Sorted serpentine mapping for storage drives

Description

BACKGROUND

1. Technical Field

This disclosure relates to disk drives and data storage systems for computer systems. More particularly, the disclosure relates to systems and methods for using sorted serpentine mapping for disk drives.

2. Description of the Related Art

Disk drives typically comprise a disk and a head connected to a distal end of an actuator arm which is rotated about a pivot by a voice coil motor (VCM) to position the head radially over the disk. The disk comprises a plurality of radially spaced, concentric tracks for recording user data sectors and embedded servo sectors. The embedded servo sectors comprise head positioning information (e.g., a track address) which is read by the head and processed by a servo controller to control the velocity of the actuator arm as it seeks from track to track.

A multi-head disk drive comprises multiple heads and disk surfaces, where each head is configured to write data to and read data from a respective disk surface. Each disk surface may be formatted into a number of data zones, where each data zone is associated with a particular data transfer rate. When a sequence of data is written to or read from the disk drive, format variations among the disk surfaces in the disk drive may result in undesirable performance drawbacks, such as large fluctuations in data access rates.

BRIEF DESCRIPTION OF THE DRAWINGS

Systems and methods that embody the various features of the invention will now be described with reference to the following drawings, in which:

FIG. 1A illustrates a disk drive according to one embodiment of the invention.

FIG. 1B illustrates the format of a servo sector according to one embodiment of the invention.

FIG. 2A illustrates a platter according to one embodiment of the invention.

FIG. 2B illustrates a disk drive with multiple disk surfaces and heads according to one embodiment of the invention.

FIG. 3 illustrates a combination of a host system and a disk drive according to one embodiment of the invention

FIG. 4A illustrates serpentine mapping according to one embodiment of the invention.

FIG. 4B illustrates sorted data zones mapping according to one embodiment of the invention.

FIG. 4C illustrates sorted serpentine mapping according to one embodiment of the invention.

FIG. 5 further illustrates sorted serpentine mapping of FIG. 4C.

FIG. 6 illustrates a flow diagram of sorted serpentine mapping according to one embodiment of the invention.

FIG. 7 illustrates a flow diagram for performing storage access operations according to one embodiment of the invention.

DETAILED DESCRIPTION OF SPECIFIC EMBODIMENTS

While certain embodiments are described, these embodiments are presented by way of example only, and are not intended to limit the scope of protection. Indeed, the novel methods and systems described herein may be embodied in a variety of other forms. Furthermore, various omissions, substitutions and changes in the form of the methods and systems described herein may be made without departing from the scope of protection.

Overview

A multi-head disk drive may comprise multiple heads and multiple platters, each of which may for example comprise two disk surfaces. Each head may be configured to access a respective disk surface (e.g., read data from and write data to the associated disk surface). Disk surfaces may further comprise a plurality of data tracks, which may be grouped into a plurality of cylinders. Especially in disk drives where adjacent tracks seeks are more efficient than switching from a particular head to a next head, it may be advantageous to map data tracks using serpentine mapping, which determines the sequence by which the physical locations of the disk drive are mapped to logical addresses. When serpentine mapping is used, a particular head can map all tracks in a particular cluster or zone by switching between mapping in forward (e.g., away from the outer diameter of a platter) and reverse directions. After making a pass over all tracks in the cluster, the disk drive can switch to a next head and repeat this process for all tracks associated with the next head. This mapping can be continued until all tracks have been mapped. However, one problem with serpentine mapping is that a first logical address of the disk drive (e.g., LBA 0) may not be mapped to a physical location in the disk drive having the highest data rate. In many applications, such as bootstrapping of an operating system during initialization, benchmarking, etc., it is important to have the first logical address mapped to a disk drive location having the highest data rate.

