JPH03259460A - Disk control system - Google Patents

Abstract

PURPOSE:To attain a high-speed processing by including a unused block of a spare area into a skew value to set the head address of an adjacent cylinder or an adjacent track. CONSTITUTION:Plural cylinders (1-1)-(1-m) or tracks of a disk on which a data is recorded are split into plural blocks 2 respectively. Then a defect block corresponding to the cylinders (1-1)-(1-m) or tracks is detected and the block of a spare area 3 is allocated as an alternate block with respect to the defect block. Moreover, a value subtracting the unused block number of the spare area 3 from the block number of a cylinder skew is used as a new cylinder skew, and the head address of an adjacent cylinder or track is obtained and set from the address of a final block in the operating block of the cylinders (1-1)-(1-m) or tracks. Thus, the unused area of the spare area 3 is effectively used to shorten the skew time.

Description

[Detailed Description of the Invention] [Summary] Regarding a disk control method for speeding up cylinder skew, the purpose is to effectively utilize unused blocks in a spare area to speed up cylinder skew, and to write or write data. A plurality of cylinders or a plurality of tracks of a disk to be read are each divided into a plurality of blocks, an arbitrary number of blocks are set as spare areas corresponding to the cylinders or tracks, a defective block corresponding to the cylinder or track is detected, and a defective block corresponding to the cylinder or track is detected. A block in the spare area is allocated as a replacement for the defective block, and the value obtained by subtracting the number of unused blocks in the spare area from the number of blocks in the cylinder skew value is used as the new cylinder skew value. The configuration is such that the start address of the adjacent cylinder or trunk is determined and set from the address of the last block in the block.

[Industrial application field]

The present invention relates to a disk control method for speeding up cylinder skew.

When recording continuous large-capacity data such as image data on a disk device such as a magnetic disk device or an optical disk device,
Since the capacity of one cylinder or track is insufficient, recording is performed over multiple cylinders or multiple tracks. In such a case, a time is required to move a head such as a magnetic head or an optical head to an adjacent cylinder or an adjacent track, that is, a skew time is required.

Furthermore, if the head address of each cylinder or track is set on the same radius of the disk, a head moved to an adjacent cylinder or track will need rotational waiting time until it reaches the head track. . Therefore, it has been proposed to set the start address in accordance with the skew time.

Furthermore, even if there is a defective area due to a structural defect on the disk, a means is adopted to form a spare area to replace the defective area, and this spare area is formed to correspond to a cylinder or a track. .

[Conventional technology]

A spare area is formed within a track or cylinder in order to repair a defective area on a disk of a recording medium, and in FIG. Since the disk is divided, a plurality of spare blocks SP are formed therein, and the spare block SP is allocated in place of the defective block in the track, so that the disk including the defective block can be used. In addition, in Fig. 5, the head HDI to HD
A plurality of spare blocks SP are formed corresponding to the cylinder accessed by n, and the spare block SP is allocated in place of a bad block in the cylinder, so even if there is a cylinder containing a bad block, the disk can still be used. can do.

In order to reduce the rotation waiting time on a disk in which such spare blocks SP are formed, as shown in FIG. The time required to perform this, that is, the skew value (for example, 4 blocks) is set as the start address of the next cylinder. For example, even if the spare area is all used, the starting address (0) of the next cylinder starts from the last block of this spare area.
) The cylinder skew can be performed so that the rotation waiting time becomes almost zero.

By setting the start address of each cylinder as described above, the rotation waiting time can be reduced compared to the case where the start address of each cylinder is set on the same radius.

[Problems to be solved by the invention] With the increase in the capacity of disk devices, there is a demand for high-speed data processing, and attempts have been made to shorten the seek time of the head, but the problem is close to its limit. It has become. Also, by configuring the start address of the cylinder to be shifted sequentially according to the skew value rather than on the same radius, the rotation waiting time can be shortened, but if there are no defective blocks, the spare area will be completely empty. The block becomes a used block, and even in that case, the spare area is wasted and reserved for skew.

The present invention aims to speed up cylinder skew by effectively utilizing unused blocks in a spare area.

