Patent Application
20150109700
This disclosure relates to data storage devices, and more particularly to techniques for optimizing capacity and the data transfer rates of a storage device.
Data storage devices can be incorporated into a wide range of devices, including large scale data centers, laptop or desktop computers, tablet computers, digital video recorders, set-top boxes, digital recording devices, digital media players, video gaming devices, video game consoles, cellular telephones, and the like. Data storage devices may include hard disk drives (HDD). HDDs include one or multiple magnetic disks having positive or negative areas of magnetization. Bits of data may be represented using the positive and negative areas of magnetization. Blocks of data are arranged to form tracks on a rotating disk surface. A magnetic transducer may be used to read data from the disk and write data to a disk. Different magnetic recording techniques may be used to store data to the disk. Magnetic recording techniques include, for example, longitudinal magnetic recording (LMR), perpendicular magnetic recording (PMR), and shingled magnetic recording (SMR). Heat assisted magnetic recording (HAMR) may be used with LMR, PMR, or SMR.
During operation, the disk of a HDD may be subjected to vibrations caused by normal operation of the disk or by ambient conditions. In some cases, ambient conditions may introduce severe vibrations that inhibit the ability of a HDD to properly perform read and/or write operations and data transfer rates may degrade.
In general, this disclosure describes techniques for writing data to a storage medium. In particular, this disclosure describes techniques for defining areas of a storage disk and adaptively writing data to areas of the storage disk. The techniques of this disclosure may be particularly useful for selecting an area within a write buffer area, including an E-region of an SMR disk, to write data to.
According to one example of the disclosure, a method of writing data to a buffer region of a shingled magnetic recording disk, comprises calculating a position accuracy value associated with a data writing head, selecting one of a first zone and a second zone of the buffer region based on the position accuracy value, wherein the first zone includes an initial track density that is greater than an initial track density of the second zone, and writing data to the selected zone.
According to another example of the disclosure, a shingled magnetic disk comprises a data buffer region including a first zone and a second zone, wherein the first zone includes an initial track density that is greater than an initial track density of the second zone and a processing unit configured to calculate a position accuracy value associated with a data writing head, select one of a first zone and a second zone of the buffer region based on the position accuracy value, and write data to the selected zone.
According to another example of the disclosure a non-transitory computer-readable storage medium has instructions stored thereon that upon execution cause one or more processors of a hard disk drive to calculate a position accuracy value associated with a data writing head, select one of a first zone and a second zone of a buffer region of a shingled magnetic disk based on the position accuracy value, wherein the first zone includes an initial track density that is greater than an initial track density of the second zone, and write data to the selected zone.
According to another example of the disclosure, an apparatus for writing data to a buffer region of a magnetic disk comprises means for calculating a position accuracy value associated with a data writing head, means for selecting one of a first zone and a second zone of the buffer region based on the position accuracy value, wherein the first zone includes an initial track density that is greater than an initial track density of the second zone, and means for writing data to the selected zone.
The details of one or more examples are set forth in the accompanying drawings and the description below. Other features, objects, and advantages will be apparent from the description and drawings, and from the claims.
In general, this disclosure describes techniques for writing data to a storage medium. In particular, this disclosure describes a disk with different areas. The different areas of the disk may be included within a designated write buffer area of a disk. A designated write buffer area of a disk may be an E-region of a SMR disk. Techniques described in this disclosure may be used for selecting an area of the disk to write data. Data may be written to each area of the disk using one or more of a respective track pitch, track density, and/or write inhibit value.
Disks of hard disk drives (HDD) may be subjected to vibrations caused by normal operation of the disk drive or by ambient conditions. For example, an HDD may be coupled to a device with speakers that introduce severe vibrations. Vibrations may reduce the ability of a write head to accurately maintain an alignment with respect to a track of a disk. Severe vibrations may cause Track Mis-Registration (TMR) and a HDD may not be able to perform write operations due to TMR.
In some cases, HDD disks are designed to accommodate predictable vibrations caused by normal operations. However, accommodating less predictable severe vibrations caused by ambient conditions is more difficult. In some cases, a HDD may repeatedly retry a write operation until a time when vibrations are reduced such that a write operation can be performed. However, write operation retries may cause data transfer rates to deteriorate greatly during prolonged periods of vibration. Further, in some cases, a write operation may time-out after a number of unsuccessful write retries. In other cases, a HDD may write data to a disk using relatively greater track densities at inner diameters in order to achieve a consistent TMR rate across the disk based on predicted operating conditions. However, such HDDs may not adequately address severe vibrations that occur outside of the predicted operating conditions and if track densities are based on worst case conditions, capacity may be limited. Further, such HDDs may not be configured to adaptively perform write operations based on vibrations that are actually being experienced by the HDD.
