disk device

Abstract

A magnetic disk device is provided that reduces the time required for reordering. A controller assigns numerical information of a first scale corresponding to the arrangement order of a plurality of first sectors, the plurality of first sectors being obtained by radially dividing the recording surface of the magnetic disk into units of first storage areas capable of writing a first amount of data using an SMR method. The controller assigns numerical information of a second scale corresponding to the arrangement order of the second sectors, the plurality of second sectors being obtained by radially dividing the recording surface into units of second storage areas capable of writing a first amount of data using a CMR method. The controller sets one or more first sectors and one or more second sectors as accessible. The controller selects one of the first and second scales and converts numerical information of sectors to be accessed by a plurality of commands into numerical information of the selected scale. The controller then executes the plurality of commands in an order based on the converted numerical information.

Description

[0001] This application claims the benefit of priority based on Japanese Patent Application No. 2021-153994 (filing date: September 22, 2021), the entire contents of which are incorporated herein by reference. Technical Field

[0002] This embodiment relates to a magnetic disk device. Background Art

[0003] In recent years, when a disk drive receives multiple commands from a host, it performs a reordering operation to rearrange the execution order of the multiple commands to an order different from the order in which they were received. This reordering operation shortens the total travel distance of the magnetic head, thereby improving access efficiency to the disk. Summary of the Invention

[0004] An object of one embodiment is to provide a magnetic disk device that reduces the time required for reordering work.

[0005] According to one embodiment, a magnetic disk device is connectable to a host. The magnetic disk device includes: a magnetic disk having a recording surface; a magnetic head for accessing the recording surface; and a controller. The controller sets one or more first sectors among a plurality of first sectors as first storage areas for writing data using the SMR (Shingled Magnetic Recording) method, wherein the plurality of first sectors are obtained by radially dividing the recording surface into units of first storage areas capable of writing a first amount of data using the SMR method. The controller assigns numerical information of a first scale corresponding to the radial arrangement order of the plurality of first sectors to each of the plurality of first sectors. The controller sets one or more second sectors among a plurality of second sectors as second storage areas for writing data using the CMR (Conventional Magnetic Recording) method, wherein the plurality of second sectors are obtained by radially dividing the recording surface into units of second storage areas capable of writing the first amount of data using the CMR method. The controller assigns numerical information of a second scale corresponding to the radial arrangement order of the plurality of second segments to each of the plurality of second segments. Upon receiving a plurality of commands from the host computer requesting access to either the first storage area or the second storage area, the controller selects one of the first and second scales and, for each of the plurality of commands, converts the numerical information of the accessed segment into numerical information of a third scale, which is the selected scale. The controller then executes the plurality of commands in the order based on the numerical information of the third scale obtained by converting the numerical information of the accessed segment. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] Figure 1 This is a diagram showing an example of the configuration of the magnetic disk device according to the first embodiment.

[0007] Figure 2 This is a schematic diagram showing an example of the configuration of the buffer memory according to the first embodiment.

[0008] Figure 3 This is a diagram showing an example of the structure of the magnetic disk according to the first embodiment.

[0009] Figure 4 This is a schematic diagram for explaining the SMR method among the recording methods used in the magnetic disk device of the first embodiment.

[0010] Figure 5 This is a schematic diagram for explaining the CMR method among the recording methods used in the magnetic disk device of the first embodiment.

[0011] Figure 6 This is a schematic diagram for explaining positions to which the SMR storage area according to the first embodiment can be allocated.

[0012] Figure 7 This is a schematic diagram for explaining positions to which the CMR storage area according to the first embodiment can be allocated.

[0013] Figure 8 This is a schematic diagram for explaining the difference in arrangement between the SMR segment and the CMR segment according to the first embodiment.

[0014] Figure 9 This is a schematic diagram showing an example of allocation of the SMR storage area and the CMR storage area on the magnetic disk according to the first embodiment.

[0015] Figure 10 This is a flowchart showing an example of the reordering operation according to the first embodiment.

[0016] Figure 11 This is a schematic diagram showing an example of the execution order of a plurality of access commands before the reordering operation according to the first embodiment.

[0017] Figure 12 This is a schematic diagram showing an example of the execution order of a plurality of access commands after the reordering operation according to the first embodiment.

[0018] Figure 13 This is a flowchart showing the writing operation according to the second embodiment.

[0019] Description of labels

[0020] 1 Magnetic disk device, 11 Magnetic disk, 11a Servo area, 11b Track, 12 Spindle motor, 13 Ramp, 15 Actuator arm, 16 VCM, 21 Motor driver, 22 Head, 22r Read head, 22w Write head, 23 HDC, 24 Preamplifier, 25 RWC, 26 Processor, 27 RAM, 28 FROM, 29 Buffer memory, 30 Controller, 40 Host, 50 Recording surface, 60 SMR storage area, 70 CMR storage area, 80 Unused area, 291 Command queue, 292 Data cache. DETAILED DESCRIPTION

[0021] Hereinafter, a magnetic disk device according to an embodiment will be described in detail with reference to the drawings. However, the present invention is not limited to these embodiments.

[0022] (First embodiment)

[0023] As methods for writing data to magnetic disks, two methods are known: SMR (Shingled Magnetic Recording) and CMR (Conventional Magnetic Recording). These methods are collectively referred to as recording methods. Details of SMR and CMR will be described later.