[Means to solve the problem]

The disk control method of the present invention aims to increase speed by including unused blocks in the spare area in the skew value and setting the start addresses of adjacent cylinders or adjacent tracks. explain.

Multiple cylinders 11 to 1- of a disk for recording data
Divide m or a plurality of tracks into a plurality of blocks 2, respectively, and set an arbitrary number of blocks SPI to SP4 as spare areas 3 corresponding to cylinders 1-1 to 1-m or tracks, and make them corresponding to cylinders 1-1 to 1m or tracks. Detects a defective block, allocates a block in spare area 3 to this defective block as a replacement, and subtracts the number of unused blocks in spare area 3 from the number of blocks in the cylinder skew value to determine the new cylinder skew. As a value, the start address of the adjacent cylinder or adjacent track is determined and set from the address of the last block among the used blocks of cylinders 1-1 to 1-m or the track.

[Effect]

When the spare area 3 corresponding to cylinders 1-1 to 1-m or tracks is set to 4 blocks SPI to SP4, and the skew value is set to 4 blocks, for example, if there is one defective block in cylinder 1-1, the spare area 3 Area block SP
Since I is allocated as a replacement, the number of unused blocks in spare area 3 is 3 blocks, and the skew value is 4.
When three unused blocks are subtracted from the block, it becomes one block, and the start address O of the next cylinder 1-2 becomes the block position obtained by adding 1 to the start address 0 of the previous cylinder 1-1.

Similarly, if there are two defective blocks in cylinder 1-2 and blocks SPI and SP2 in spare area 3 are allocated to replace the defective blocks, the number of unused blocks becomes two. Therefore, the start address O of the next cylinder 1-3 is a block position that is +2 to the start address O of the previous cylinder 1-2.

Therefore, by effectively utilizing the unused area of spare area 3,
Skew time can be reduced.

〔Example〕

Embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 2 is a block diagram of main parts of an embodiment of the present invention, and 11
．． 12 is an interface section, 13 is a register section, 14 is a data transfer processing section, 15 is a control storage section, 16 is a microprogram section, and 17 and 18 are tables A and B.

The interface section 11 is connected to, for example, an interface section of a host computer via a path or the like, and the interface section 12 is connected to an interface section of a drive section of a disk device. Further, in the control storage section 15, a microprogram section 16 DEG tables A, B 17, 18, etc. are formed.

According to the program stored in the microprogram unit 16, controls such as forming a disk format, detecting a defective block, and setting a start address are performed.

FIG. 3 is a flowchart of an embodiment of the present invention, showing the case where the first block address of the next cylinder is set by steps (1) to 03). First, defective information is merged (1), and defective information is sorted. (2). That is, defect information indicating defective blocks corresponding to cylinders or tracks of the disk is read from the disk and transferred to the data transfer processing section 14 via the interface section 12. Further, if there is disk defect information held on the host computer side, it is transferred to the data transfer processing section 4 via the interface section 11.

Collect disk defect information and calculate the number of defective blocks for each cylinder 3), write the number of defective blocks in the cylinder corresponding area of table A (4), and when the writing is completed, set the cylinder number CY L No. 5), the number of bad blocks is read from the area of cylinder number CY L No, O in table A and is sent to table B (
6). Then, by adding the sum of the skew values to the cylinder final block address, the first block address of the next cylinder is determined (7).

Then, the number of bad blocks BB set in table B
and the number of spare blocks SPB are compared (8), and if the number of bad blocks BB is less than the number of spare blocks SPB,
The number of defective blocks BB is subtracted from the number of spare blocks SPB to obtain the number of unused blocks DB (9), and the number of unused blocks DB is calculated from the first block address BA of the next cylinder obtained in step (7). A new starting block address BA is obtained by subtracting 0ω, the starting block address BA is written to table B QD, and the number of bad blocks BB is calculated as the number of spare blocks SP.
If the number is the same as B or more, the number of spare blots SP
Since all blocks B are used, the number of unused blocks DB becomes zero, and the first block address BA of the family cylinder obtained in step (7) is set in table B (
l I).