Some HDD disks include a data buffering or staging area where data is written before the data is written to another area of disk. For example, disks where tracks are shingled, such as shingled magnetic recording (SMR) disks, may include a data buffering region (e.g., an E-region) and a regular data storage region (e.g., an I-region). Data may be initially written to an E-region and then written to an I-region. Vibrations caused during write operations associated transferring data from a data buffer may be especially problematic due to the relatively limited amount of disk storage capacity reserved for data buffering (e.g., vibrations may result in buffer overflows). In view of the above, this disclosure provides techniques for adaptively writing data to defined areas of a magnetic disk based on determined vibration conditions. While the techniques of this disclosure may be particularly useful for a HDD including a SMR disk with a write buffer area, the techniques described herein may be readily applied to other types of HDDs as well.
Disk 102 includes a stack of one or more disks having magnetic material deposited on one or both sides thereof. Disk 102 may be composed of a light aluminum alloy, ceramic/glass, or other suitable substrate that magnetic material may be deposited thereon. Using electromagnetic techniques, data may be stored on disk 102 by orientating an area of the magnetic material. Data stored on disk 102 may be organized as data blocks. Data blocks are typically 512 bytes or 4 KB in size, but may be other sizes as well. The magnetic areas of disk 102 may be arranged into a set of radially-spaced concentric tracks, illustrated in
Magnetic material of disk 102 may be configured according to one of a plurality of magnetic recording techniques. Examples of magnetic recording techniques include longitudinal magnetic recording (LMR) and perpendicular magnetic recording (PMR). Additional magnetic recording techniques include shingled magnetic recording (SMR) and heat assisted magnetic recording (HAMR). SMR is a type of PMR that increases bit density compared to conventional PMR by allowing tracks to be written in a manner that allows overlap of one or more adjacent tracks. HAMR may be used in conjunction with LMR, PMR, or SMR techniques to achieve higher areal storage density.
The bit density, distance between tracks (referred to as track pitch), and track density of disk 102 may vary according to a magnetic recording technique. Track density may be defined according to the number of tracks per radial length (e.g., track per inch (TPI)). Track pitch may be defined as a target or nominal distance between tracks. Further, as illustrated in
In one example, disk 102 is configured to store data using SMR. In contrast to conventional PMR tracks, which can be written in any order, shingled tracks are written according to a particular sequence. The tracks of an SMR disk are typically organized into a plurality of shingled regions (referred to as I-regions) in which tracks are written sequentially from an inner diameter (ID) to an outer diameter (OD) or in some examples, from OD to ID. In order to improve the performance of SMR disks, a portion of the disk may be allocated to a so-called “exception region” (i.e., E-region, which may also be referred to as write buffer area or E-cache).
An E-region may be used as a staging area for data which will ultimately be written to an I-region. An E-region may be configured to have either shingled or unshingled tracks. It should be noted that in some examples, hard disk drive 100 may include a so-called hybrid E-region where a hybrid E-region includes areas of disk 102 and portion data SRAM 130 which may be used by hard disk drive 100 for data staging. Typically, only a minor portion of a disk's area is an E-region (e.g., 5%-10%). In one example, R0 of disk 102 may be an E-region and R1 of disk 102 may be an I-region. In other examples, disk 102 may include multiple I-regions and multiple E-regions where I-regions and E-regions are located at various radial positions of disk 102 (e.g., disk 102 may include concentric circular bands of E-regions and I-regions). Further, as described above, disk 102 may include multiple disks. Thus, particular E-regions and I-regions need not be included on the same disk (e.g. an E-region may be located on a first disk and an I-region may be located on a second disk). Further, it should be noted that although not shown disk 102 may include other types of regions including, for example, guard regions and write-twice cache regions. It should be noted that other types of magnetic recording techniques may incorporate techniques that allocate areas of a disk to be used as a write buffer area. Thus, disk 102 may be configured according to any combination of magnetic recording techniques where R0 or R1 is a write buffer area of a magnetic disk 102 and the other area is the area where data is ultimately written.
As described above, a region may include a plurality of zones defined according to a number of radially adjacent tracks or according to a radial distance.
As described above, data stored on disk 102 may be organized as data blocks. In some examples, address indirection may be used to write blocks of data to disk 102. That is, host file systems may use logical block addresses (LBAs) in commands to read and write blocks of data to disk 102 without regard for actual locations (physical block address (PBA)) used internally by hard disk drive 100. Address indirection may be particularly useful to shield a host from the complexities associated with SMR. As described in detail below, address indirection is typically implemented in the controller portion of the drive's architecture (e.g., hard disk controller 122) using indirection tables.