In addition, each disk surface of the disk drive may also be formatted into a number of data zones, where each data zone is associated with a particular data rate. The format may vary from disk surface to disk surface within a disk drive. This may be due to differences in performance (e.g., data rate) of respective heads and/or other factors. For example, tracks and/or cylinders located near the outer diameter of disk platters tend to have higher data rates than tracks and/or cylinder located near the inner diameter because while the angular velocity remains constant, the linear velocity decreases as heads move closer to the inner diameter (and the radius decreases). Accordingly, tracks and/or cylinders may be grouped into data zones based on their proximity to the outer diameter. A plurality of data zones may be sorted in accordance with data rates, and heads may be configured to map tracks starting from a zone having the highest data rate and continuing with a zone having the next highest data rate. However, one problem with using this approach is that seek times may be longer when compared to using serpentine mapping. Because for a particular zone the disk drive may remain on tracks associated with a high data rate head for a greater number of tracks in relation with lower data rate head(s), the distance traversed during switching to a lower data rate head (e.g., the distance between the last track associated with the high data rate head and first track associated with the lower data rate head) may not be constrained. Accordingly, seek times within a particular zone may be longer than those achieved using serpentine mapping.

In some embodiments of the present invention, a plurality of physical heads of a disk drive is associated with or mapped to a plurality of logical heads. The plurality of logical heads is sorted according to the respective data rates of the physical heads. For example, a physical head with the highest data rate is mapped to the first logical head, while the physical head with the lowest data rate is mapped to the last logical head. When serpentine mapping is utilized, the first logical address of the disk drive (e.g., LBA 0) is mapped to the physical location in the disk drive (e.g., sector) having the highest data rate.

In some embodiments of the present invention, the disk drive is further partitioned or formatted into a plurality of clusters, which are sorted by the respective data rates. For example, clusters that are located closer to the outer diameter of the platters may have higher data rates than clusters located farther from the outer diameter. The clusters can be configured to have a predetermined or fixed number of cylinders such that seek times within a particular cluster are constrained.

System Overview

FIG. 1 illustrates a disk drive 100 according to one embodiment of the invention. Disk drive 100 comprises a disk or platter 102, an actuator arm 4, a physical head 104 actuated radially over the disk 102 and connected to a distal end of the actuator arm 4. The disk 102 comprises a plurality of data tracks 10. In one embodiment, the disk 102 can comprise one or more reserved tracks. The disk drive further comprises a controller 14.

In one embodiment, the disk 102 comprises a plurality of servo sectors 24₀-24_N that define the plurality of data tracks 10. The controller 14 processes the read signal to demodulate the servo sectors 24₀-24_N into a position error signal (PES). The PES is filtered with a suitable compensation filter to generate a control signal 26 applied to a voice coil motor (VCM) 110 which rotates the actuator arm 4 about a pivot in order to position the physical head 104 radially over the disk 102 in a direction that reduces the PES. The servo sectors 24₀-24_N may comprise any suitable position information, and in one embodiment, as is shown in FIG. 1B, each servo sector comprises a preamble 30 for synchronizing gain control and timing recovery, a sync mark 32 for synchronizing to a data field 34 comprising coarse head positioning information such as a track number, and servo bursts 36 which provide fine head positioning information. The coarse head position information is processed to position a head over a target track during a seek operation, and the servo bursts 36 are processed to maintain the head over a centerline of the target track while writing or reading data during a tracking operation.

Because the disk 102 is rotated at a constant angular velocity, the data rate is increases toward the outer diameter tracks as the linear velocity is higher in relation to that of the inner diameter tracks (where radius is smaller). Tracks 10 can be grouped into a number of zones, wherein the data rate is substantially constant across a zone, and is increased from the inner diameter zones to the outer diameter zones. As is illustrated in FIG. 1, data tracks are grouped into three physical zones 120 (Zone 1), 118 (Zone 2), and 116 (Zone 3). Zone 1 120 is located closest to the inner diameter of the disk, and Zone 3 116 is located closest to the outer diameter of the disk. While three zones are illustrated in FIG. 1, those of ordinary skill in the art will appreciate that the disk 102 can comprise any suitable number of zones.

FIG. 2A illustrates the disk or platter 102 according to one embodiment of the invention. The platter 102 comprises a top surface 102A and a bottom surface 102B, each of which comprise data tracks 10. The platter is rotated about a central axis 202. The inner diameter 206 of the platter is located closer to the central axis 202 than the outer diameter 204 of the platter.

