjusted size block 1184 Fourth zero place filler block 1188 Valid database information block 1192

Claims (1)

广1288870 Patent Application Scope A method for realizing database allocation and management by using at least one data storage drive, the method comprising: establishing a first partition table and a second partition table, wherein the first partition table is the second partition a mirroring of the table, wherein the first and/or second partition table defines a location and a size of one or more partitions in the data storage drive, wherein the data storage drive is one of the at least one data storage of the data storage device; And generating a first database information block and a second database information block, wherein the first database information block is a mirror of the second database information block, and the first or second database information block is used For identifying and linking one or more partitions, the first and/or second database information blocks are stored in each of the aforementioned partitions of the at least one data storage drive. 2. The method of claim 1, wherein the method further comprises: identifying at least one or more bit slices and panes in the one or more partitions, and binding the one or more programs Blocks and panes to form the above database. 3 The method of claim 4, wherein the data storage drive comprises a hard disk drive. And 4 methods for adjusting the size of the database partition by using at least one resource storage drive. The method comprises: first generating - the first database information block and the second database information block: the first and second database information blocks Located at the beginning of each partition, the first and second database information blocks are used to identify the above partitions. A method for adding a partition to a database by using at least one data storage drive, 31 τ' 〆" Ο. - U · 爿曰 臧 正) is replacing the page! 288870 The method includes: First, generating a first-database information block And a second database information block, wherein the first and second information information blocks are located at the beginning of the subdivision area to identify the partition; and secondly, generating a storage in the first and second database information blocks First, the first block of the first value; again, generating a second value of the L stored second value indicating the size before the partition is increased; and _ after fetching, generating a third block of the third value, The third value indicates the size of the above-mentioned partition increase. 6 systems for realizing database allocation and management by using at least one data storage drive, the system comprising: a memory; a software resident in the memory; 々-仃a processor of the above software, wherein the processor creates a yth partition table and a second partition table by executing the software, wherein the first partition table is a mirror of the second partition table And the second partition table provides location and size information of at least one of the at least one data memory; the processor further generates a first-lean information block and a second database information block by executing the software. The first database information block is a mirror of the above-mentioned second and second database information blocks, and the first or second database information block is used for identification and chaining - rugged coffee, loading - # / silk: miscellaneous 7. The system of claim 6, wherein the system of claim 6 wherein the first and second data 32 1288870 engine, 13⁄4 correction replaces the I library information block The system of claim 6, wherein the at least one data storage drive is a hard disk drive. 33