Patent Application
20100325384
This application claims priority under 35 USC ยง119 from Korean Patent Application No. 10-2009-0055980, filed on Jun. 23, 2009, in the Korean Intellectual Property Office, the disclosure of which is incorporated herein in its entirety by reference.
1. Field of the General Inventive Concept
The present general inventive concept relates to a data storage device, and more particularly, to a method of efficiently accessing a data storage medium in response to a request from a host, a data storage device having a function to perform the method, and a computer readable recording medium having recorded thereon a program to execute the method.
2. Description of the Related Art
In general, a hard disk drive (HDD) is a type of data storage device and is formed as a single storage device to perform access operations of reading data recorded on a disk (or a platter) or writing user data to the disk according to a host command.
Often, the host command includes logical block address (LBA) information that is unrelated to a current location of a head of a data storage device. Thus, when a disk is accessed according to the host command, a cylinder head sector (CHS) that is a real address of stored data or data to be stored in the data storage device and which is based on the LBA information included in the host command, may not match the current location of the head of the data storage device.
When the host command indicates a write command and when the CHS based on the LBA information does not match the current location of the head of the data storage device, as described above, the data storage device performs additional operations such as a seek operation and a head switching operation so as to move the head to a location of the CHS based on the LBA information, thereby increasing access latency of the data storage device.
The present general inventive concept provides a data storage medium accessing method of accessing a data storage medium of a data storage device according to a virtual address (VA).
The present general inventive concept also provides the data storage device to access the data storage medium according to the VA.
The present general inventive concept also provides a computer readable recording medium having recorded thereon a program to execute the data storage medium accessing method.
Additional aspects and utilities of the present general inventive concept will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the general inventive concept.
Embodiments of the present general inventive concept may also be achieved by providing a data storage medium accessing method including the operations of converting a logical block address (LBA) included in a command received from a host into a virtual address (VA), and converting the VA into a real address of a data storage medium included in a data storage device, and accessing the data storage medium using the real address, wherein the VA is an address that is based on a location of a reserved block of the data storage medium during a data write operation, wherein the reserved block is a valid block that is ready to have data written thereto, and wherein a current location of a head included in the data storage device in the data write operation is set according to the location of the reserved block.
If the command is a write command, the operation of converting the LBA may include the operations of selecting reserved block information from among one or more pieces of reserved block information of the data storage medium, wherein the one or more pieces of reserved block information are in a first queue, moving the head to a reserved block of the data storage medium, wherein the reserved block corresponds to the selected reserved block information, allocating the VA to correspond to the selected reserved block information to the LBA included in the command, wherein when data is written to the reserved block, setting the selected reserved block information as active block information, wherein when the number of the active block information is equal to or greater than a predetermined number, inserting mapping information about mapping between the LBA and the VA into a second queue.
The operation of inserting the mapping information may include the operations of checking whether a VA that exists in the second queue is identical to the VA included in the mapping information before inserting the mapping information, and if the identical VA exists in the second queue, setting block information to which the identical VA is allocated as invalid block information, and transmitting the invalid block information to a third queue.
When a status of one or more pieces of invalid block information of the third queue indicates a full status or when an operation status of the data storage device is in an idle time, the operation of converting the LBA may further include the operations of sorting the one or more pieces of invalid block information in the third queue, and transmitting sorted block information to the first queue.
If the command is a read command, the operation of converting the LBA may further include the operation of searching for a VA to correspond to the LBA by using the mapping information of the second queue.
Embodiments of the present general inventive concept may also be achieved by providing a data storage device including a head disk/data storage medium assembly (HDA), and a controller to convert a logical block address (LBA) included in a command received from a host into a virtual address (VA) based on a reserved block of a data storage medium included in the HDA and accessing the data storage medium using the VA, wherein the reserved block is a valid block that is ready to have data written thereto.
The controller may control an interface between the host and the HDA via three layers, and the three layers may include a first layer to classify the command received from the host into a data command and a non-data command, a second layer to convert an LBA included in a data command received from the first layer into a VA, and a third layer to operate the HDA according to access with respect to the data storage medium based on the VA, and the non-data command.