FIG. 2B illustrates a disk drive 210 with multiple disk surfaces (102A, 102B, 102C, and 102D) and physical heads (104A, 104B, 104C, and 104D) according to one embodiment of the invention. Each physical head 104_i is attached to a distal end of a respective actuator arm 4_i (4A, 4B, or 4C), and is rotated about a pivot by a voice coil motor (VCM) 110 in order to actuate the physical head 104_i radially over the disk surface 102_i. Each head is configured to access data tracks 10 of the respective disk surface 102_i. Disk surfaces are rotated about central axis 202.

In one embodiment, tracks across disk surfaces 102_i are grouped into a plurality of cylinders, wherein each track of the set of tracks in a particular cylinder is located substantially same distance from the central axis 202. For example, cylinder 220 comprises four tracks 222, 224, 226, and 228 located respectively on disk surfaces 102A, 102B, 102C, and 102D. Each of these tracks is accessed (e.g., written and read) and mapped by a respective physical head 104A, 104B, 104C, and 104D.

FIG. 3 illustrates a combination 300 of a host system and a disk drive according to one embodiment of the invention. As is shown, a disk drive 320 (e.g., a hard disk drive) includes a controller 330 (which can comprise the controller 14) and a magnetic storage module 360. The magnetic storage module 360 comprises magnetic media 364 (e.g., one or more disks 102) and a cache buffer 362 (e.g., DRAM). The cache buffer 362 can comprise volatile storage (e.g., DRAM, SRAM, etc.), non-volatile storage (e.g., non-volatile solid-state memory, such as flash integrated circuits, Chalcogenide RAM (C-RAM), Phase Change Memory (PC-RAM or PRAM), Programmable Metallization Cell RAM (PMC-RAM or PMCm), Ovonic Unified Memory (OUM), Resistance RAM (RRAM), NAND memory, NOR memory, EEPROM, Ferroelectric Memory (FeRAM), Magnetoresistive RAM (MRAM), or other discrete NVM (non-volatile memory) chips), or any combination thereof. In one embodiment, the cache buffer 362 can store system data.

The controller 330 can be configured to receive data and/or storage access commands from a storage interface module 312 (e.g., a device driver) of a host system 310. Storage access commands communicated by the storage interface 312 can include write data and read data commands issued by the host system 310. Read and write commands can specify a logical address (e.g., LBA) used to access the disk drive 320. The controller 330 can execute the received commands in the magnetic storage module 360.

Disk drive 320 can store data communicated by the host system 310. In other words, the disk drive 320 can act as memory storage for the host system 310. To facilitate this function, the controller 330 can implement a logical interface. The logical interface can present to the host system 310 disk drive's memory as a set of logical addresses (e.g., contiguous address) where host data can be stored. Internally, the controller 330 can map logical addresses to various physical locations or addresses in the magnetic media 364.

Examples of Mapping

FIG. 4A illustrates serpentine mapping 400A according to one embodiment of the invention. Serpentine mapping can be performed by the controller 330 and/or controller 14. Data tracks 410 of a plurality of disk surfaces are accessed and mapped by a plurality of physical heads 104A, 104B, 104C, and 104D. Data tracks are grouped into a plurality of clusters 404A, 406A, and 408A. The clusters are sorted according to the respective data rates. For example, the data rate of cluster 404A (which is located closest to the outer diameter) may be higher than the data rate of cluster 406A, which in turn may have higher data rate than that of cluster 408A (which is located closest to the inner diameter). The clusters can have a predetermined or fixed number of cylinders (e.g., 64 or less, 128 or less, 128 or more, etc.). While four physical heads and three clusters are illustrated in FIG. 4A, those of ordinary skill in the art will appreciate that any suitable number of heads and clusters can be utilized.

In one embodiment, the disk drive is mapped beginning with track 402A (located closest to the outer diameter) associated with the first physical head (e.g., physical head 104A). Mapping is continued with subsequent tracks of cluster 404A associated with the physical head 104A. When the last track within the cluster 404A is mapped, mapping is switched to the next physical head (e.g., physical head 104B) and tracks associated with that physical head are mapped in reverse direction. In other words, tracks associated with the physical head 104B are mapped beginning with the track in the same cylinder as the last mapped track associated with the physical head 104A and ending with the track associated with the physical head 104B located closest to the outer diameter.

