Method for adjusting rw parameters of a disk device, adjusting device and disk device

Abstract

Provided is a RW parameter adjustment method, adjustment device, and disk device for a disk device, capable of balancing a plurality of required standards well within a limited test time and capable of shortening the adjustment time. The RW parameter adjustment method of the disk device of the embodiment performs a prior registration process, measures characteristics by changing conditions of a plurality of read / write-related parameters that determine capacity / characteristics for each head of various heads of a test disk device, creates a database for each head in which the measured results of the characteristics are registered, performs an adjustment process for an adjustment target disk device by changing the conditions of the parameters to measure characteristics for each head, searches for a database of a head of similar characteristics from the database of each head, and sets a proper balance for capacity / characteristics from all hit conditions based on the characteristics of the found database by estimation of a neural network computer.

Description

[0001] Related applications

[0002] This application enjoys priority over Japanese Patent Application No. 2022-042712 (filed on March 17, 2022). This application incorporates the entire contents of that earlier application by reference. Technical Field

[0003] The implementation relates to a method for adjusting the RW parameters of a disk device, as well as an adjustment device and a disk device. Background Technology

[0004] When achieving high capacity, hard disk drives (HDDs) require adjustments to optimal read / write (RW) parameters such as TPI (Track per Inch), BPI (Bits per Inch), Iw (Writer Current), and OSA (Overshoot Amplifier). This is especially true when combining the head and media, where various standards must be met individually, and adjustments must be made within a limited testing timeframe.

[0005] In the existing RW parameter adjustment, because the mechanism of determining BPI / TPI / Iw / OSA is serial, it is difficult to balance and meet various requirements and standards well. In addition, the adjustment takes time. Summary of the Invention

[0006] As mentioned above, the existing RW parameter adjustment methods for disk devices have the following technical problems: because the mechanism for determining the requirements of RW parameters such as BPI / TPI / Iw / OSA is serial, it is difficult to balance and satisfy various requirements. In addition, the adjustment takes time.

[0007] The purpose of this implementation is to provide a method for adjusting the RW parameters of a disk device, as well as an adjustment device and a disk device, which can balance and meet multiple requirements within a limited test time and shorten the adjustment time.

[0008] One embodiment of the disk device RW parameter adjustment method involves pre-registration processing, measuring the characteristics by changing multiple read / write related parameters that determine capacity / characteristics for each head of the test disk device, and creating a database for each head that records the measurement results of the characteristics.

[0009] For the adjustment process of the disk device to be adjusted, the characteristics are determined according to the condition of changing the parameters of each head, the database of heads with similar characteristics is searched from the database of each head, and based on the characteristics of the found database, the appropriate balance setting for capacity / characteristics is made from all hit conditions through the estimation of the neural network computer. Attached Figure Description

[0010] Figure 1 This is a block diagram illustrating the configuration of an RW parameter adjustment device for a disk device RW parameter adjustment method based on an implementation method.

[0011] Figure 2 It means Figure 1 A flowchart of the RW parameter adjustment method applicable to the RW parameter adjustment device shown.

[0012] Figure 3 It means Figure 1 A conceptual diagram illustrating an example of the use of a neural network PC for adjusting RW parameters.

[0013] Figure 4 This is a block diagram illustrating the configuration of a disk device adjusted by the RW parameter adjustment method of this embodiment.

[0014] Figure 5 This is a diagram illustrating an example of error rate optimization for a disk device after RW parameter adjustment using the adjustment method of this embodiment.

[0015] Figure 6 This is a diagram illustrating an example of accuracy optimization of a disk device by adjusting the RW parameters using the adjustment method of this embodiment.

[0016] Figure 7 This is a diagram illustrating an example of TPIM optimization for a disk device whose RW parameters have been adjusted using the adjustment method of this embodiment. Detailed Implementation

[0017] Hereinafter, embodiments will be described with reference to the accompanying drawings. Furthermore, the disclosure is merely an example and is not intended to limit the invention by the content described in the following embodiments. Modifications readily conceived by those skilled in the art are naturally included within the scope of the disclosure. To make the description clearer, the drawings sometimes schematically illustrate changes in the dimensions, shapes, etc., of various parts in the actual embodiments. In multiple drawings, sometimes the same reference numerals are used to label corresponding elements, and detailed descriptions are omitted.