[0024] In the first embodiment, both an area for writing data in the SMR method and an area for writing data in the CMR method can be set on the recording surface of the magnetic disk. A magnetic disk device will be described that can perform a reordering operation in a short time even when both an area for writing data in the SMR method and an area for writing data in the CMR method are set on the recording surface of the magnetic disk.

[0025] In addition, the area where data is written using the SMR method is described as the SMR storage area. The SMR storage area is an example of the first storage area. The area where data is written using the CMR method is described as the CMR storage area. The CMR storage area is an example of the second storage area.

[0026] Hereinafter, the magnetic disk device of the first embodiment will be described in detail. The magnetic disk device of the first embodiment will be referred to as a magnetic disk device 1 .

[0027] Figure 1 This is a diagram showing an example of the configuration of the magnetic disk device 1 according to the first embodiment.

[0028] The magnetic disk device 1 is connected to a host computer 40. The magnetic disk device 1 can receive access commands from the host computer 40. The access commands include a write command for requesting to write data and a read command for requesting to read data.

[0029] The magnetic disk device 1 includes one or more magnetic disks 11. Figure 1 In the figure, for ease of understanding, the magnetic disk device 1 is assumed to include a magnetic disk 11. The magnetic disk device 1 accesses the magnetic disk 11 according to an access command. Accessing the magnetic disk 11 includes writing data to the magnetic disk 11 and reading data from the magnetic disk 11.

[0030] The magnetic disk 11 is accessed via the magnetic head 22. Specifically, the magnetic disk device 1 includes, in addition to the magnetic disk 11, a spindle motor 12, a ramp 13, an actuator arm 15, a voice coil motor (VCM) 16, a motor driver 21, a magnetic head 22, a hard disk controller (HDC) 23, a preamplifier 24, a read / write channel (RWC) 25, a processor 26, a RAM (Random Access Memory) 27, a FROM (Flash Read Only Memory) 28, and a buffer memory 29.

[0031] The magnetic disk 11 is rotated at a predetermined rotation speed around a rotation axis by a spindle motor 12 . The spindle motor 12 is driven by a motor driver 21 .

[0032] The magnetic head 22 accesses the magnetic disk 11 using its write head 22w and read head 22r. The write head 22w writes to the magnetic disk 11. The read head 22r reads from the magnetic disk 11. The magnetic head 22 is mounted on the front end of the actuator arm 15. The magnetic head 22 moves radially along the magnetic disk 11 via the VCM 16 driven by the motor driver 21. When the rotation of the magnetic disk 11 stops, for example, the magnetic head 22 moves onto the ramp 13.

[0033] When reading data from the magnetic disk 11 , the preamplifier 24 amplifies the signal read from the magnetic disk 11 , outputs the amplified signal, and supplies it to the RWC 25 . Furthermore, the preamplifier 24 amplifies the signal corresponding to the write target data supplied from the RWC 25 , and supplies the amplified signal to the magnetic head 22 .

[0034] The HDC 23 controls data transmission and reception with the host computer 40 via the I / F bus, controls the buffer memory 29 , and performs error correction processing on read data.

[0035] Various information is temporarily stored in the buffer memory 29 .

[0036] Figure 22 is a schematic diagram showing an example of the configuration of the buffer memory 29 according to the first embodiment. As shown in this figure, the buffer memory 29 is provided with a command queue 291 and a data buffer 292 .

[0037] The data buffer 292 is a buffer area that temporarily stores data transferred between the magnetic disk device 1 and the host 40 .

[0038] The command queue 291 is a queue-structured area that stores access commands received from the host 40. One or more access commands stored in the command queue 291 are executed sequentially. Furthermore, the command queue 291 may be located in a location other than the buffer memory 29. For example, the command queue 291 may be located in the RAM 27 described later.

[0039] The buffer memory 29 is composed of, for example, a volatile memory capable of high-speed operation. The type of memory constituting the buffer memory 29 is not limited to a specific type. The buffer memory 29 can be composed of, for example, DRAM (Dynamic Random Access Memory), SRAM (Static Random Access Memory), or a combination thereof.

[0040] Return the description to Figure 1 The RWC 25 demodulates the write target data supplied from the HDC 23 and supplies the demodulated data to the preamplifier 24. The RWC 25 also demodulates the signal read from the magnetic disk 11 and supplied from the preamplifier 24 and outputs the demodulated signal to the HDC 23 as digital data.

[0041] The processor 26 is, for example, a CPU (Central Processing Unit) and is connected to a RAM 27 , a FROM (Flash Read Only Memory) 28 , and a buffer memory 29 .

[0042] The RAM 27 is composed of, for example, a DRAM, an SRAM, or a combination thereof. The RAM 27 is used as a working memory by the processor 26. The RAM 27 is used as an area for loading firmware programs or an area for storing various management data.

[0043] FROM 28 is a nonvolatile memory. Processor 26 controls the entire magnetic disk device 1 according to the firmware program pre-stored in FROM 28 and magnetic disk 11. For example, processor 26 loads the firmware program pre-stored in FROM 28 and magnetic disk 11 into RAM 27 and controls the motor driver 21, preamplifier 24, RWC 25, HDC 23, and other components according to the loaded firmware program.