Mapping is then switched to the next physical head (e.g., physical head 104C) and tracks associated with that physical head are mapped in forward direction. This process continues until all tracks of cluster 404A associated with the physical heads 104A-104D have been mapped. At that point, mapping proceeds with the track associated with the first physical head (e.g., physical head 104A) in the next cluster (e.g., cluster 406A). This process continues until tracks in all clusters have been mapped. As explained above, one problem with serpentine mapping is that the first logical of the disk drive (e.g., LBA 0 corresponding to 402A) may not be mapped to a physical location in the disk drive (e.g., {cylinder, head, sector} tuple) having the highest data rate.

FIG. 4B illustrates sorted data zones mapping 400B according to one embodiment of the invention. Sorted data zones mapping can be performed by the controller 330 and/or controller 14. Data tracks 410 can be formatted or grouped into a number of zones, where each data zone is associated with a particular data rate. In particular, the data rate can be substantially constant across a zone, and can increase from the inner diameter zones to the outer diameter zones. As is illustrated in FIG. 4B, data tracks are grouped to form two zones 412 (Zone 1) and 414 (Zone 2). Zone 412 is located closest to the outer diameter of the disk, and zone 414 is located closest to the outer diameter of the disk. Zones can be sorted according to the respective data rates. For example, because zone 412 is located closer to the outer diameter, it has a higher data rate than zone 414. While two zones are illustrated in FIG. 4B, those of ordinary skill in the art will appreciate that the disk drive can comprise any suitable number of zones.

In one embodiment, the disk drive is mapped beginning with track 402B (located closest to the outer diameter) of the physical head determined to have the highest data rate (e.g., physical head 104B) and continuing with subsequent tracks of zone 412 associated with the physical head 104B. When the last track of zone 412 associated with the physical head 104B is mapped, mapping switches to the physical head having the next highest data rate (e.g., physical head 104D).

Tracks associated with the physical head 104D are mapped beginning with track 402B′ (located closest to the outer diameter) and continuing with subsequent tracks of zone 412 associated with the physical head 104D. When the last track of zone 412 associated with the physical head 104D is mapped, mapping is switched to the physical head having the next highest data rate (e.g., physical head 104A). Tracks associated with the physical head 104A are mapped beginning with track 402B″ (located closest to the outer diameter) and continuing with subsequent tracks of zone 412 associated with the physical head 104A. When the last track of zone 412 associated with the physical head 104A is mapped, mapping is switched to the physical head having the lowest data rate (e.g., physical head 104C). Tracks associated with the physical head 104C are mapped beginning with the track located closest to the outer diameter and continuing with subsequent tracks of zone 412 associated with the physical head 104C. When the last track associated with the physical head 104C is mapped, mapping proceeds with a track (located closest to the outer diameter) associated with the physical head having the highest data rate (e.g., physical head 104B) in the next zone 414. This process continues until tracks in all zones have been mapped.

Using the sorted data zones approach, the first logical address of the disk drive (e.g., LBA 0 corresponding to 402B) is mapped to a physical location in the disk drive having the highest data rate. However, as explained above, one problem with using this approach is that seek times may not be deterministic. Because in a particular zone the disk drive may remain on tracks associated with a high data rate head (e.g., 104B) for a greater number of tracks in relation to physical heads having lower data rates, the distance traversed during switching to a lower data rate head (e.g., the distance between the last track associated with the high data rate head and first track associated with the lower data rate head, such as the distance between the last track in zone 412 associated with the physical head 104B and track 402B′) may not be constrained. Accordingly, seek times within a particular zone may be longer than those achieved using serpentine mapping.