[0018] First, in the existing RW parameter adjustment method for disk devices, the drive current Iw is temporarily determined to obtain the desired O / W (Over / Write). Then, after temporarily determining the OSA that satisfies the desired error rate, the TPI / BPI that satisfies the desired characteristics is determined. Therefore, the TPI / BPI is adjusted, the OSA is adjusted again, and this process is repeated several times to determine the TPI / BPI / Iw / OSA.

[0019] In the existing methods described above, although adjustments are made to meet the desired capacity / characteristics, deviations occur due to the serial adjustments. In addition, the number of measurements required after changing TPI / BPI / Iw / OSA is also much larger.

[0020] Therefore, in the RW parameter adjustment of this embodiment, by performing a pre-registration process, the characteristics are measured by changing multiple read / write related parameters such as TPI / BPI / Iw / OSA for each head of the various heads of the test disk device in advance. A database of measured characteristics is created for each head, and heads with similar characteristics are searched from the database. Based on the characteristics of the found database, an appropriate balance setting is made for capacity / characteristics from all hit conditions through estimation implemented by a neural network computer, thereby solving the existing problems.

[0021] Figure 1 This is a block diagram illustrating the configuration of an RW parameter adjustment device in a disk device RW parameter adjustment method based on an implementation method. Figure 1 In this diagram, 100 is the Neural Network Computer (hereinafter referred to as Neural Network PC), 200 is the Host (hereinafter referred to as Host PC), and 300 is the Machine Interface (hereinafter referred to as Machine IF). Machine IF 300 contains n disk drives (hereinafter referred to as HDDs) 400-1 to 400-n.

[0022] The aforementioned neural network PC100 has a database storage unit 101 that stores a database representing the characteristics of each head, and its processing operations are broadly divided into registration processing and adjustment processing.

[0023] Figure 2 This is a flowchart representing the processing actions of the aforementioned neural network PC100. (a) shows the registration process, and (b) shows the adjustment process. Figure 3 This is a conceptual diagram representing an example of the use of the aforementioned neural network PC100.

[0024] First, during the registration process, various RW heads for experimental HDD400-1 to 400-n, housed in machine IF300 via host PC200, such as... Figure 2As shown in (a), the heads are specified sequentially (step S11), the conditions (input RW parameters) of TPI / BPI / Iw / OSA are changed in a large number (many, many) mode, the characteristics of O-BER / F-BER / TPIM / ATI for each input parameter are measured (step S12), the characteristic results obtained by measuring each head are databased and registered in the database storage unit 101 (step S13), and the entire head is processed repeatedly (step S14).

[0025] As a database, for example, 100 heads of HDDs are prepared in advance. Adjusted values ​​for parameters determining the capacity / characteristics of each head / zone are used to modify the TPI / BPI / Iw / OSA to meet the maximum areal density of the criteria. A large number of measurements are performed on the characteristics of O-BER / F-BER / TPIM / ATI (correspondence: TPI→O-BER, BPI→F-BER, Iw→TPIM, OSA→OW), and the data is databased for each head. The large number of measurements are implemented, for example, with 15,000 combinations (models). The neural network PC100 used is as follows... Figure 3 As shown, it has an input layer (adjustment value), intermediate layer 1 / intermediate layer 2, and output layer (characteristic value), and can arbitrarily set the number of inputs and outputs. Various activation functions / number of layers / patch size / evaluation functions are considered to be the best settings.

[0026] In the adjustment process, the HDD being adjusted is used, such as... Figure 2 (b) As shown, heads are specified sequentially (step S21), and a small number of measurements of 20 points are performed on each head / zone using a combination (model) of TPI / BPI / Iw / OSA and O-BER / F-BER / TPIM / ATI (step S22). Then, a similar database (model) is searched using RMS (Root Mean Square) (step S23). By using neural network processing of this database, the O-BER / F-BER / TPIM / ATI for each TPI / BPI / Iw / OSA is estimated with perfect accuracy, thereby estimating the optimal parameters when the assumption follows the database (step S24). The parameters of the specified head are then set to the estimated optimal parameters (step S25). This process is repeated for all heads (step S26). Thus, the result corresponding to the conditions can be estimated, and the optimal parameter settings can be balanced. Finally, the parameter adjustment log is registered in the internal memory of the adjusted HDD so that it can be confirmed at the factory (step S27).

[0027] Furthermore, while the above-described adjustment process demonstrates serial processing of HDDs, it is also possible to process a large number of disk devices in parallel. Additionally, the pre-registration process allows for the modification of database registration information, and the adjustment process can include searching the modified database. Furthermore, the database storage unit 101 can be integrated within the neural network PC 100.