The controller in the second layer may convert the LBA into the VA by using a first queue to manage one or more pieces of reserved block information of the data storage medium included in the HDA, a second queue to manage at least one piece of mapping information about mapping between the LBA and the VA, and a third queue to manage one or more pieces of invalid block information of the data storage medium, and when a status of the one or more pieces of invalid block information of the third queue indicates a full status or when an operation status of the data storage device is in an idle time, the controller may sort the one or more pieces of invalid block information, and may transmit the sorted block information to the first queue.
If the command is a write command and reserved block information from among the one or more pieces of reserved block information of the first queue is selected in the second layer, the controller may operate the third layer whereby a head included in the HDA is moved to a reserved block that corresponds to the selected reserved block information and that is of the data storage medium, and if the command is a read command, the controller in the second layer may search for a VA to correspond to an LBA included in the command by using the mapping information of the second queue.
According to another aspect of the present general inventive concept, there is provided a computer readable recording medium having recorded thereon a program to execute the data storage medium accessing method described above.
The data storage device of claim 6, wherein the VA is based on a cylinder head sector (CHS) that is the real address of the reserved block
Embodiments of the present general inventive concept may also be achieved by providing a computer readable recording medium to contain computer-readable codes as a program to execute the a data storage medium accessing method, the method including converting a logical block address (LBA) comprised in a command received from a host into a virtual address (VA), and converting the VA into a real address of a data storage medium comprised in a data storage device, and accessing the data storage medium using the real address, wherein the VA is an address that is based on a location of a reserved block of the data storage medium during a data write operation, wherein the reserved block is a valid block that is ready to have data written thereto, and wherein a current location of a head comprised in the data storage device in the data write operation is set according to the location of the reserved block.
Embodiments of the present general inventive concept may also be achieved by receiving a host command and classifying the content of the host command into a data command and a non-data command converting a logical block address (LBA) included in the host command into a virtual address (VA) to correspond to a real address of a reserved block in the data storage medium, and moving a head unit of the HDA to a location of the reserved block of the data storage medium to access the data storage medium with respect to the VA.
The data command may include a write command and the method may further include writing data included in the host command to the reserved block of the data storage medium, and mapping the VA of the written to reserved block to the LBA of the host command.
The data command may include a read command and the method further may further include searching for the VA by using an LBA in the read command, moving the head unit of the HDA to a cylinder head sector (CHS) of the data storage medium corresponding to the searched for VA, and reading data from the correct CHS in the data storage medium.
Embodiments of the present general inventive concept may also be achieved by receiving a host command from a host that includes a data write command, converting a logical block address (LBA) included in the host command into a virtual address (VA) to correspond to reserved block information of the data storage medium, selecting reserved block information among a plurality of pieces of reserved block information managed by a first queue of a controller, selecting a reserved block including a VA in which to write a data instruction of the write command, wherein the VA corresponds to a cylinder head sector (CHS) of the reserved block of the data storage medium, and moving the a head of a head disk assembly to the reserved block to correspond to the selected reserved block information to write data to the reserved block, wherein the reserved block information in the first queue includes a CHS and VA of the reserved block.
The method may also include mapping a VA included in the reserved block to the LBA included in the data write command, setting the selected reserved block information as active block information when data is written to the reserved block of the data storage medium and allocating the mapping information into a second queue of a controller.
The method may also include checking whether a plurality of selected reserved block information of the data storage medium has been written to a predetermined number of active block information and checking whether at least one piece of mapping information in the second queue includes a VA that is the same as a VA included in the reserved block information that is set as the active block information
If the VA in the second queue is not the same VA included in the active block information, inserting the predetermined number pieces of active block information into the second queue.
If the VA in the second queue is the same VA included in the active block information, setting a block information status of a VA pre-existing in the second queue to be invalid block information and transmitting the invalid block information to a third queue
Embodiments of the present general inventive concept may also be achieved by receiving a host command and classifying the content of the host command into a data command and a non-data command, controlling access to the data storage medium using first, second and third layers of a controller, searching for a virtual address (VA) in a first queue of the second layer by using a logical block address (LBA) included in the host command when the host command received in the first layer is a read command, calling the third layer from the second layer to move a head unit to access a cylinder head sector (CHS) of the data storage medium corresponding to the searched VA, and reading data from the CHS data storage medium.
