JPH02297761A - Magnetic disk device - Google Patents

Abstract

PURPOSE:To eliminate the processing time of a head switching operation and rotational delay, etc., required for an alternate processing and to accelerate alternate processing time by performing all the alternate processings at a semiconductor disk part. CONSTITUTION:The device is equipped with the semiconductor disk part 7 having a semiconductor storage element such as a RAM, etc., a hard magnetic disk part 6, and a control part 4, and when the read and write is performed, a read operation or a write operation for the alternation of the magnetic disk part is performed setting the semiconductor disk part 7 as the one dedicated for alternation. Thereby, since seek time, head switching time, and rotation latency time occupying the most part in a processing at the hard magnetic disk part 6 can be eliminated and the alternate processing can be realized only by performing a read or write processing on the semiconductor disk part 7, the speed of the alternate processing can be accelerated.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a magnetic disk device, and particularly to a large-capacity magnetic disk device.

[Conventional technology]

In a conventional method for replacing defective sectors in magnetic disk drives, a management table is provided in a specific track on a magnetic disk as an alternative management means to manage whether or not alternative processing is to be performed. An alternative method is to allocate sectors to several replacement sectors prepared for each track, and if there are a large number of bad sectors and replacement processing cannot be performed on the same track, allocating them to the excess bad sectors. On the other hand, a method is adopted in which a specific cylinder is used as an alternative cylinder and the track belonging to that cylinder is substituted sector by sector.

Therefore, if there are a large number of bad sectors and replacement processing cannot be performed within the bamboo sector, the head must be switched to the replacement cylinder in order to access the replacement-only sector.
A head movement operation (referred to as a seek operation) must be performed. In addition, it is necessary to read or write to the replacement management table, change the head to the original track, and perform a seek operation in order to continue accessing the next sector after the replacement processing.

[Problem to be solved by the invention]

The above-mentioned replacement processing in conventional magnetic disk drives always includes head switching operations, seek operations, and rotation wait processing when a track with a defective sector is replaced with a track or replacement management table. , This takes a processing time of several ms' to several ms, which has the drawback of reducing the processing speed of the magnetic disk device.

[Means to solve the problem]

The magnetic disk device of the present invention includes a semiconductor disk unit having a semiconductor storage element such as a RAM, a hard magnetic disk unit, and a control unit, and when reading or writing, the magnetic disk unit is used exclusively as a substitute for the semiconductor disk unit. It is designed to perform read and write operations to replace the .

In this way, by performing all alternative processing in the semiconductor disk unit, processing time such as head switching operations, seek operations, and rotation waits that were required for alternative processing is eliminated, and the alternative processing time can be speeded up. This made it possible.

〔Example〕 Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention.

In FIG. 1, 1 is a basic processing unit, 2 is a main memory,
3 is a system bus; 4 is a disk control unit that transmits and receives signals between the basic processing unit 1 and main storage unit 2 and the hard magnetic disk unit 6 and semiconductor disk unit 7, and controls input and output of both disk units; , 7 is a semiconductor disk section.

Next, the operation of the magnetic disk device configured as described above will be explained.

The hard magnetic disk unit 6 and semiconductor disk unit 7 shown in FIG. 1 will be explained using examples of initialization of initial medium defects, alternative processing at the time of reading data, and data saving processing when the semiconductor disk unit 7 is powered off. do.

The hard magnetic disk section 6 shown in FIG. 1 is an example in which one with ESD I specifications is used, and the ID section (meaning the address and status of each sector on the medium) is shown in FIG. As shown in the figure, it has a 5-byte structure. In Fig. 2, the target address is indicated by the contents of 4 bytes of the sector ID field 1a, and the flag field 2a indicates a normal free sector if 00'', a defective sector if 02'', and alternative processing if 03''. sector.

The flowchart shown in FIG. 3 shows the operation for initial medium defects in the hard magnetic disk unit 6 using the semiconductor disk unit 7, and the flowchart shown in FIG. 6 shows an alternative processing operation when reading data from the unit 6 and the semiconductor disk unit 7. The processing operation shown in FIG. 4 is based on the premise that the hard magnetic disk section 6 and the semiconductor disk section 7 have been initialized as shown in FIG.

For driving the hard magnetic disk unit 6 and semiconductor disk unit 7 for reading and writing data, a startup command is issued from the processing device 1 (see FIG. 1) to the disk control unit 4, and the disk control unit 4 It interprets the startup command and applies drive to the hard magnetic disk unit 6 or the semiconductor disk unit 7.

As shown in FIG. 3, first, a data portion is read from the O sector of the head 1 (hereinafter referred to as read data) (reference numeral 11) for the medium defect information cylinder in the hard magnetic disk unit 6, and Recognize which sector has a defect (reference number 12). Create a replacement sector for the next recognized defective sector in the semiconductor disk unit 7 (reference number 13). This process Change the flag byte (information representing the media state of the sector) in the ID part to 03'' (representing the sector that has undergone alternative processing), and write the ID part of the sectors starting from the first sector (hereinafter referred to as write ID). ) are performed sequentially (reference number 14
). This alternative processing is performed by counting up the sectors of the semiconductor disk unit 7 one by one until all recognized defective sectors are counted up (reference numerals 15 and 16). Furthermore, each sector that has undergone this alternative processing is
Set up an alternative management table for the specific cylinder above,
The presence or absence of alternative processing is managed (reference numeral 16).