[0028] Figure 4 This is a block diagram showing the specific configuration of an HDD that is adjusted by the aforementioned RW parameter adjustment device.

[0029] HDD1 consists of a head-disk assembly (HDA) 10, a head amplifier integrated circuit (hereinafter referred to as head amplifier IC) 17, and a system-on-a-chip (SOC) 20.

[0030] The HDA10 includes: a disk 11, a spindle motor (SPM) 12, an arm 13, and a voice coil motor (VCM) 16. The disk 11 is rotated by the SPM 12. A load beam 14 is mounted at the front end of the arm 13, and a read / write head 15 is mounted at the front end of the load beam 14. The arm 13, driven by the VCM 16, moves the read / write head 15 to a designated position on the disk 11.

[0031] The read head 15 is a configuration in which the read head element and the write head element are separately mounted on a slider. The read head element reads the data recorded on the disk 11. The write head element writes data to the disk 11.

[0032] The head amplifier IC16 includes a read amplifier and a write driver. The read amplifier amplifies the read signal read by the read head element and transmits it to the read / write (R / W) channel 22. On the other hand, the write driver transmits the write current corresponding to the write data output from the R / W channel 22 to the write head element.

[0033] SOC 20 includes: a microprocessor (CPU) 21, a read / write channel 22, a disk controller 23, and a positioning controller 24. CPU 21 is the main controller of the drive, performing servo control for positioning the read / write head 15 via positioning controller 24 and read / write control for data execution via head amplifier IC 16. R / W channel 22 includes a read channel for performing signal processing for reading data and a write channel for performing signal processing for writing data. Disk controller 23 performs interface control for controlling data transfer between the host system (not shown) and R / W channel 22. Furthermore, positioning controller 24 can be implemented either in hardware or in software (firmware).

[0034] The memory 25 includes both volatile and non-volatile memory. For example, the memory 25 includes a buffer memory composed of DRAM and flash memory. The non-volatile memory of the memory 25, together with a storage unit (not shown) storing programs required for processing by the CPU 21, has a parameter storage unit 26 that stores the RW parameter adjustment values ​​during the aforementioned RW parameter adjustment process, and an adjustment log storage unit 27 that stores an adjustment log indicating that the adjustment is complete. That is, the adjustment log storage unit 27 stores the parameter adjustment log of the aforementioned step S27; by reading this log, it is easy to confirm that the HDD head adjustment has been optimized.

[0035] Furthermore, the storage units 26 and 27 can be stored in any storage area within the HDD1, even if they are not stored in the memory 25.

[0036] The effects of the above adjustment methods are explained below.

[0037] Figure 5 This is a graph illustrating an example of optimized estimated accuracy of the HDD after RW parameter adjustment using the adjustment method of this embodiment. Figure 6 This is a graph representing an example of error rate optimization for this HDD. Figure 7 This is a diagram representing the TPIM optimization example for this HDD. In Figure 6 In this context, Conv represents the error rate set using the existing adjustment method, and New represents the error rate set using the adjustment method of the implementation method. Additionally, in... Figure 7 In this context, Conventional (WS) represents the TPIM characteristics set by existing adjustment methods, while NewRW (Est) and NewRW (Act) represent the static and dynamic TPIM characteristics set by the adjustment methods of the implementation method, respectively.

[0038] In the above configuration, the estimated accuracy is as follows: Figure 5 As shown, the usable precision in the adjustment can be identified. Using this precision, the search results for settings that meet the required standards are obtained. Figure 6 The error rate is shown. For example... Figure 6 As shown, it can be confirmed that the RW parameter, which differs from existing methods, is the optimal one. As a characteristic at this time, for example, regarding TPIM characteristics, such as... Figure 7 As shown, the results are as expected, confirming that while maintaining the same capacity as the existing system, the deviation in characteristics has been reduced, and there have been further improvements.

[0039] In summary, by using the adjustment method and device of this embodiment, multiple requirements can be met in a balanced and satisfactory manner within a limited test time, and the adjustment time can be shortened.

[0040] Furthermore, embodiments of the present invention have been described; however, these embodiments are merely illustrative and not intended to limit the scope of the invention. These new embodiments can be implemented in various other ways, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included within the scope and spirit of the invention, and are encompassed within the scope of the invention as described in the claims and thereafter.