Embodiments of the present general inventive concept may also be achieved by providing a data storage apparatus to perform a data storage medium accessing method, including a head disk/data storage medium assembly (HDA) that includes at least one data storage medium, a controller to receive a data command from a host that includes a write command, the controller including a plurality of layers to manage access to the at least one data storage medium, and a memory to store active block information including virtual addresses (VAs) of data written to reserved blocks of the data storage medium and output the active block information to one of the layers of the controller to represent the VAs of stored data in the data storage medium.
The controller may also include a first layer to receive and classify host commands from a host; and a second layer to convert logical block addresses (LBAs) included in the host commands to VAs that represent real addresses of the data storage medium, to reserve and allocate block information within the data storage medium and to store VAs to correspond to the written to reserved blocks of the data storage medium.
A plurality of memory queues may also be disposed in the controller to store reserved blocks of the data storage medium, to store mapping allocation information for VAs of reserved blocks, and to recycle unused block information to be stored for re-use.
A head unit may also be included to move to a desired write location of a reserve block based on a VA included within a selected reserve block of the data storage medium.
Embodiments of the present general inventive concept may also be achieved by providing an apparatus to access a storage location in a data storage medium including a head disk/data storage medium assembly (HDA) including at least one data storage medium, a controller to receive host commands including logical block addresses (LBAs) from a host and to control the interface between the host and the HDA based on the host commands, the controller including a first queue to select a plurality of reserved blocks in the data storage medium that include reserved block information including virtual addresses (VAs) corresponding to cylinder head sectors (CHSs) of the reserved blocks, a memory to store active block information to correspond to reserved block information of the reserved blocksa second queue to store mapping information of the VAs that correspond to the LBAs of the host commands, and a third queue to recycle invalid reserved block information to the first queue, and a head unit to be moved by the controller to a location of the selected reserved blocks selected from the free queue and to write data from the host command into the reserved blocks.
The controller may form active block information from the selected reserved block information that includes the VAs of the reserved blocks and stores the active block information in the memory.
The above and/or other features and utilities of the present general inventive concept will become apparent and more readily appreciated from the following description of the exemplary embodiments, taken in conjunction with the accompanying drawings, in which:
Exemplary embodiments of the present general inventive concept will be more clearly understood from the following detailed description taken in conjunction with the accompanying drawings in which:
Reference will now be made in detail to the embodiments of the present general inventive concept, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the like elements throughout. The embodiments are described below in order to explain the present general inventive concept by referring to the figures
The controller 110 controls an interface that transmits information including communications and data between a host (not illustrated) and the HDA 120.
The host (not illustrated) may be an apparatus that operates according to Windows-based or other operating system programs. For example, the host may be a computer system operated locally to control the data storage device 100 or may be located remote to the data storage device 100 via a network connection or through the internet. The computer system may include one or a plurality of computer terminals that may access the data storage device 100. The data storage device 100 may use a virtual address (VA) to access one or a plurality of data storage mediums included in the data storage device 100 by using any number of alphanumeric and other characters as address values, regardless of the type of operating system program of the host.
The virtual address may correspond to a physical address of a track, cylinder, sector, cluster, head, or a combination of these locations that identify a desired memory location of stored information or information to be stored within one or more data storage mediums 12 of the HDA 120. The data storage device 100 may also include a memory 130 to store data and operational information of the controller 110 and head disk assembly 120. The memory 130 may be disposed within the controller 110, external to the controller 100, or have portions within and external to the controller 110.
In the case where the data storage device 100 is a hard disk drive (HDD), the HDA 120 may be a head disk assembly. The HDA 120 may have a structure similar to that illustrated in
The HDA 120 may include at least one data storage medium 12 rotated by a spindle motor 14. The data storage medium 12 may be a magnetic disc or a disc. The HDA 120 includes a head 16 located close to a surface of the data storage medium 12. The head 16 may be referred to as a converter.