The format of the management table is as illustrated in FIG. After the replacement with the semiconductor disk unit 7 is completed, replacement processing of the hard magnetic disk unit 6 is performed. This process changes the flag byte of the ID section to 02'' for the recognized defective sector, changes the remaining 4 bytes to an alternative ID, and performs a write ID on the defective sector of the hard magnetic disk section 6 (reference number 17)
. This process continues until the replacement process for all defective sectors is completed (reference numerals 18 and 19). By performing the processing operations shown in FIG. 3 described above, initial medium defects in the hard magnetic disk section 6 and the semiconductor disk section 7 are initialized.

Next, referring to FIG. 4, a description will be given of a read processing operation for both disk portions whose initial defects have been initialized by the processing operation of FIG. 3.

First, write a flag byte to the hard magnetic disk unit 6.
Read data is performed sequentially as ``o'' (reference numeral 21
). From this result, we detect cases in which the target sector cannot be found (hereinafter referred to as no data error), and cases in which the target sector is found but the data section of that sector is defective (hereinafter referred to as data error). do(
Reference number 22). Through this process, sectors that have been replaced and sectors that have become defective due to subsequent defects are detected.

If the process ends normally (reference numeral 23), the process is counted up to the next sector and continues until the final sector. If an error occurs (reference numeral 24), sectors that cannot be saved by retrying are treated as later failures if they are data errors (reference numeral 25).

Read ID for the sector that caused a no-data error
26), if it can be read normally, it is the replaced sector and the read ID is 27).
Valid data is obtained by reading data from the semiconductor disk unit 7 to the replacement sector indicated by the data (reference numeral 28). If the read data is abnormally terminated, the replacement management table of the semiconductor disk unit 7 is used as the read data (reference numeral 29), and the ID of the error sector is compared with the read data to confirm whether or not there is a replacement for the error sector. If no error sector is found in this comparison process (reference numeral 30), it is determined to be a late defect (reference numeral 25)
If found, it is a sector in which both the ID part and the data part are defective, and alternative processing is performed on a track-by-track basis (reference numeral 31).For late defects, this processing routine is exited and separate processing is performed. Although the above description has been made regarding the 0 or more reading process in which alternative processing is performed, the data writing process is also performed in the same way as the process shown in FIG.

Next, the processing operation for saving data before the power is turned off in the semiconductor disk section 7 will be explained with reference to FIG.

First, the flag byte is set to 03'' from the first sector of the semiconductor disk unit 7, and data is sequentially read (reference number 4).
1), For data that has been successfully read, refer to reference numeral 42). Write sequentially from the first sector of the head 2 (the part that reads or writes to the medium surface) of the medium defect information cylinder in the hard magnetic disk unit 6. Data processing is performed (reference numeral 43). Through this process, the data in the semiconductor disk section 7 can be saved to the medium defect information cylinder of the hard magnetic disk section 6.

The above processing is performed while incrementing the sector count by 1 until a no-data error occurs in the semiconductor disk unit 7 (reference numeral 44), and all saving is completed. However, when a no-data error occurs, the error sector ID is performed (reference number 45), and the flag byte is
OO'' (reference number 46), it is not regarded as a normal end and is processed as a subsequent error (reference number 4
7).

FIG. 7 shows a hard magnetic disk section 6 and a semiconductor disk section 7.
FIG. 3 is an operation explanatory diagram showing the access order and evacuation operation of the read operation. In FIG. 7, operations regarding error determination and trial processing are omitted.

In FIG. 7, the read operation starts from the data read/write cylinder 3b on the magnetic disk 1b at the board 1c, and then the read 2c is performed. At this time,
Since sector #2 is defective, read 3c is performed as an alternative process to the semiconductor disk 2b. After the substitution is completed, the program returns to the read 4c and performs a normal read operation of the reads 5c and 6c and the magnetic disk 1b. Next, the semiconductor disk 2b
To save data from the medium defect information cylinder 4b, the alternative data of the cylinder 5b is used as a write operation 7c. Regarding the evacuation operation, the operation shown in FIG. 6 is performed. Furthermore, when resuming operations after evacuation, the data read from the hard magnetic disk unit 6 is written to the semiconductor disk unit 7 in the reverse operation of FIG. By this data recovery operation, the operation shown in FIG. 4 can be performed again.

〔Effect of the invention〕

As explained above, the magnetic disk device of the present invention initializes initial defects by the operation shown in FIG. 3, and performs alternative processing at the time of read/write by the operation shown in FIG. This eliminates seek time, head switching time, and rotational wait time, which are a large burden on processing in the magnetic disk unit, and enables alternative processing to be achieved by only reading or writing to the semiconductor disk unit. This has the effect of increasing processing speed.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing one embodiment of the present invention, FIG. 2 is a format diagram showing the structure of the ID section on the magnetic disk medium of the embodiment of FIG. 1, and FIG. FIG. 4 is a flowchart showing the initialization operation of the initial defective medium replacement process for the hard magnetic disk unit in the embodiment of FIG. 1, and FIG. FIG. 5 is a flowchart showing an alternative sector read processing operation after processing; FIG. 5 is a format diagram showing an alternative management table written to the semiconductor disk unit in the embodiment of FIG. 1;
FIG. 6 is a flowchart showing the processing operation for saving data in the semiconductor disk unit to the initial defect information cylinder of the hard magnetic disk unit in the embodiment shown in FIG. 1, and FIG. FIG.

1...Basic processing unit, 2...Main storage device, 3...
System bus, 4... Disk control unit, 5... Disk control bus, 6... Hard magnetic disk unit, 7...
・Semiconductor disk section.

Claims (1)

[Claims] A semiconductor disk unit having a semiconductor storage element, a hard magnetic disk unit, and a control unit are provided, the semiconductor disk unit is used exclusively as a replacement, and the semiconductor disk unit performs read or write operations for replacing the hard magnetic disk unit. A magnetic disk device characterized in that it is configured to perform the following operations.