KR19980078302A - How to Optimize Runtime Readers for Hard Disk Drives - Google Patents

Abstract

end. TECHNICAL FIELD OF THE INVENTION The invention described in the claims relates to a method of optimizing read factors for read / write channels in a hard disk drive.

I. An object of the present invention is to provide a method for optimizing a runtime readout factor to reliably and efficiently perform data reading by selecting a read factor optimized for a runtime of a hard disk drive.

All. Summary of the Invention Solution of the Invention: Two data sets, a read factor change function set which is a set of read factor functions valid for read error recovery, and an error recovery frequency indicating the frequency at which an error is recovered by performing the read factor change function when a read error occurs. A method of optimizing the execution time read factor of a hard disk drive having an error recovery frequency list that is a set of items having a function, wherein when a data read error occurs, a read factor change function indicated by each item constituting the error recovery frequency list indicates a read error. The process of performing the read factor changing function and attempting to reread the data while changing it sequentially until recovery, and if the error is recovered during data rereading, increases the error recovery frequency of the item and then the error recovery frequency exceeds the preset threshold. The reader of the item It is characterized in that the change function is set to the value of the changed read factor as a new reference read factor and stored in the nonvolatile memory.

la. Significant use of the invention: Can be used to optimize read / write channel readers for hard disk drives.

Description

How to optimize runtime readers for hard disk drives

The present invention relates to a method for optimizing a read factor of a hard disk drive read / write channel circuit, and more particularly, to a method for optimizing a read factor during runtime of a drive.

A hard disk drive is a secondary memory device of a computer system that records and reads digital information by using a head and media. Although a hard disk drive has a relatively small probability of writing and reading due to the characteristics of the head / media, a write / read error generally occurs. The causes of these errors include external causes such as thermal noise, external electrical noise, mechanical disturbances caused by external vibrations or shocks, and uneven application of the magnetic material on the disk media. There are internal causes, such as when distortions occur. Even when the cause of the above error occurs, the error is usually not generated when the magnitude of the noise / disturbance is small or the magnitude of the original read signal is large. In other words, when the signal to noise ratio is high, it has high resistance to the cause of error. On the other hand, in the hard disk drive, the digital information signal is written to the head / media through the read / write channel during recording, and the signal input from the head / media during reading is amplified by the preamplifier and converted into the digital information signal from the read / write channel circuit. Is restored. At this time, the read / write channel circuit sets an optimal reference recording / reading parameter for signal restoration in accordance with the recording / reading characteristics of the head / media. In the case of a hard disk drive, the recording / reading characteristics are generally slightly different due to the characteristic variation of the head / media used in each set. Therefore, in order to compensate for this, an optimal recording / reading factor can be experimentally obtained for each set at the time of manufacture of a hard disk drive. Hereinafter, a conventional method for obtaining an optimal recording / reading factor to be performed in manufacturing will be described.

First, the read factors to be optimized as the first step are determined. Examples of readout factors include cut-off frequency, data threshold, boost level, and the like. Thereafter, as a second step, the read factors are set to various values and the reading experiment is performed. The reading experiment is to investigate the reading error frequency when using the reading factor by adding appropriate disturbance. As a third step, the reading factor representing the smallest error frequency is selected through the reading experiment. At this time, there are various methods for selecting, but the specific selection method is omitted since it is not deeply related to the gist of the present invention. The read factor selected as the fourth step is then stored in non-volatile memory of the hard disk drive. Thereafter, the reader stored in the memory is read out every time the power is applied and used as a reader of the hard disk drive.

In the conventional method described above, the optimum read factor of each set of hard disk drives is determined at the time of manufacture, and thus there is no adaptability according to the use environment. In addition, since the optimal factor selection process applied at the time of manufacture is performed in a relatively short time, a non-optimal value may be set as a read factor depending on the situation at the time of testing. Therefore, a higher error rate can be obtained than when operating the optimized read factor. This increases the probability of problems such as delays in data transmission caused by failure to read data in the event of an error, and may act as a factor of degrading the reliability and performance of the hard disk drive.

Accordingly, an object of the present invention is to provide a method for optimizing a runtime read factor for selecting a read factor optimized for a runtime of a hard disk drive so as to reliably and efficiently perform data reading.

In order to achieve the above object, the present invention provides two types of read factor change functions, which are sets of read factor functions effective for read error recovery, and an error recovery frequency indicating a frequency at which an error is recovered by performing the read factor change function when a read error occurs. A method of optimizing a runtime read factor of a hard disk drive having an error recovery frequency list that is a set of items having data, the method comprising: reading a read factor change function indicated by each item constituting the error recovery frequency list when a data read error occurs; The process of executing the read factor changing function and attempting to reread the data while changing sequentially until the error recovery, and if the error is recovered during the data rereading, increases the error recovery frequency of the item and then resets the corresponding error recovery frequency to a preset threshold. If exceeded, The read factor changing function is characterized in that the value of the changed read factor is set as a new reference read factor and stored in the nonvolatile memory.

1 is a block diagram of a typical hard disk drive.