The head 16 may detect a magnetic field of the data storage medium 12 or may magnetize the data storage medium 12, and thus respectively read data from the data storage medium 12 or write data to the data storage medium 12. In general, the head 16 is coupled to the surface of the data storage medium 12. Although the head 16 is illustrated as a single head in relation to
The head 16 may be integrated into a slider 20. The slider 20 has a structure capable of generating an air bearing or air gap between the head 16 and the surface of the data storage medium 12. The slider 20 is coupled to a head gimbal assembly (HGA) 22 that is attached to an actuator arm 24 having a voice coil 26. The voice coil 26 is located close to a magnetic assembly 28 so as to define a voice coil motor (VCM) 30. A current supplied to the voice coil 26 generates a torque that rotates the actuator arm 24 with respect to a bearing assembly 32. The rotation of the actuator arm 24 allows the head 16 to move while crossing the surface of the data storage medium 12.
In general, data is stored in a plurality of circular tracks 34 of the data storage medium 12. Each of the circular tracks 34 generally includes a plurality of sectors. Each of the sectors includes a data field and an identification field. The identification field includes a gray code to identify a sector and a track (cylinder). The head 16 moves while crossing the surface of the data storage medium 12 so as to read or write information in different tracks and sectors.
As illustrated in
The reserved block is a valid storage block in the data storage medium 12 and is ready to have data written thereto. The virtual address is defined based on a cylinder head sector (CHS) that is a real address of the reserved block. The virtual address may be converted to a real address of the data storage medium 12 included in the data storage device 100, and the data storage medium 12 may be accessed using the real address. Virtual addresses and their corresponding CHS locations may be stored in table form in the memory 130 and accessed by the controller 110. For example, the virtual address may be defined as illustrated in Table 1.
As illustrated in Table 1, reserved blocks of the data storage medium 12 may be adjacent to each other or may be blocks between which some blocks are skipped. The skipped blocks may be bad blocks. The reserved blocks may thus be arranged in an order that allows access despite the presence of bad blocks or blocks being used to store other information.
Thus, when data is written based on the virtual address, the data can be written according to an order of the reserved blocks of the data storage medium 12 to correspond to the virtual addresses. Also, virtual addresses may be defined in such a manner that a number representing a reserved block may be allocated to the CHS according to a page stored in a table. Thus, the virtual address may be referred to as a virtual page number (VPN).
In order to access the data storage medium 12 according to the virtual addresses, the controller 110 may be divided into three layers as illustrated in
That is, as illustrated in
The first layer 111 may classify a command received from the host into a data command and a non-data command, and may be referred to as a HDD interface layer (HIL). The data command may include a read command and a write command. The non-data command may include a setting feature command to change a setting of the HDA 120, a standby command, a reset command, or the like.
The second layer 112 may convert an LBA, which is included in a data command received from the HDD interface layer 111, into a virtual address (VA), and may be referred to as a virtual mapping layer (VML). The virtual mapping layer 112 may be implemented to use three memory queues so as to convert the LBA into the VA. The three memory queues may work in conjunction with the memory 130 to manage address information of data to be written to and read from the at least one data storage mediums 12.
The third layer 113 may convert the VA transmitted from the second layer 112 into the CHS, and thus, access the data storage medium 12 of the HDA 120, and may be referred to as the access and operation layer. The access and operation layer 113 may also operate the HDA according to access of the at least one data storage medium 12 with respect to the CHS based on the VA and the non-data command.
A first queue 301 manages and stores at least one reserved block information (RBI) having different priorities of the data storage medium 12 included in the HDA 120, and may be referred to as a free queue. Each RBI may correspond to a reserved block having a different CHS in the data storage medium 12 or plurality of mediums. A second memory queue 302 manages at least one piece of mapping information regarding mapping between the LBA and the VA, and may be referred to as an allocation or mapping queue. A third queue 303 manages one or more pieces of invalid block information of the data storage medium 12, and may be referred to as a garbage queue. The VAs, RBI, mapping information, and invalid block information may be stored in the memory 130 during the various stages of the write and read processes in the present general inventive concept.