Sorted Serpentine Mapping

FIG. 4C illustrates sorted serpentine mapping 400C according to one embodiment of the invention. Sorted serpentine mapping can be performed by the controller 330 and/or controller 14. Any suitable serpentine mapping pattern may be employed, including cylinder serpentine and surface serpentine mapping patterns. In addition, any suitable non-serpentine mapping pattern may be employed. In one embodiment, physical heads are assigned or mapped to logical heads sorted by the data rate of the associated physical heads. For example, the logical heads can be sorted according to the decreasing data rate of the associated physical heads. Data rates of physical heads can be determined using calibration. The mapping between physical and logical heads can be stored, for example, in the cache buffer 362, with a persistent copy stored in the magnetic media 364. Using sorted serpentine mapping, the first logical address of the disk drive is mapped to a physical location in the disk drive having the highest data rate. In addition, cylinders are grouped into clusters of predetermined or fixed size so that seek distances within a particular cluster are deterministic and constrained.

As is illustrated in FIG. 4C, physical head 104B determined to have the highest data rate is mapped to the first logical head (e.g., logical head 0). Physical head 104D determined to have the second highest data rate is mapped to the second logical head (e.g., logical head 1). Physical head 104A determined to have the next highest data rate is mapped to the third logical head (e.g., logical head 2). Physical head 104C determined to have the lowest data rate is mapped to the last logical head (e.g., logical head 3).

In one embodiment, the disk drive is mapped beginning with track 402C (located closest to the outer diameter) associated with the first logical head (e.g., logical head 0 mapped to the physical head 104B). Because the first logical head is mapped to the physical head having the highest data rate, the first logical address of the disk drive (e.g., LBA 0) is mapped to a physical location with the highest data rate. Cylinders are further grouped into clusters of predetermined or fixed size. Three such clusters 404A, 406A, and 408A are illustrated in FIG. 4C. While four physical heads and three clusters are illustrated in FIG. 4C, those of ordinary skill in the art will appreciate that any suitable number of heads and clusters can be utilized.

Mapping continues with tracks associated with the first logical head subsequent to track 402C. When the last track of the cluster 404A associated with the first logical head is mapped, mapping is switched to the next logical head (e.g., logical head 1 mapped to the physical head 104D), and tracks associated with that logical head are mapped in reverse direction. In other words, tracks associated with the logical head 1 are mapped beginning with the track in the same cylinder as the last mapped track associated with the logical head 0 and ending with the track associated with the logical head 1 located closest to the outer diameter. Mapping is then switched to the next logical head (e.g., logical head 2 mapped to the physical head 104A), and tracks associated with that logical head in cluster 404A are mapped in forward direction. This process continues until all tracks in cluster 404A associated with all logical heads have been mapped. At that point, mapping proceeds with the track associated with the first logical head (e.g., logical head 0 mapped to the physical head 104B) in the next cluster (e.g., cluster 406A). This process continues until all tracks in all clusters have been mapped.

By using sorted serpentine mapping 400C, the disk drive can achieve substantially monotonically decreasing data rates within a particular cluster. This is due to starting the traversal with tracks associated with the physical head mapped to the logical head associated with the highest data rate and continuing the traversal until tracks associated with the physical head mapped to the logical head associated with the lowest data rate are mapped. When mapping switches to a subsequent cluster having a lower data rate, there may be increase in the data rate because of the switch from the physical head having the lowest data rate (tracks associated with which were mapped last) to the physical head having the highest data rate (tracks associated with which are being currently mapped).

FIG. 5 further illustrates sorted serpentine mapping 400C of FIG. 4C. Cylinders are grouped into a plurality of clusters C₁ (520), C₂ (522), C₃ (524), C₄ (526), C₅ (528), C₆ (530) through C_N (532) sorted according to the respective data rate of the associated cluster. For example, clusters located closer to the outer diameter of the platters (e.g., farthest from the central axis 202) associated with the surfaces 102A, 102B, 102C, and 102D have higher data rates than clusters located closer to the inner diameter of the platters. Four physical heads 104A, 104B, 104C, and 104D access and map tracks associated with the surfaces 102A, 102B, 102C, and 102D. While four physical heads are illustrated in FIG. 5, those of ordinary skill in the art will appreciate that any suitable number of heads can be utilized.