[0044] The configuration including the HDC 23, the RWC 25, and the processor 26 may also be regarded as the controller 30. The controller 30 may also include other elements such as the RAM 27, the FROM 28, or the buffer memory 29.

[0045] Figure 3 This is a diagram showing an example of the structure of the magnetic disk 11 of the first embodiment. Servo information is written on the recording surface 50 formed on the surface of the magnetic disk 11 by a servo writer or the like before shipment. The servo information includes sector / cylinder information and a burst pattern. The sector / cylinder information provides servo addresses in the circumferential and radial directions of the magnetic disk 11. The sector / cylinder information is used for seek control to move the magnetic head 22 to the target track. The burst pattern is used for tracking control to maintain the magnetic head 22 on the target track. In addition, the servo information can also be written to the magnetic disk 11 after shipment by self-servo writing (SSW). Figure 3 Schematic diagram of the servo area 11 a shown in FIG. 1 shows radially arranged servo areas 11 a as an example of the arrangement of the servo areas in which servo information is written.

[0046] A plurality of concentric tracks 11b of different radii are arranged at predetermined intervals (also referred to as track pitch) on the recording surface 50 of the magnetic disk 11. A large number of sectors are continuously formed on each track 11b. The magnetic head 22 writes and reads data to and from each sector.

[0047] The arrangement of the plurality of tracks 11b varies depending on the recording method. Figure 4 and Figure 5 The SMR method and the CMR method are explained.

[0048] Figure 4This is a schematic diagram illustrating the SMR method, a recording method used in the magnetic disk device 1 of the first embodiment. In the SMR method, when data (recorded as first data) is written to a certain track and then data (recorded as second data) is written to a track radially adjacent to the track, the tracks are arranged so that the second data partially overlaps with the first data. That is, according to the SMR method, data on one of two radially adjacent tracks on the magnetic disk 11 is written so that data on the other track partially overlaps.

[0049] For example, data on track #2 is written so as to overlap with a portion of data already written on track #1. Furthermore, data on track #3 is written so as to overlap with a portion of data already written on track #2. In other words, the SMR method repeatedly overlaps data on one track with data already written on an adjacent track.

[0050] As a result, the width of each track, in other words, the track pitch TP, becomes narrower than the width (WHw) of the write head 22w of the magnetic head 22, thereby improving the recording density.

[0051] However, with the SMR method, the track pitch TP is narrower than the width WHw of the write head 22w. Therefore, when updating a portion of the data on multiple tracks, the data on the tracks adjacent to the updated data may be corrupted. To prevent data corruption, with the SMR method, the data on multiple tracks, including the portion of data, is updated collectively. The area of ​​multiple tracks updated collectively is called a band. The amount of data that can be written to a band, or in other words, the size of a band, is common to all bands.

[0052] Figure 5 This is a schematic diagram for explaining the CMR method among the recording methods used in the magnetic disk device 1 of the first embodiment. As shown in this figure, according to the CMR method, each track is configured so as not to overlap with the radially adjacent tracks. In other words, the CMR method is a method in which data of two tracks adjacent to each other in the radial direction of the magnetic disk 11 are written in a manner that does not overlap with each other. According to the CMR method, the width of each track is the same as the width (WHw) of the write head 22w, so data at any position can be updated. Therefore, according to the CMR method, compared with the SMR method, the recording density is low, but on the other hand, high random access performance can be obtained.

[0053] The controller 30 can dynamically change the setting of the area where data is written in the SMR manner, that is, the SMR storage area. The controller 30 sets the SMR storage area for the recording surface 50, or releases the set SMR storage area. Releasing the set SMR storage area refers to a state in which the area set as the SMR storage area can be set again as the SMR storage area (or the CMR storage area described later). In the future, setting the area of ​​the recording surface 50 as the SMR storage area (or CMR storage area) is recorded as allocating the SMR storage area (or CMR storage area) to the area of ​​the recording surface 50. Releasing the set SMR storage area (or CMR storage area) is recorded as releasing the allocation of the SMR storage area (or CMR storage area).

[0054] Figure 6 It is a schematic diagram for illustrating the position where the SMR storage area can be allocated involved in the first embodiment. According to the example shown in this figure, the recording surface 50 is divided into multiple zones in the radial direction. Here, as an example, the recording surface 50 is divided into K partitions (partition #0, partition #1, ..., partition #K-1). A partition is the smallest unit in which the recording density can be set individually. That is, within a partition, the recording density is uniform, but the recording density may be different between two different partitions. In addition, the recording density is determined by a combination of the setting of the track pitch and the setting of the bit pitch.

[0055] Each partition is further divided into multiple segments. Each segment is described as an SMR segment. Each SMR segment has a capacity that can write data of the size of the tape when data is written using the SMR method.

[0056] Furthermore, the capacity written to each track 11b decreases toward the inner periphery of the recording surface 50. Therefore, the number of tracks per SMR zone increases, and the radial width of the SMR zone increases, as the recording surface 50 moves toward the inner periphery. Furthermore, the recording density is set uniformly within the zone, so the width of each SMR zone is set uniformly.

[0057] The controller 30 assigns a number to each SMR segment corresponding to the order of arrangement of all SMR segments in the radial direction. This number is recorded as the SMR segment number. Figure 6 In the example shown, the outermost SMR segment is assigned #0 as the SMR segment number, and each SMR segment is assigned an integer number corresponding to the amount of deviation from the SMR segment assigned #0.