Referring again to
As illustrated in
Slider 106 is configured to read and write data to disk 102 according to a magnetic recording technique, for example, any of the example magnetic recording techniques described above. Slider 106 may include read and write heads corresponding to each of a plurality of disks included as part of disk 102. Further, slider 106 may include one or more read and write heads for each disk. Slider 106 is coupled to preamplifier 116. Preamplifier 116 may also be referred to as arm electronics (AE). Preamplifier 116 is configured to select a correct head from a plurality of heads for a particular read or write operation. Preamplifier 116 is configured to drive head 106 with a write current, during a write operation. Further, preamplifier 116 is configured to amplify read signals from slider 106, during a read operation using a programmable head bias current. Preamplifier 116 may also be configured to detect errors during each of the read and write operations. Preamplifier 116 may include a signal adaptive filter (SAF) for thermal asperity (TA) recovery during a read operation.
Preamplifier 116 receives data to be written to disk 102 from read/write data channel unit 118. Further, preamplifier 116 provides data read from disk 102 to read/write data channel unit 118. Data may originate from a host device and may be communicated to read/write data channel unit 118 via host interface unit 136 and processing unit 120. Host interface unit 136 provides a connection between hard disk drive 100 and a host device. Host interface unit 136 may operate according to a physical and logical characteristics defined according to a computer bus interface. Example standardized interfaces include ATA (IDE, EIDE, ATAPI, UltraDMA, SATA), SCSI (Parallel SCSI, SAS), Fibre Channel, and PCIe (with SOP or NVMe).
As illustrated in
Hard disk controller 122 generally represents the portion of processing unit 120 configured to manage the transfer of blocks of data to and from host interface unit 136 and read/write data channel unit 118. Hard disk controller 122 may be configured to perform operations to manage data buffering and may interface with host interface unit 136 according to a defined computer bus protocol, as described above. For example, hard disk controller 122 may receive and parse packets of data from host interface unit 136. Further, hard disk controller 122 may be configured to communicate with host. For example, hard disk controller 122 may be configured to report errors to host and format disk 102 based on commands received from host.
As described above, hard disk controller 122 may be configured perform address indirection. That is hard disk controller 122 may translate the LBAs in host commands to an internal physical address, or an intermediate address from which a physical address can ultimately be derived. It should be noted in for a hard disk drive that utilizes SMR the physical block address (PBA) of a logical block address (LBA) can change frequently. Further, for an SMR hard disk drive, the LBA-PBA mapping can change with every write operation because the hard disk drive may dynamically determine the physical location on the disk where the data for an LBA will be written. Further, the data for the same LBA may be written to a different location the next time the host LBA is updated. In addition, a hard disk drive, such as hard disk drive 100, may autonomously move data between write caches in RAM, write caches on disk 102, E-regions on disk 102 and I-regions on disk 102. In some examples, the LBAs for the data may stay the same regardless of where the hard disk drive 100 has the data stored. Further, background processes, such as, for example, defragmentation may also be executed autonomously by hard disk drive 100 to move data sectors from one PBA to another while the LBA stays the same. Thus, hard disk drive 102 may be configured to use address indirection in associated with multiple types of write operations.
Interface processor 124 generally represents the portion of processing unit 120 configured to interface between servo processor 126 and hard disk controller 122. Interface processor 124 may perform predictive failure analysis (PFA) algorithms, data recovery procedures, report and log errors, perform rotational positioning ordering (RPO) and perform command queuing. In one example, interface processor may be an ARM processor.
As described above, data is typically written to or read from disk 102 in blocks which are contained within a sector of a particular track. Disk 102 may also include one or more servo sectors within tracks. Servo sectors may be circumferentially or angularly-spaced and may be used to generate servo signals. A servo signal is a signal read from disk 102 that may be used to align slider 106 with a particular sector or track of disk 102. Server processor 126 generally represents the portion of processing unit 120 configured to control the operation of spindle assembly 104 and voice coil motor assembly 110 to insure slider 106 is properly positioned with respect to disk 102. Servo processor 126 may be referred to as a Servo Hardware Assist Real-time Processor (SHARP). Servo processor 126 may be configured to provide closed loop control for any and all combinations of: slider position on track, slider seeking, slider settling, spindle start, and spindle speed. Further, servo processor 126 may include a set of instructions to calculate servo related variables.