An operation of the virtual mapping layer 112 using the first, second and third queues 301, 302 and 303 will now be described. When the data command that includes a LBA is received from the HDD interface layer 111 and the data command is a write command, reserved block information managed and stored by the free queue 301 is selected. In the case where a plurality of pieces of reserved block information exist in the free queue 301, reserved block information having a first priority from among the plurality of pieces of reserved block information is selected. A reference numeral 304 of
At this time, the VML 112 of the controller 110 may call or signal the access and operation layer 113, and move the head 16 to a reserved block of the data storage medium 12, wherein the reserved block corresponds to the selected reserved block information 304. Accordingly, data may be directly written to the reserved block of the data storage medium 12 so that an access time of a data write operation may be reduced. The reserved block information stored in and selected from the free queue 301 may include a CHS, a VA corresponding to the CHS, and metadata, which exist with respect to corresponding reserved blocks. The metadata may include a write count, a read count, and a retry count which exist with respect to the corresponding reserved blocks.
As illustrated in
When a number of reserved blocks of the data storage medium 12 have been set and stored as pieces of active block information 305 and the number of written to reserved blocks are made to be equal to or greater than a predetermined number N via the aforementioned write process, the controller 110 uses the VML 112 to check whether at least one piece of mapping information that already exists in the second allocation (mapping) queue 302 includes one or more VAs equal to the VAs included in the reserved block information that is set as the active block information 305. The at least one piece of mapping information pre-stored in the allocation queue 302 or in the memory 130 may include at least one LBA, a VA allocated to the LBA, a CHS corresponding to the VA, and meta information allocated to each block. The meta information may include meta information defined in the free queue 301. As a result of the check, if the same VA is not included in the at least one piece of mapping information already existing in the second allocation queue 302, the controller 110 inserts at least N pieces of active block information from the memory 130 that correspond to the written to reserved blocks of the data storage medium 12 into the allocation queue 302. As a result, the allocation queue 302 may include VAs corresponding to the CHSs of the written to reserved blocks.
However, as a result of the check, if it is determined that the at least one VA included in the reserved block information that is set as the active block information is the same VA that is included in the at least one piece of mapping information that already exists in the allocation queue 302, the controller 110 may set block information in which the VA included in the at least one piece of mapping information already existing in the allocation queue 302 to be allocated as invalid block information, and thus transmit the invalid block information to the garbage queue 303 (refer to a process 306 illustrated in
When a status with respect to the one or more pieces of invalid block information of the garbage queue 303 indicates a full status, or when an operation status of the data storage device 100 is in an idle state or idle mode, the garbage queue 303 may sort the one or more pieces of invalid block information according to meta information about the one or more pieces of invalid block information, and transmit the sorted block information to the free queue 301 (refer to a process 307 illustrated in
Also, when the data command received from the HDD interface layer 111 is the read command, the controller 110 using the virtual mapping layer 112 searches for a VA in the allocation queue 302 by using an LBA included in the read command. When the VA in the allocation queue 302 is searched for, the controller 110 using the virtual mapping layer 112 may call the third access and operation layer 113, move the head 16 to a CHS corresponding to the searched VA, and thus, can read data from the data storage medium 12.
As described above, the data storage device 100 according to the current embodiment may write data in an order of valid blocks by separately managing a reserved block and an active block by using the free queue 301 and the allocation queue 302, and thus, may reduce access latency. Also, the data storage device 100 may easily search for a VA stored in the allocation queue 302 during a data read operation. In addition, when the data storage device 100 accesses the data storage medium 12, the data storage device 100 performs wear leveling on a data area of the data storage medium 12 by using the write count information, the read count information, the bad block information, the retry information, or the moving line of a head obtained by considering the most minimum movement of the head 16. By doing so, the data storage device 100 may reduce occurrence of an adjust track write (ATW) or adjust track read (ATR) phenomenon.