[0058] The method of assigning SMR segment numbers is not limited to the above. For example, the innermost SMR segment may be assigned #0 as the SMR segment number, and each SMR segment may be assigned an SMR segment number with an integer value corresponding to the amount of deviation from the SMR segment assigned #0.

[0059] The SMR segment number can be considered as numerical information indicating the radial position using a scale whose unit is the width of the SMR segment. The SMR segment number, which is numerical information, is recorded as numerical information of the SMR scale. The numerical information of the SMR scale is an example of numerical information of the first scale. Hereinafter, an SMR segment assigned #X (where X is an integer greater than or equal to 0) as the SMR segment number may be recorded as SMR segment #X.

[0060] The controller 30 allocates or releases the SMR storage area in units of SMR segments, thereby aligning the boundaries of the SMR storage area with the boundaries of the SMR segments.

[0061] Furthermore, the controller 30 can dynamically change the settings of the area where data is written using the CMR method, that is, the CMR storage area. The controller 30 allocates or de-allocates the CMR storage area to the recording surface 50. Similar to the SMR storage area, the controller 30 can allocate or de-allocate the CMR storage area using storage areas of the same size as the SMR band.

[0062] Figure 7 This is a schematic diagram for explaining positions to which the CMR storage area according to the first embodiment can be allocated.

[0063] according to Figure 7 In the example shown, each partition is divided into multiple segments. Each segment is described as a CMR segment. When data is written in the CMR format, each CMR segment has a capacity that can write data of the same size as the size of a tape in the SMR format.

[0064] Regarding the CMR zone, the width of the CMR zone in the radial direction increases as it moves toward the inner circumference of the recording surface 50. The recording density is set uniformly within the partition, and therefore the width of each CMR zone is set uniformly.

[0065] The controller 30 assigns a number to each CMR segment corresponding to the order of arrangement of all CMR segments in the radial direction. This number is recorded as a CMR segment number. Figure 7In the example shown, the outermost CMR segment is assigned #0 as the CMR segment number. Furthermore, each CMR segment is assigned a CMR segment number corresponding to the amount of offset from the CMR segment assigned #0. Hereinafter, a CMR segment assigned #X (where X is an integer greater than or equal to 0) as the SMR segment number may also be referred to as CMR segment #X.

[0066] As long as the CMR segment #0 and SMR segment #0, which serve as the segment number references, are located at the same radial position, the method of assigning CMR segment numbers is not limited to the above. For example, if the innermost SMR segment is assigned #0 as the SMR segment number, the innermost CMR segment is assigned #0 as the CMR segment number.

[0067] The CMR segment number can be considered as numerical information indicating the radial position using a scale whose unit is the width of the CMR segment. The CMR segment number as numerical information is recorded as numerical information of the CMR scale. The numerical information of the CMR scale is an example of numerical information of the second scale.

[0068] The controller 30 allocates or releases the CMR storage area in units of CMR segments, thereby aligning the boundaries of the CMR storage area with the boundaries of the CMR segments.

[0069] As already described, the SMR method allows for a smaller track pitch than the CMR method. Consequently, when comparing SMR zones and CMR zones within the same partition, the width of the SMR zones is smaller than that of the CMR zones. Furthermore, within the same partition, the number of SMR zones is greater than that of CMR zones.

[0070] Figure 8 This diagram illustrates the differences between the configurations of SMR segments and CMR segments in the first embodiment. The diagram shows the configurations of SMR segments and CMR segments in a partition #i (where i is an integer from 0 to K-1) among K partitions.

[0071] according to Figure 8 In the example shown, nine SMR extents #n through #n+8 are arranged in partition #i in this order. Furthermore, seven CMR extents #m through #m+6 are arranged in this order in partition #i. As can be seen from this figure, the extent boundaries differ between the SMR and CMR extent arrangements.

[0072] The controller 30 allocates SMR storage areas and CMR storage areas in a mutually exclusive manner. That is, the controller 30 cannot allocate a CMR segment that only partially overlaps with an area already allocated as an SMR storage area as a CMR storage area. The controller 30 cannot allocate an SMR segment that only partially overlaps with an area already allocated as a CMR storage area as an SMR storage area.

[0073] As mentioned above, the boundaries between the SMR and CMR segments differ. This can create gaps between the SMR and CMR storage areas. These gaps are referred to as "idle areas" ("idle areas 80" described later). In other words, "idle areas" are areas that are not allocated as either SMR or CMR storage areas. These areas are not used for storage.

[0074] Figure 9 Schematic diagram showing an example of allocation of the SMR storage area and the CMR storage area to the magnetic disk 11 according to the first embodiment.

[0075] As mentioned above, the SMR storage area and the CMR storage area may be dynamically allocated. Therefore, the SMR storage area 60 and the CMR storage area 70 may be mixed on the recording surface 50 of the magnetic disk 11. Figure 9 In the illustrated example, the SMR storage area 60 a , the CMR storage area 70 a , and the SMR storage area 60 a are allocated in this order from the outer peripheral side.

[0076] Furthermore, the space between the SMR storage area 60 a and the CMR storage area 70 a and the space between the CMR storage area 70 a and the SMR storage area 60 b are each treated as an idle area 80 not used for storage.