In one embodiment, physical heads are assigned or mapped to logical heads sorted by the data rate of the associated physical heads. As explained above, physical head 104B determined to have the highest data rate is mapped to the first logical head (e.g., logical head 0), physical head 104D determined to have the next highest data rate is mapped to the second logical head (e.g., logical head 1), physical head 104A determined to have the next highest data rate is mapped to the third logical head (e.g., logical head 2), and physical head 104C determined to have the lowest data rate is mapped to the last logical head (e.g., logical head 3). Disk surfaces 102A-102D are mapped beginning with track 502 (located closest to the outer diameter) associated with the first logical head (e.g., logical head 0 mapped to the physical head 104B). Mapping continues with tracks associated with the first logical head subsequent to track 506.

When the last track of the cluster 520 associated with the first logical head is mapped, mapping is switched to the next logical head (e.g., logical head 1 mapped to the physical head 104D), and tracks associated with that logical head are mapped in reverse direction starting with track 504. In other words, tracks associated with the logical head 1 are mapped beginning with the track in the same cylinder as the last mapped track associated with the logical head 0 and ending with the track associated with the logical head 1 located closest to the outer diameter. Mapping is then switched to the next logical head (e.g., logical head 2 mapped to the physical head 104A), and tracks associated with that logical head are mapped in forward direction starting with track 506 and continuing with subsequent tracks in cluster 520 associated with the logical head 2. When the last track associated with that logical head is mapped, mapping is switched to the last logical head (e.g., logical head 3 mapped to the physical head 104C), and tracks associated with the last logical head are mapped in reverse direction starting with track 508. When the track associated with the logical head 3 located closest to the outer diameter is mapped, mapping proceeds with track 510 associated with the first logical head in the next cluster 522. This process continues until all tracks in all clusters have been mapped.

FIG. 6 illustrates a flow diagram 600 of sorted serpentine mapping according to one embodiment of the invention. The process 600 can be performed by the controller 330 and/or controller 14. In one embodiment, the process 600 begins in block 602, where it maps the physical head determined to have the highest data rate to the first logical head (e.g., logical head 0). In block 604, the process maps a track in a current cluster (e.g., cluster having the highest data rate) and associated with the physical head mapped to the first logical head, and where the track has the highest data rate of all tracks of the disk drive. For example, this track can correspond to the track located closest to the outer diameter in the cluster having the highest data rate (e.g., track 402C and/or 502). This entire track or a segment in this track can be mapped as the first logical address of the disk drive (e.g., LBA 0).

In block 606, the process 600 continues mapping subsequent tracks until the last track in the current cluster (e.g., cluster where the track mapped in block 604 is located) associated with the physical head mapped to the first logical head is mapped. These tracks are part of a plurality of cylinders. In block 608, the process 600 maps the physical head having the next highest data rate to the next logical head (e.g., logical head 1). The process 600 also switches to that physical head. In block 610, the process 600 maps the track in the same cylinder as the last mapped track (e.g., associated with the previous physical head). In block 612, the process 600 continues to map tracks associated with the current physical head in opposite direction. In other words, if tracks associated with the previous physical head were mapped in forward direction (e.g., toward the inner diameter tracks), mapping is performed in reverse direction. Conversely, if tracks associated with the previous physical head were mapped in reverse direction (e.g., toward the outer diameter tracks), mapping is performed in forward direction. In another embodiment, the direction of mapping is not reversed (e.g., same direction is maintained). Mapping in block 612 is continued until all tracks in the current cluster associated with the current physical head have been mapped.

In block 614, the process 600 determines whether all tracks in the current cluster associated with all physical heads have been mapped. If not, the process 600 transitions to block 608, where the process maps the physical head having the next highest data rate to the next logical head. The process 600 switches to that physical head, and mapping continues as described above. If all tracks of the current cluster have been mapped, the process 600 transitions to block 618, where it determines whether all tracks in all clusters have been mapped. If not, the process 600 transitions to block 620, where it switches to the physical head having the highest data rate (e.g., mapped to the first logical head). The process 600 also switches to the cluster adjacent to the current cluster. In one embodiment, the process 600 switches to the cluster having the next highest data rate, such as a cluster that is immediately adjacent in the direction toward the inner diameter. The process 600 transitions to block 604, and mapping continues as is described above. In one embodiment, mapping terminates when all tracks in all clusters have been mapped. Mapping can be restarted from the track having the highest data rate of all tracks of the disk drive (e.g., track mapped to the first logical address in the disk drive). In another embodiment, mapping may not necessarily map all tracks in all clusters using the above-described approach. For example, a track or tracks in a particular cluster or clusters, or an entire cluster or a plurality of clusters can be mapped using any other suitable mapping scheme.