For example, a servo signal read from disk 102 may be used by servo processor 126 to determine the amount of displacement of slider 106 from the center of a track. The value of this displacement may be referred to as run out. Run out may include repeatable run out (RRO) caused by spindle assembly or non-repeatable run out (NRRO) caused by internal or external vibrations. Run out is one example of a position accuracy associated with a read or write included in slider 106. If a measured value of NRRO is greater than a predetermined threshold, a write operation may be inhibited. This threshold may be referred to as write inhibit slice (Winh). In addition to servo signals, sensors such as vibration measurement unit 138 may provide information to servo processor 126 and/or processing unit 120 that allow processing unit 120 to calculate a position accuracy value associated with a head of slider 106. Vibration measurement unit 138 may be any device configured to measure vibration of hard disk drive 100 and/or disk 102 and provide a measurement to processing unit 120. In one example, vibration measurement unit 138 may include an accelerometer.
As described above, a position accuracy value associated with a data writing head, e.g., run out, may be used to select one of zone 202, zone 204, and zone 206 of disk 102 to write data to. Further, data may be adaptively written to each of zone 202, zone 204, and zone 206 using respective track densities, track pitches, and/or write inhibit values based on a position accuracy value.
In the example illustrated in
As illustrated in
Track density calculator 306 may be configured to receive a run out value and determine a track density.
where hd(t) is the head position, t is time, and w is the window width.
Track density calculator 306 may calculate a track density based on a received calculated run out value. In one example, the run out value and the calculated track density have an inverse relationship, i.e., as run out increases, the track density decreases. In other examples, the determined run out value and the calculated track density may have a different relationship. For example, the relationship between track density and run may be based on the following equation:
td(t)=25400/(ntp+(ro(t)−nro/C),
where td(t) is the track density in kTPI, ntp is a nominal track pitch in nm, which may be set as a constant value, nro is a nominal run out in nm, which may be set as a constant value, as C is a constant scaling factor.
Referring again to
Track density sampling unit 502 may be configured to receive a write command and sample a track density value. A write command may include a data block and an LBA. In one example, track sampling unit 502 is configured to receive a plurality of track density values from track density calculator 306 and sample one or more track density values in conjunction with receiving a write command. The track density calculator output illustrated in graphical form in
Write zone selection unit 504 may be configured to receive one or more sampled track density values and/or an average track density value and select one of a plurality of zones to write data based on the one or more track density values. In one example, write zone selection unit 504 may select one of a plurality of zones within a buffer area of a disk, e.g., zone 202, zone 204, and zone 206 within an E-region of an SMR disk. In one example, each of zone 202, zone 204, and zone 206 may be configured such that data is written to one of zone 202, zone 204, and zone 206 using a respective track density. For example, zone 202 may be configured such that data is written to zone 202 using a track density of 300,000 tracks per inch (i.e., 300 k TPI), zone 204 may be configured such that data is written to zone 204 using a track density of 325,000 tracks per inch (i.e., 325 k TPI), and zone 206 may be configured such that data is written to zone 206 using a track density of 350,000 tracks per inch (i.e., 350 k TPI). Thus, region R0 may be configured such that tracks are written at higher densities as the radial distance from the center of disk 102 increases. In another example, region R0 may be configured such that tracks are written at lower densities as the radial distance from the center of disk 102 increases.
In one example, write zone selection unit 504 may be configured to select a zone such that a data is written to the zone with the highest density that does not exceed the track density. For example, in the case where zone 202 has a track density of 300 k TPI, zone 204 has a track density of 325 k TPI, and zone 206 has a track density of 350 k TPI, and a sampled track density is 340 k TPI, write zone selection unit 504 may be configured to select zone 204. In another example, where zone 202 has a track density of 300 k TPI, zone 204 has a track density of 325 k TPI, and zone 206 has a track density of 350 k TPI, write zone selection unit 504 may be configured to select one of zone 202, zone 204, and zone 206 based on the following run out value ranges: (1) select zone 206 if run out is less than or equal to 1 nm; (2) select zone 204 if run out is greater than 1 nm and less than 2 nm; (3) select zone 202 if run out is greater than or equal to 2 nm. In one example, the run out value used to select one of zone 202, zone 204, and zone 206 may be based on a one second time window occurring immediately before a write command is received. It should be noted that for examples where write zone selection unit 504 selects a zone based on a run out value range, track density calculator 306 may output a track density value such that write zone selection unit 504 selects a desired zone based on a run out range. In this manner, run out calculator 304, track density calculator 306, track density sampling unit 502, and write zone selection unit 504 may be highly integrated.