[0077] The SMR storage areas 60a and 60b are areas where one or more SMR segments with consecutive SMR segment numbers are collectively allocated as SMR storage areas. The CMR storage area 70a is an area where one or more CMR segments with consecutive CMR segment numbers are collectively allocated as CMR storage areas.

[0078] Figure 10 This is a flowchart showing an example of the reordering operation according to the first embodiment.

[0079] When receiving a new access command from the host 40 ( S101 ), the controller 30 stores the access command in the command queue 291 ( S102 ).

[0080] Then, the controller 30 determines whether the magnetic head 22 is currently accessing the SMR zone allocated as the SMR storage area ( S103 ).

[0081] At the timing of the processing of S103, when the head 22 is accessing the SMR segment allocated to the SMR storage area (S103: Yes), the controller 30 converts the segment number of the access destination segment into numerical information of the SMR scale for the unexecuted access commands in the command queue 291 (S104).

[0082] For example, the amount of data that can be written to a unit area of ​​the magnetic disk 11 using the SMR method is expressed as Vsmr, and the amount of data that can be written to the same unit area using the CMR method is expressed as Vcmr. Furthermore, the unit area can be a partition or the entire recording surface 50 of the magnetic disk 11. The ratio R1 of the width of the SMR zone to the width of the CMR zone is given by the following equation (1).

[0083] R1=Vcmr / Vsmr···(1)

[0084] For example, according to Figure 8 In the example shown, for partition #i, 9 bands of data can be written in the SMR method, and 7 bands of data can be written in the CMR method. Therefore, R1 is calculated as 7 / 9.

[0085] Furthermore, the recording density may vary between partitions. However, the ratio of the amount of data that can be written using SMR to the amount that can be written using CMR is the same regardless of the partition. That is, R1 is a common value in any partition.

[0086] When the access destination of a certain access command is a CMR segment allocated to the CMR storage area and the segment number of the CMR segment is Ncmr, the controller 30 converts Ncmr into numerical information Nsmr of the SMR scale using the following equation (2).

[0087] Nsmr=Ncmr / R1···(2)

[0088] Furthermore, the SMR segment number corresponds to the numerical information of the SMR scale. Therefore, when the access destination of a certain access command is an SMR segment allocated to the SMR storage area, the controller 30 does not convert the segment number of the SMR segment.

[0089] Through the above conversion, the positions of the segments of the access destinations involved in multiple access commands can be evaluated using numerical information on a unified scale. In this case, the numerical information on the SMR scale can be used for evaluation. Thus, even when the access destinations involved in unexecuted access commands include CMR segments, the controller 30 can still evaluate the positional relationship between the access destinations, such as the distance between the access destinations.

[0090] At the timing of the processing of S103, when the head 22 is accessing the CMR segment allocated as the CMR storage area instead of the SMR segment allocated as the SMR storage area (S103: No), the controller 30 converts the segment number of the access destination segment into numerical information of the CMR scale with respect to the unexecuted access commands in the command queue 291 (S105).

[0091] Specifically, when an access command involves an SMR segment allocated to an SMR storage area and the segment number of the SMR segment is Nsmr, the controller 30 converts Nsmr into CMR scale numerical information Ncmr using the following formula (3).

[0092] Ncmr=Nsmr*R1···(3)

[0093] Furthermore, the CMR segment number corresponds to the numerical information of the CMR scale. Therefore, when the access destination of a certain access command is a CMR segment allocated to the CMR storage area, the controller 30 does not convert the segment number of the CMR segment.

[0094] Through the above conversion, the positions of the segments of the access destinations involved in each of the multiple access commands can be evaluated using numerical information on a unified scale. In this case, numerical information on the CMR scale can be used for evaluation. Thus, even when the access destinations involved in the unexecuted access commands include SMR segments, the controller 30 can still evaluate the positional relationship between the access destinations, such as the distance between the access destinations.

[0095] After S104 or S105, the controller 30 determines the execution order of the unexecuted access commands in the command queue 291 based on the numerical information of the unified scale (S106). Then, the reordering operation ends.

[0096] In S106 , the controller 30 can determine the execution order of the access commands by an arbitrary method based on the numerical information of the unified scale.

[0097] In one example, the controller 30 selects the access destination closest to the current position of the magnetic head 22 from the access destinations based on a plurality of access commands as the movement destination of the magnetic head 22. Then, the controller 30 selects another access destination close to the selected access destination as the next movement destination of the magnetic head 22. Later, the controller 30 calculates the movement path of the magnetic head 22 for accessing all the access destinations based on the plurality of access commands by repeatedly selecting the access destination close to the selected access destination as the next movement destination. Then, the controller 30 determines the execution order of the plurality of access commands so that the magnetic head 22 moves along the movement path obtained by the calculation.

[0098] In other examples, the controller 30 may perform complex calculations to minimize the total moving distance of the magnetic head 22 .

[0099] Furthermore, as previously mentioned, the width of each segment increases as the partition becomes closer to the inner periphery. Therefore, the controller 30 can also modify the unified scaled numerical information representing the location of each access destination segment based on the partition to which the segment belongs, so that the unified scaled numerical information representing the location of each access destination segment better corresponds to the actual location. Furthermore, the controller 30 can also determine the execution order of multiple access commands based on the modified numerical information. Furthermore, whether or not to modify the unified scaled numerical information representing each access destination segment is arbitrarily determined by the designer.