The data storage device 100 may include not only the HDD using CHS but also all storage mediums that may have a specific address value at an area of the storage mediums. Accordingly, the VA may be applied to not only the HDD but also to other storage mediums.
As described above, since at least the data storage medium 12 is sequentially accessed according to a virtual page number (VPN) that is the VA that corresponds to reserved block information in the free queue 301 and selected therefrom, a seek operation and/or a head switching operation may be reduced or eliminated when data is written, and thus, an access time may be reduced.
Referring to
Referring to
In operation 505, the controller 110 uses the VML 112 to map a VPN included in the selected reserved block information to an LBA included in the received data command. Accordingly, the LBA included in the received data command is mapped with the VPN included in the selected reserved block information. In operation 506, the controller 110 calls the access and operation layer 113 and writes data to the reserved block of the data storage medium 12.
In operation 507, when the controller 110 determines that the data is written to the reserved block, in operation 508, the controller 110 sets the selected reserved block information as active block information, stores the active block information in memory 130, and determines whether a number of pieces of active block information is equal to or greater than a predetermined number N (operation 509). If the number of the active block information is not equal to or greater than N, the controller 110 returns to operation 501 and then re-performs the aforementioned operations 501 through 509 to accumulate additional active block information.
If the number of the active block information is equal to or greater than N, the controller 110 inserts mapping information about mapping between the LBA and the VPN into the second (or allocation) queue 302 (operation 510).
In operation 511, if the received data command is the read command, the controller 110 searches for a VPN in the mapping information of the second queue 302, wherein the VPN corresponds to an LBA included in the read command, and calls the access and operation layer 113, thereby accessing the data storage medium 12 so as to perform a data read operation according to the searched VPN.
In operation 501, if another command other than the data command is received, the other command may be a non-data command, and thus, the controller 110 calls the access and operation layer 113 directly from the HDD interface layer 111, and controls operations of the HDA 120 according to the non-data command.
Before inserting the active block information into the allocation queue 302, the controller 110 determines whether a pre-stored VPN, which could be identical to a VPN included in the active block information to be inserted, is in the mapping information that already exists in the second queue 302 (operation 610).
If the identical VPN exists, the controller 110 may set block information corresponding to the identical VPN existing in the second queue 302 as invalid block information, and then transmits the invalid block information to the third queue 303 (operation 611). After that, the controller 110 inserts the active block information into the second queue 302 (operation 612). If the identical VPN does not exist, the controller bypasses operation 611 and inserts the active block information into the second queue 302 at 612, similar to operation 510 described above.
If the third queue 303 is in a full status, or an operation of the data storage device 100 is in an idle mode, the controller 110 may sort the invalid block information of the third queue 303 (operation 614). The sorting operation is performed in a similar manner with that described in relation to
Meanwhile, in operation 613, if the status of the third queue 303 is not the full status, or if the data storage device 100 is not in the idle time, the controller 110 returns to operation 601 and re-performs the aforementioned operations.
The present general inventive concept can be implemented as a method, an apparatus, or a system. The present general inventive concept can also be embodied as computer-readable codes on a computer-readable medium. The computer-readable medium can include a computer-readable recording medium and a computer-readable transmission medium. The computer-readable recording medium is any data storage device that can store data as a program which can be thereafter read by a computer system. Examples of the computer-readable recording medium include read-only memory (ROM), random-access memory (RAM), CD-ROMs, magnetic tapes, floppy disks, and optical data storage devices. The computer-readable recording medium can also be distributed over network coupled computer systems so that the computer-readable code is stored and executed in a distributed fashion. The computer-readable transmission medium can transmit carrier waves or signals (e.g., wired or wireless data transmission through the Internet). Also, functional programs, codes, and code segments to accomplish the present general inventive concept can be easily construed by programmers skilled in the art to which the present general inventive concept pertains.
While the present general inventive concept has been particularly illustrated and described with reference to exemplary embodiments thereof, it will be understood that various changes in form and details may be made therein without departing from the spirit and scope of the following claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2009-55980 | Jun 2009 | KR | national |