FIG. 7 illustrates a flow diagram 700 for performing storage access operations according to one embodiment of the invention. The process 700 can be performed by the controller 330 and/or controller 14. In one embodiment, the process 700 begins in block 702, where it determines the logical address (e.g., LBA) associated with a storage access operation, such as read data command and/or write data command. In block 704, the process determines the physical address or address of physical location in the disk drive that is associated with the specified logical address. In one embodiment, the physical address can be in a form of a {cylinder, logical head, sector} tuple. Those of ordinary skill in the art will appreciate that any suitable address format can be used.

In block 706, the process 700 determines the physical head associated with or mapped to the logical head corresponding to the logical address. In one embodiment, the process 700 can look up the physical head using the mapping stored in the cache buffer 362. Once the physical head has been determined, the process 700 transitions to block 708. The process 700 accesses the physical location in the disk drive that corresponds to the logical address specified by the storage access operation. Depending on the type of storage access operation, the process can retrieve data stored in the physical location (and in physical locations adjacent or subsequent to the physical location) and/or store host data in the physical location (and in physical locations adjacent or subsequent to the physical location).

CONCLUSION

Using sorted serpentine mapping, a plurality of physical heads of the disk drive is associated with or mapped to a plurality of logical heads. The plurality of logical heads is sorted according to the respective data rates of the physical heads, such as in the order of decreasing data rates. Data tracks are mapped using serpentine mapping that utilizes the sorted order of the logical heads. The first logical address of the disk drive (e.g., LBA 0) is mapped to a physical location (e.g., sector) in the disk drive having the highest data rate. In addition, when cylinders are grouped into a plurality of clusters of predetermined of fixed size and the plurality of clusters is sorted according to the data rate, seek times within a particular cluster are constrained. Furthermore, the disk drive can achieve substantially monotonically decreasing data rates within a particular cluster. Moreover, the average throughput of the disk drive is increased. Improved performance can thereby be attained.

Other Variations

Those skilled in the art will appreciate that in some embodiments, other types of mapping can be implemented. For example, in one embodiment, sorted serpentine mapping can be performed without reversing the direction of mapping after physical heads are switched. In addition, the actual steps taken in the disclosed processes, such as the process illustrated in FIGS. 6 and 7, may differ from those shown in the figures. Depending on the embodiment, certain of the steps described above may be removed, others may be added. Accordingly, the scope of the present disclosure is intended to be defined only by reference to the appended claims.

While certain embodiments have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the protection. Indeed, the novel methods and systems described herein may be embodied in a variety of other forms. Furthermore, various omissions, substitutions and changes in the form of the methods and systems described herein may be made without departing from the spirit of the protection. The accompanying claims and their equivalents are intended to cover such forms or modifications as would fall within the scope and spirit of the protection. For example, the systems and method disclosed herein can be applied to hard disk drives, hybrid hard drives, and the like. In addition, other forms of storage (e.g., DRAM or SRAM, battery backed-up volatile DRAM or SRAM devices, EPROM, EEPROM memory, etc.) may additionally or alternatively be used. As another example, the various components illustrated in the figures may be implemented as software and/or firmware on a processor, ASIC/FPGA, or dedicated hardware. Also, the features and attributes of the specific embodiments disclosed above may be combined in different ways to form additional embodiments, all of which fall within the scope of the present disclosure. Although the present disclosure provides certain preferred embodiments and applications, other embodiments that are apparent to those of ordinary skill in the art, including embodiments which do not provide all of the features and advantages set forth herein, are also within the scope of this disclosure. Accordingly, the scope of the present disclosure is intended to be defined only by reference to the appended claims.