2 is a readout factor optimization process flow diagram in accordance with one embodiment of the present invention;

Hereinafter, an operation according to an exemplary embodiment of the present invention will be described in detail with reference to the accompanying drawings. In the following description and drawings, numerous specific details are set forth in order to provide a more comprehensive understanding of the invention, such as reader items, reader change functions, and specific processing flows. It will be apparent to those skilled in the art that the present invention may be practiced without these specific details. And detailed descriptions of well-known functions and configurations that may unnecessarily obscure the subject matter of the present invention will be omitted.

1 shows a block diagram of a general hard disk drive, and shows an example of a hard disk drive having two disks 10 and four heads 12 corresponding thereto. Referring to FIG. 1, the disks 10 are typically stacked and mounted on the drive shaft of the spindle motor 34 and each disk surface corresponds to one head 12. The disk 10 typically consists of a plurality of tracks arranged concentrically and includes a parking zone and a data zone where the head 12 can be located when the drive is not in use (powered off). do. The head 12 moves horizontally on the disk 10 by driving a rotary voice coil motor (VCM) 28 and writes or reads data on the disk 10 surface. The preamplifier 16 preamplifies and applies the read signal picked up by one of the heads 12 to the read / write channel circuit 18 when reading data. Encoded write data applied from the read / write channel circuit 18 is driven to drive the corresponding one of the heads 12 to be recorded on the disc 10. At this time, the preamplifier 16 selects one of the heads 12 under the control of a disk data controller (DDC) 36. The read / write channel circuit 18 decodes the read signal applied from the preamplifier 16 to generate read data RDATA. The read / write channel circuit 18 encodes the write data WDATA applied from the DDC 36 and applies it to the preamplifier 16. The read / write channel circuit 18 also demodulates the head position information which is a part of the servo information recorded on the disk 10 to generate a position error signal (PES). The PES generated from the read / write channel circuit 18 is applied to the A / D converter 20, and the A / D converter 20 converts the applied PES into a digital step value corresponding to its level so as to provide a microcontroller ( 22). The DDC 36 records data received from the host computer on the disk 10 through the read / write channel circuit 18 and the preamplifier 16 or transfers data reproduced from the disk 10 to the host computer. . The DDC 36 also interfaces the communication between the host computer and the microcontroller 22.

On the other hand, the microcontroller 22 controls the DDC 36 in response to the data write / read command received from the host computer, and controls the track search and track following. At this time, the micro controller 22 controls the track tracking by using the PES value input from the A / D converter 20 and responds to various servo control signals output from a gate array (not shown). Perform control. The D / A converter 24 converts a control value for position control of the heads 12 generated from the microcontroller 22 into an analog signal and outputs the analog signal. The VCM driver 26 generates a current I (t) for driving the actuator by a signal applied from the D / A converter 24 and applies it to the VCM 28. The VCM 28 located on the other side of the actuator having the heads 12 attached to one side moves the heads 12 horizontally on the disk 10 in response to the current direction and current level input from the VCM driver 26. Let's do it. The motor controller 30 controls the spindle motor driver 32 according to a control value for rotation control of the disks 10 generated from the microcontroller 22. The spindle motor driver 32 drives the spindle motor 34 to rotate the disks 10 under the control of the motor controller 30. The buffer memory 38 connected to the DDC 36 temporarily stores data transferred between the disk 10 and the host computer under the control of the DDC 36. The nonvolatile memory 40 connected to the microcontroller 22 stores a control program and an error recovery frequency list according to an embodiment of the present invention. The error recovery frequency list is shown in Table 1 below. As described above, each of the items is sequentially assigned a reader change function, but in general, the most effective reader change function is assigned first.

On the other hand, prior to describing the read factor optimization process according to an embodiment of the present invention, several terms are defined. First, the reference reader can be changed as the read factor reference value of the hard disk drive, which is the nonvolatile memory 40 of the drive. Are stored in. That is, the reference reader is read from the nonvolatile memory whenever the drive is powered and used as a read factor. On the other hand, the set of reader change functions represents a set of functions for performing a method of changing a read factor effective for error recovery. The error recovery frequency list has the same number of items as the number of elements in the set of read factor change functions, and each item has the following two data. One is a read factor changing function indicating the address of the reader unique change function address, and the other is an error recovery frequency indicating the frequency at which an error is recovered by performing the read factor changing function when a read error occurs.

Hereinafter, an execution time read factor optimization process according to an embodiment of the present invention will be described.

First, the read factor optimization process according to an embodiment of the present invention can be largely divided into an initialization process and a runtime operation as illustrated in FIG. 2. In the above initialization process, the error recovery frequency list sequentially assigns one read factor change function to each item, but generally assigns the most effective read factor change function first. The error recovery frequency list of each item is initialized to zero. In this case, the initialized reference reader and the error recovery frequency list are stored in the nonvolatile memory 40 and may be stored in the disk 2 of the hard disk drive, that is, the maintenance area. In addition, the above-described initialization operation needs to be performed only once when the hard disk drive is manufactured. The reference reader is read from the nonvolatile memory 40 to the microcontroller 22 internal memory every time the hard disk drive is powered on and initialized. Subsequent write / read operations are performed using the read factors stored in the internal memory. Hereinafter, a process of optimizing a read factor during runtime of a hard disk drive will be described.