[0100] The controller 30 executes the unexecuted access commands in the command queue 291 in the determined order until a new access command is received from the host 40 .

[0101] use Figure 11 and Figure 12 A specific example of the reordering operation according to the first embodiment will be described.

[0102] Figure 11 This is a schematic diagram showing an example of the execution order of multiple access commands before the reordering operation involved in the first embodiment. Figure 8 Examples of the configuration of SMR segments and the configuration of CMR segments are shown.

[0103] according to Figure 11 In the example shown, SMR zones #n+2 and #n+3 are each allocated as an SMR storage area, and CMR zones #m and #m+4 are each allocated as a CMR storage area. Furthermore, the area between CMR zone #m and SMR zone #n+2, and the area between SMR zone #n+3 and CMR zone #m+4, respectively, correspond to the idle area 80.

[0104] And, in Figure 11 , the moving path of the magnetic head 22 is shown in FIG. 1 when a plurality of access commands are executed in the order of the received timing. Specifically, Figure 11 Figure 2 shows the execution sequence of the controller 30 before the reordering operation is performed when the controller 30 receives, in the following order, an access command targeting CMR segment #m, an access command targeting CMR segment #m+4, an access command targeting SMR segment #n+2, and an access command targeting SMR segment #n+3. This sequence causes the magnetic head 22 to make multiple radial reciprocating movements, resulting in a very long total travel distance. Therefore, executing these four access commands requires considerable time.

[0105] The controller 30 passes Figure 10 The series of operations shown above changes the execution order of the four access commands. Specifically, the controller 30 determines the new execution order as follows.

[0106] When executing an access command targeting CMR segment #m, the controller 30 determines that the current position of the magnetic head 22 is accessing a CMR segment allocated as a CMR storage area. Consequently, the controller 30 unifies the segment numbers of the target segments to CMR scale numerical information for the access commands targeting CMR segment #m, CMR segment #m+4, SMR segment #n+2, and SMR segment #n+3.

[0107] The processing of the segment numbers for the unified access destination segments will be described in detail. Furthermore, for ease of understanding, the reference is made to SMR segment #n and CMR segment #m, which are located at the same radial position. That is, both n and m are set to 0. Consequently, the access destinations for the four access commands are set to CMR segment #0, CMR segment #4, SMR segment #2, and SMR segment #3.

[0108] Regarding CMR segments #0 and CMR segments #4, which are the access destinations based on the four access commands, the segment numbers are numerical information of the CMR scale and therefore do not require conversion. The controller 30 converts SMR segments #2 and SMR segments #3, which are the access destinations based on the four access commands.

[0109] according to Figure 8In the example shown, R1 is calculated as 7 / 9. Therefore, the controller 30 converts the segment numbers of SMR segment #2 and SMR segment #3 into numerical information on the CMR scale by multiplying the segment numbers by 7 / 9 according to equation (3).

[0110] The controller 30 performs the conversion using the calculation shown in equation (3) on the segment number "2" of SMR segment #2, obtaining 1.56 as the converted numerical information. Furthermore, the controller 30 performs the conversion using the calculation shown in equation (3) on the segment number "3" of SMR segment #3, obtaining 2.33 as the converted numerical information.

[0111] Thus, the controller 30 can obtain the segment numbers of CMR segment #0, CMR segment #4, SMR segment #2, and SMR segment #3 as numerical information on the CMR scale. Specifically, the controller 30 obtains 0 as the numerical information on the CMR scale for CMR segment #0, 4 as the numerical information on the CMR scale for CMR segment #4, 1.56 as the numerical information on the CMR scale for SMR segment #2, and 2.33 as the numerical information on the CMR scale for SMR segment #3.

[0112] The segment numbers that are unified into numerical information of the CMR scale can be considered to represent the positions of each segment. The controller 30 determines the execution order of the four commands based on the segment numbers of the numerical information that are unified into the CMR scale. Specifically, the controller 30 takes the CMR segment #0 that is currently located at position "0" on the CMR scale, where the head 22 is located, as the starting point, and selects the segments with the closest distance of the head 22's single movement from CMR segment #4 that is located at position "4" on the CMR scale, SMR segment #n+2 that can be considered to be located at position "1.56" on the CMR scale, and SMR segment #n+4 that can be considered to be located at position "2.33" on the CMR scale. In this way, the path of the head 22 starting from CMR segment #0 and passing through SMR segment #2 and SMR segment #3 to CMR segment #4 can be derived. The controller 30 changes the execution order of the four access commands so that the head 22 moves along this path. Figure 12 This shows the execution order of the four access commands after the reordering operation is performed.

[0113] In this way, Figure 11 Compared with the path of the magnetic head 22 shown in FIG. 1 , a path with a shorter total moving distance of the magnetic head 22 can be obtained. As a result, the total time required to execute the four access commands is shortened compared to before the reordering operation.

[0114] Here, a technique compared with the first embodiment is described. The technique compared with the first embodiment is described as a comparative example. According to the comparative example, the controller determines the access destination in units of tracks and determines the execution order of multiple access commands based on the determined positions in units of tracks.