Zone format selection unit 506 may be configured to format a selected zone. For example, in some cases it may be advantageous to adaptively determine a write inhibit value, a track pitch, and/or a track density for a particular situation. For example, in the example where zone 202 has a track density of 300 k TPI, zone 204 has a track density of 325 k TPI, and zone 206 has a track density of 350 k TPI, after one of zone 202, zone 204, or zone 206 is selected, a write inhibit value may be determined for writing data to the selected zone, where the write inhibit value seeks to accommodate predicted conditions (e.g., vibration conditions) that may occur during a write operation. Further, in some cases, one of zone 202, zone 204, or zone 206 may be full. That is, data capacity of a zone may be used with data that needs to be transferred out of the buffer (e.g., needs to be moved to the I-region).
In this case, an alternative zone may be selected and zone format selection unit 506 may be configured to change the format of the alternative zone that is selected. For example, in the example described above where zone 202 has a track density of 300 k TPI, zone 204 has a track density of 325 k TPI, and zone 206 has a track density of 350 k TPI and an calculated track density is 340 k TPI, if zone 204 is full, write zone selection unit 504 may select zone 206 as an alternative to zone 204 and zone format selection unit 506 may format zone 206 to have the same track density as zone 204 (i.e., 325 k TPI) or, alternatively, format selection unit 506 may format zone 206 to have a track density of 340 k TPI. Further, zone format selection unit 506 may be configured to format zone 202, zone 204, and zone 206 according to any position accuracy conditions (e.g., track densities may be reduced when sustained high levels of vibrations are experienced). In this manner, an initial track density of a zone may be adjusted.
Logical and physical address table updating unit 508 may be configured to receive a selected zone, a zone format, and/or information associated with a write command and output a physical address. In one example, logical and physical address table updating unit 508 may output a physical address to read/write data channel unit 118. Further, logical and physical address table updating unit 508 may be configured to update a logical and physical address table, where a logical and physical address table maps a logical address (e.g., a track number within Zone 202) to a physical address (e.g., a radial distance). As described above, hard disk drive 100 may use address indirection to map logical address to a physical address.
It should be noted that while the above examples describe writing data to a zone based on a position accuracy value associated with a data writing head, the techniques described herein are also applicable to transferring data from an E-region to an I-region. For example, data may be transferring from the E-region to an I-region (e.g., written to the E-region and/or read from the I-region) only when a vibration measurement signal is less than a threshold.
In one or more examples, the functions described may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, the functions may be stored on or transmitted over, as one or more instructions or code, a computer-readable medium and executed by a hardware-based processing unit. Computer-readable media may include computer-readable storage media, which corresponds to a tangible medium such as data storage media, or communication media including any medium that facilitates transfer of a computer program from one place to another, e.g., according to a communication protocol. In this manner, computer-readable media generally may correspond to (1) tangible computer-readable storage media which is non-transitory or (2) a communication medium such as a signal or carrier wave. Data storage media may be any available media that can be accessed by one or more computers or one or more processors to retrieve instructions, code and/or data structures for implementation of the techniques described in this disclosure. A computer program product may include a computer-readable medium.
By way of example, and not limitation, such computer-readable storage media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage, or other magnetic storage devices, flash memory, or any other medium that can be used to store desired program code in the form of instructions or data structures and that can be accessed by a computer. Also, any connection is properly termed a computer-readable medium. For example, if instructions are transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. It should be understood, however, that computer-readable storage media and data storage media do not include connections, carrier waves, signals, or other transient media, but are instead directed to non-transient, tangible storage media. Disk and disc, as used herein, includes compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk and Blu-ray disc, where disks usually reproduce data magnetically, while discs reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media.
Instructions may be executed by one or more processors, such as one or more digital signal processors (DSPs), general purpose microprocessors, application specific integrated circuits (ASICs), field programmable logic arrays (FPGAs), or other equivalent integrated or discrete logic circuitry. Accordingly, the term “processor,” as used herein may refer to any of the foregoing structure or any other structure suitable for implementation of the techniques described herein. Also, the techniques could be fully implemented in one or more circuits or logic elements.
The techniques of this disclosure may be implemented in a wide variety of devices or apparatuses, including a wireless handset, an integrated circuit (IC) or a set of ICs (e.g., a chip set). Various components, modules, or units are described in this disclosure to emphasize functional aspects of devices configured to perform the disclosed techniques, but do not necessarily require realization by different hardware units. Rather, as described above, various units may be combined in a hardware unit or provided by a collection of interoperative hardware units, including one or more processors as described above, in conjunction with suitable software and/or firmware.
Various examples have been described. These and other examples are within the scope of the following claims.