[0041] Label Explanation

[0042] 100…Neural Network Computer, 101…Database Storage Unit, 200…Main Host, 300…Machine Interface, 400-1 to 400-n…Disk Drive (HDD), 10…Head-Disk Assembly (HDA), 11…Disk, 12…Spindle Motor (SPM), 13…Arm, 14…Load Beam, 15…Head, 16…Voice Coil Motor (VCM), 17…Head Amplifier Integrated Circuit, 20…System-on-Chip (SOC), 21…Microprocessor (CPU), 22…R / W Channel, 23…Disk Controller, 24…Position Controller, 25…Memory, 26…Parameter Storage Unit, 27…Adjustment Log Storage Unit.

Claims

1. A method for adjusting the RW parameters of a disk device, A pre-registration process is performed, in which the characteristics are measured by changing multiple read / write related parameters that determine the capacity / characteristics for each head of the test disk device, and a database of the measurement results is created for each head, registering the characteristics. For the adjustment process of the disk device to be adjusted, the characteristics are determined according to the condition of changing the parameters of each head, the database of heads with similar characteristics is searched from the database of each head, and based on the characteristics of the found database, the appropriate balance setting for capacity / characteristics is made from all hit conditions through the estimation of the neural network computer.

2. The method for adjusting the RW parameters of a disk device according to claim 1, In the pre-registration process, the characteristic is measured extensively, while in the adjustment process, the characteristic is measured only a small number of times.

3. The method for adjusting the RW parameters of a disk device according to claim 1, The neural network computer has an input layer for adjusting values, a first intermediate layer, a second intermediate layer, and an output layer for characteristic values, and can arbitrarily set the number of inputs and outputs.

4. The method for adjusting the RW parameters of a disk device according to claim 1, In the pre-registration process, the registration information in the database is allowed to be modified, and the modification process includes searching the database after the registration information has been modified.

5. The method for adjusting the RW parameters of a disk device according to claim 1, In the adjustment process, an adjustment log representing the execution of the parameter adjustment is stored in the internal memory of the disk device that is the object of the adjustment.

6. A device for adjusting the RW parameters of a disk drive, comprising: Disk devices; A housing that houses multiple disk drives; and In the information processing device, during the pre-registration process, characteristics are measured by changing multiple read / write related parameters that determine capacity / characteristics for each head of the disk drive prepared for testing via the housing device, and a database of the measurement results of the characteristics is created for each head. In the adjustment process, characteristics are measured by changing the parameters for each head of the disk drive prepared via the housing device as the adjustment target, and a database of heads with similar characteristics is searched from the database of each head. Based on the characteristics of the found database, an appropriate balance setting for capacity / characteristics is made from all hit conditions by inferring through a neural network.

7. The RW parameter adjustment device for a disk drive according to claim 6, In the pre-registration process, the characteristic is measured extensively, while in the adjustment process, the characteristic is measured only a small number of times.

8. The RW parameter adjustment device for a disk drive according to claim 6, The information processing device is a neural network computer, which has an input layer for adjusting values, a first intermediate layer, a second intermediate layer, and an output layer for characteristic values, and can arbitrarily set the number of inputs and outputs.

9. The RW parameter adjustment device for a disk drive according to claim 6, In the pre-registration process, the registration information in the database is allowed to be modified, and the modification process includes searching the database after the registration information has been modified.

10. The RW parameter adjustment device for a disk drive according to claim 6, In the adjustment process, an adjustment log representing the execution of the parameter adjustment is stored in the internal memory of the disk device that is the object of the adjustment.

11. The RW parameter adjustment device for a disk drive according to claim 6, The information processing device has an internal storage unit for storing the database.

12. A disk drive, connected as an adjustment target to an RW parameter adjustment device, The RW parameter adjustment device A pre-registration process is performed, in which the characteristics are measured by changing multiple read / write related parameters that determine the capacity / characteristics for each head of the test disk device, and a database of the measurement results of the characteristics is created for each head. For the adjustment process of the disk device to be adjusted, characteristics are determined according to the conditions under which the parameters are changed for each head. A database of heads with similar characteristics is searched from the database of each head. Based on the characteristics of the found database, an appropriate balance setting is made for capacity / characteristics from all hit conditions through estimation by a neural network computer. The disk device includes an internal memory that records an adjustment log indicating that the parameter adjustment was performed through the adjustment process.

Images