[0115] However, the process of determining the access destination in units of tracks requires a relatively long time. Therefore, according to the comparative example, the reordering work requires a lot of time.

[0116] In contrast, according to the first embodiment, the access destination is determined in units of segments, which are storage areas of the size of a band that aggregates multiple tracks, and the execution order of multiple access commands is determined based on the determined positions in units of segments.

[0117] Determining the position in units of sectors is simpler than determining the position in units of tracks. Therefore, according to the first embodiment, the time required for the reordering operation can be shortened compared to the comparative example.

[0118] Furthermore, even when the recording surface 50 of the magnetic disk 11 contains a mixture of SMR storage areas and CMR storage areas, a simple conversion of the segment numbers using the ratio R1 can yield standardized position information related to the segment positions. Consequently, even when the recording surface 50 of the magnetic disk 11 contains a mixture of SMR storage areas and CMR storage areas, reordering can be performed in a short amount of time.

[0119] As described above, according to the first embodiment, the controller 30 sets one or more of the multiple SMR zones, each of which is obtained by dividing the recording surface 50 radially into storage areas capable of storing tape-sized data using the SMR method, as SMR storage areas. Furthermore, the controller 30 sets one or more of the multiple CMR zones, each of which is obtained by dividing the recording surface 50 radially into storage areas capable of storing tape-sized data using the CMR method, as CMR storage areas. The controller 30 assigns numerical SMR scale information corresponding to the radial arrangement order of the multiple SMR zones to each SMR zone. The controller 30 also assigns numerical CMR scale information corresponding to the radial arrangement order of the multiple CMR zones to each CMR zone. Furthermore, when receiving multiple access commands from the host 40 requesting access to either the SMR storage area or the CMR storage area, the controller 30 selects one of the SMR scale and the CMR scale and, for each of the multiple access commands, converts the numerical information of the access destination segment into numerical information of the selected scale. The controller 30 then executes the multiple access commands in the order based on the converted numerical information.

[0120] This can reduce the time required for the reordering work.

[0121] Furthermore, according to the first embodiment, the controller 30 converts the numerical information into the unified scale based on the ratio of the amount of data that can be written to the unit area using the SMR method to the amount of data that can be written to the unit area using the CMR method.

[0122] According to the first embodiment, the controller 30 performs conversion based on the ratio of the amount Vsmr of data writable in the unit area of ​​the magnetic disk 11 using the SMR method to the amount Vcmr of data writable in the same unit area using the CMR method.

[0123] Thus, conversion is achieved through simple calculation.

[0124] In the above description, the controller 30 performs conversion based on the ratio R1 between Vsmr and Vcmr. The controller 30 may perform conversion based on the inverse of R1, that is, the ratio between Vcmr and Vsmr.

[0125] In the above description, the controller 30 selects the scale to be unified in the conversion process for the storage area in the SMR storage area and the CMR storage area that the magnetic head 22 accesses according to the access command being executed at the time of receiving the new access command.

[0126] The timing used as a criterion for selecting the scale to be unified is not limited to the timing of receiving a new access command. For example, the controller 30 may select a scale corresponding to the storage area accessed by the magnetic head 22 based on the last executed access command.

[0127] The method of selecting the scale is not limited to the above, and the controller 30 may always select the SMR scale or the CMR scale.

[0128] In the above description, the controller 30 executes the reordering operation every time a new access command is received. However, the trigger for executing the reordering operation is not limited to the reception of a new access command.

[0129] For example, after the controller 30 has performed a reordering operation on a plurality of access commands, even if a new access command is not received, the controller 30 executes the plurality of access commands in the execution order determined by the reordering operation. Furthermore, after the execution of the plurality of access commands is completed, the controller 30 may also perform a reordering operation on the unexecuted access commands.

[0130] Furthermore, the reordering operation according to the first embodiment can be applied even when SMR storage areas and CMR storage areas are not mixed on the magnetic disk 11. For example, the technology of the embodiment can be applied in either the case where the entire recording surface 50 is used as an SMR storage area or the case where the entire recording surface 50 is used as a CMR storage area.

[0131] In either case, whether the entire recording surface 50 is used as an SMR storage area or the entire recording surface 50 is used as a CMR storage area, the controller 30 operates, for example, as follows. Specifically, the controller 30 assigns a segment number corresponding to the order in which the multiple segments are arranged in the radial direction to each of the multiple segments obtained by dividing the recording surface 50 in the radial direction. When the entire recording surface 50 is used as an SMR storage area, each segment corresponds to a band. When the entire recording surface 50 is used as a CMR storage area, the size of each segment is arbitrary as long as each segment contains multiple tracks 11b. When the controller 30 receives multiple commands from the host 40, it obtains the segment number of the access destination segment for each of the multiple commands and determines the execution order of the multiple commands based on the obtained segment number of each access destination segment.

[0132] This can reduce the time required for the reordering operation compared to the case where the reordering operation is performed by determining the access destination in units of tracks.

[0133] (Second embodiment)

[0134] The controller 30 may also use the free area 80 as a cache memory.