2 is a flowchart illustrating a read factor optimization process during a drive runtime operation according to an exemplary embodiment of the present invention. Referring to FIG. 2, if a data reading error occurs in step 50 of FIG. 2, the microcontroller 22 proceeds to step 52 and performs a read factor changing function indicated by the first item of the error recovery frequency list. Perform. Upon rereading the data, the microcontroller 22 proceeds to step 54 to check whether the read error is recovered at the same location. If the error is not recovered and the data read error occurs again, the microcontroller 22 proceeds to step 56 to check whether the item being executed is the last item in the error recovery frequency list. If the result is not the last item, the microcontroller 22 proceeds to step 58 to perform the read factor change function indicated by the second item in the error recovery frequency list, and then attempts to reread the data and proceeds to step 54 to recover the read error. Check if it is done. In this way, the microcontroller 22 performs the reread while sequentially changing the read factor changing function until the read error is recovered. If the read error is recovered in step 54, the microcontroller 22 proceeds to step 60 to increase the error recovery frequency of the item and then proceeds to step 62 to check whether the error recovery frequency is greater than or equal to the threshold. If the error recovery frequency is greater than or equal to the threshold value, the microcontroller 22 proceeds to step 64 and newly sets the value of the read factor changed by the read factor change function of the corresponding item as a reference readout and stores it in the nonvolatile memory 40. do. In step 66, the error recovery frequency is initialized. On the other hand, if the error is not recovered even after the read factor change function indicated by all items on the error recovery frequency list is executed in step 56, the microcontroller 22 reports an unrecoverable error to the host computer, and then the embodiment of the present invention. The method for automatically changing the read factor according to the method is finished.

An embodiment will be described with reference to the read factor changing method table in Table 1 below to specify the runtime read factor optimization process as described above.

Reader Factor Reader change function One Set Lead Offtrack to 5% 2 Set Lead Offtrack to -5% 3 Set Lead Offtrack to 10% 4 Set Lead Offtrack to -10% 5 Set Lead Offtrack to 15% 6 Set Lead Offtrack to -15% 7 5% increase in data cut-off frequency 8 5% downward adjustment of data cut-off frequency 9 10% increase in data cut-off frequency 10 Adjust data cut-off frequency down 10% 11 5% increase in data threshold 12 5% lower data threshold 13 Raise the data threshold by 10% 14 10% lower data threshold 15 Increase boost level by 3% 16 Lower boost level by 3%

First, it is assumed that the reference reading factor value is as follows. In this case, an example in which the data threshold value is set lower than the optimum value and the most effective error recovery method is the upward adjustment of the data threshold value will be given.

Data cut-off frequency: 0x4a

Data Threshold: 0x60,

Boost level: 0x37

When the first read error occurs, the read factor changing function is sequentially executed from No. 1 in Table 1 above. If it is assumed that an error has been recovered in a retry attempt after performing the eleventh item of Table 1, which is the 5% increase in the data threshold, the error recovery frequency of this item is increased to 1. After changing the reference reading factor, the data reading operation is continued. As a result of the read error occurrence and recovery during several read operations, the error recovery frequency corresponding to the data threshold 5% upward adjustment function exceeds the given threshold. At this time, it is regarded that the reading error occurred because the data threshold value is inappropriate among the reference reading factors, and the value of 5% of the standard reading factor is increased as the new reference reading factor. The new reader is stored in nonvolatile memory 40. Therefore, the value of the data threshold is optimized and the error occurrence caused by the inappropriate value of the data threshold is reduced by using a new reference reading factor. In the subsequent error occurrence recovery, the next one of several read factors is optimized to have a large influence on the occurrence of the error. Thus, after a long run-time operation, the reference reader can be adaptively optimized for a specific set of hard disk drives.

As described above, the present invention converges on any particular set of hard disk drives to a reference reader that is optimal for that particular set and the environment in which that particular set is used. This operation is adaptively performed at runtime. Therefore, there is an advantage that can minimize problems such as data transmission delay when an error occurs in the prior art.

Claims (3)

An error that is a set of items having two data types, a set of read factor change functions that are sets of read function functions effective for reading error recovery, and an error recovery frequency indicating the frequency of error recovery by performing the read factor change function when a read error occurs. In the method of optimizing the runtime reader of a hard disk drive having a recovery frequency list, Performing a read factor change function and attempting to reread the data while sequentially changing the read factor change function indicated by the items constituting the error recovery frequency list when the data read error occurs until recovery of the read error; If the error is recovered during data reread, increase the error recovery frequency of the item, and if the error recovery frequency exceeds the preset threshold, set the value of the reader changed by the change factor of the item as a new reference reader. The method of optimizing the execution time readout, characterized in that consisting of a process of storing in a nonvolatile memory. The method of claim 1, wherein the items constituting the error recovery frequency list are allocated in the order of a read factor change function having a high data read error recovery rate. The method of claim 1, wherein the new reference reading factor is recorded in the maintenance area of the disk and read and used every time the power is applied.