Then, it is determined whether the cylinder number CYLNo is the maximum value (max) of the user area (+2), and if the maximum value has not been reached, the cylinder number CYLNα is set to +ILO3) and a new cylinder number CY According to L No, the number of defective blocks is read from table A and set in table B (6), and the steps (7) to
03) is repeated.

When the cylinder number CY L No reaches its maximum value, the disk is formatted based on the first block address of each cylinder in the contents of table B.

Therefore, although the first block address of each cylinder is variable, the rotation waiting time when moving to an adjacent cylinder can be brought close to zero, and the operating speed can be improved.

Although the above-described flowchart shows a case where a spare area is provided corresponding to a cylinder, it can of course be applied to a case where a spare area is provided correspondingly to a track.

If each track on the disk mentioned above is divided into 53 blocks (sectors), and one block is 512 hides, and the time to go around the track is approximately 18 ms, the processing time per block is 18÷53=0. 3 (ms).

Also, there are 3 spare blocks for each cylinder,
Assuming that the number of defective blocks is 50 and the total number of cylinders is 1600, the number of unused blocks in the spare area is 1600x3
-50=4750.

The processing time per l block (
0.3m5), 4750X0.3=140
0 (ms:l).According to the embodiment of the present invention described above,
Since this unused block is used for cylinder skew, when reading the entire disk, depending on the distribution of the number of bad blocks, the above-mentioned 1.43 time can be saved. This makes it possible to improve processing speed.

Also, the block length is made longer than in the above case, for example, 1
Assuming that the block is 4096 bytes, the track is divided into 7 blocks (sectors), and the time it takes to go around one trunk is 18 ms as in the previous case, the processing time per block is 18÷7#2. 6 (ms).

Also, if each cylinder has one spare block, the total number of cylinders is 1600, and the number of defective blocks is 50, the number of unused blocks in the spare area is 1600xl-
50=1550. Multiplying this by the processing time per block (2.6m5) yields 1550X2.6ξ
4 [S], and it is possible to save this amount of time. Therefore, as the block length increases, the degree of reduction in processing time increases.

[Effects of the Invention] As explained above, the present invention detects a defective block corresponding to the cylinders 1-1 to 1-m or tracks of a disk as a recording medium, and stores the spare area 3 for this defective block. A block is allocated as a replacement, and the value obtained by subtracting the number of unused blocks in this spare area 3 from the number of blocks in the cylinder skew value is set as the new cylinder skew value among the used blocks of cylinders 1-1 to 1-m or the track. This method calculates the start address of an adjacent cylinder or adjacent track from the address of the last block of It is possible to bring the value close to zero, which has the advantage of shortening the processing time when executing an access command to the disk.

[Brief explanation of drawings]

FIG. 1 is an explanatory diagram of the principle of the present invention, FIG. 2 is a block diagram of main parts of an embodiment of the present invention, FIG. 3 is a flowchart of an embodiment of the present invention, and FIG. 4 is an explanatory diagram of a spare area in a truck. Fifth
The figure is an explanatory diagram of a spare area within a cylinder, and FIG. 6 is an explanatory diagram of a skew in a conventional example. l-1-1-m is a cylinder, 2 is a cylinder, 3 is a spare area, and SPI to SP4 are spare broncs.

Claims (1)

[Claims] A plurality of cylinders of a disk (to which data is written or read)
1-1 to 1-m) or a plurality of tracks are each divided into a plurality of blocks (2), and the cylinders (1-1 to 1-m) are divided into a plurality of blocks (2).
m) Set an arbitrary number of blocks as spare areas (3) corresponding to the cylinders (1-1 to 1-m) or corresponding to the tracks, detect defective blocks corresponding to the cylinders (1-1 to 1-m) or corresponding to the tracks, and set the spare area for the defective blocks. The block in (3) is allocated as a replacement, and the value obtained by subtracting the number of unused blocks in the spare area (3) from the number of blocks in the cylinder skew value is set as the new cylinder skew value.
-1 to 1-m) or the address of the last block among the used blocks of the track to determine and set the leading address of the adjacent cylinder or adjacent track.