[0135] As previously mentioned, the zones are arranged differently in SMR and CMR zones, resulting in a potential idle area 80 between the SMR and CMR storage areas. Therefore, when the controller 30 attempts to access another zone (the zone designated as the next zone to be accessed) after accessing a zone, if the magnetic head 22 passes through the idle area 80 and the access to the next zone is for writing, the controller 30 temporarily writes the data to be written to the next zone to be accessed into the idle area 80. This eliminates the need to move the magnetic head 22 to the next zone to be accessed, thus shortening the total travel distance of the magnetic head 22.

[0136] Figure 13 2 is a flowchart showing the writing operation involved in the second embodiment. Figure 10 The reordering work shown in this figure determines the execution order when multiple access commands are executed.

[0137] The controller 30 uses the magnetic head 22 to access the access destination segment related to one of the access commands (S201). Figure 13 In the description, this segment is recorded as the segment of the current access object.

[0138] The controller 30 determines whether the subsequent access command in the execution order determined by the reordering operation is a write command and whether there is a free area 80 between the access destination segment of the subsequent access command and the segment currently being accessed ( S202 ).

[0139] When the subsequent access command is a write command and there is an idle area 80 between the segment of the access destination involved in the subsequent access command and the segment of the current access object (S202: Yes), the controller 30 writes the data requested to be written by the subsequent access command to the idle area 80 (S203).

[0140] When the subsequent access command is not a write command, or there is no idle area 80 between the segment of the access destination involved in the subsequent access command and the segment of the current access object (S202: No), the controller 30 executes an access to the segment of the access destination involved in the subsequent access command using the head 22 (S204).

[0141] After S203 or S204, a series of work is completed. Figure 13 In the series of operations shown, the access command processed in S201 is an example of the first access command. In addition, the subsequent access command in the execution order determined by the reordering operation is an example of the second access command.

[0142] Thus, according to the second embodiment, when there is an idle area 80 between an access destination involved in a certain access command and an access destination involved in a subsequent access command in the execution order determined by the reordering work, and the subsequent access command is a write command, the controller 30 writes the data requested to be written by the subsequent access command into the idle area 80.

[0143] This can shorten the total moving distance of the magnetic head 22 and improve the efficiency of executing a plurality of access commands.

[0144] Furthermore, the controller 30 moves the data written to the free area 80 to the segment originally designated as the access destination associated with the write command at a predetermined timing. The designer can arbitrarily set the timing of the move.

[0145] The recording method used when writing to the free area 80 may be fixed to the CMR method or the SMR method. The controller 30 may select the recording method used when writing to the free area 80 by any method.

[0146] Furthermore, not only data requested by a write command but also arbitrary data such as management data can be temporarily written.

[0147] In this way, the free area 80 can be used as a cache memory for temporarily writing data.

[0148] While several embodiments of the present invention have been described above, these embodiments are provided as examples and are not intended to limit the scope of the invention. These new embodiments can be implemented in various other ways and can be omitted, replaced, or modified without departing from the spirit of the invention. These embodiments and / or their variations are included within the scope and spirit of the invention and are included within the invention set forth in the claims and their equivalents.

Claims

1. A magnetic disk device, connectable to a host computer, comprising: a magnetic disk having a recording surface; a magnetic head that accesses the recording surface; and controller, the controller, One or more first sectors among the plurality of first sectors are set as first storage areas for writing data using a shingled magnetic recording method (SMR method), wherein the plurality of first sectors are obtained by dividing the recording surface in a radial direction using first storage areas capable of writing a first amount of data using the SMR method as units, assigning numerical information of a first scale corresponding to the arrangement order of the plurality of first segments in the radial direction to each of the plurality of first segments, One or more of the plurality of second sectors are set as second storage areas for writing data using a conventional magnetic recording method, i.e., a CMR method, wherein the plurality of second sectors are obtained by dividing the recording surface in a radial direction using the second storage area in which the first amount of data can be written using the CMR method as a unit. assigning numerical information of a second scale corresponding to the arrangement order of the plurality of second segments in the radial direction to each of the plurality of second segments, When a plurality of commands are received from the host, each of which is a command requesting access to any one of the first storage area and the second storage area, one of the first scale and the second scale is selected, and for each of the plurality of commands, numerical information assigned to the access destination segment is converted into numerical information of a third scale that is the selected scale, and the plurality of commands are executed in order based on the numerical information of the third scale obtained by converting the numerical information of the access destination segment. The controller converts the numerical information into the third scale based on a ratio of one of an amount of data writable in the third storage area using the SMR method and an amount of data writable in the third storage area using the CMR method to the other.

2. The magnetic disk device according to claim 1, The controller selects, as the third scale, a scale corresponding to a storage area accessed by the magnetic head at a predetermined timing, among the first storage area and the second storage area.

3. The magnetic disk device according to claim 2, The predetermined timing is the timing when a new command is received. The plurality of commands includes the new command.

4. The magnetic disk device according to claim 1, When a fourth storage area exists on the recording surface, the controller uses the fourth storage area as a cache memory, wherein the fourth storage area is an area that is not set as any storage area of ​​the first storage area and the second storage area.

5. The magnetic disk device according to claim 4, The controller writes data requested to be written by the second command into the fourth storage area when the fourth storage area exists between an access destination associated with a first command among the plurality of commands and an access destination associated with a second command following the first command in the order of the numerical information based on the converted third scale, and when the second command is a write command.

6. The magnetic disk device according to claim 1, The first amount is the size of one belt.

Images