JPH11317008A - Disk storage device and segment cache control method for the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce the moving amount of a head and the number of disk accessing times for random accessing to all the areas of a disk. SOLUTION: When the write commands of discontinuous addresses having small block sizes are continuously issued, transfer data is stored temporarily in a segment cache buffer 240 capable of being divided into a plurality of segments, a segment management block 222 for classifying and managing those of bits of address information indicating disk writing sides set in positional relations on a disk medium for continuous disk writing as the same group is generated and registered for each segment in a segment management table 220 and, when writing conditions are established, en-block disk writing is performed in accordance with the segment management block 222 belonging to a proper group.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a disk storage device having a buffer memory for temporarily storing data transferred between a host and a disk medium, and more particularly, to a buffer area of the buffer memory divided into a plurality of segments. The present invention relates to a disk storage device suitable for efficiently performing writing to a disk medium by dividing the data and performing ranking management according to addresses, and a segment cache control method in the device.

[0002]

2. Description of the Related Art A disk storage device of recent years generally includes a buffer memory for temporarily storing data transferred between a host (a host system such as a host computer) and a disk medium.

Japanese Patent Application Laid-Open No. 8-328747 discloses a technique for efficiently managing a buffer memory (hereinafter referred to as prior art). In the prior art described in the above publication, as shown in FIG. 7, for example, a segment B for burst transfer and a segment A for both reading and writing are secured in (a cache buffer area of) the buffer memory 71, In each case, the size and start address are changed according to the timing at the time of data transfer.

In the disk storage device described in the above publication having such a buffer memory 71, when a write command instructing, for example, 16-block disk writing is received from the host 72, the data is continuously stored in the buffer memory 71 without delay. Transferred. The transferred data is stored in the burst transfer segment B on the buffer memory 71. Here, it is assumed that the size of the segment B is 16 blocks (16 sector size), the size of the segment A is 20 blocks (20 sector size), and the size of one sector is 512 bytes.

In the example of FIG. 7, the data transferred from the host 72 is the start address a1 of the segment B indicated by the HP (host pointer), for example, 10000H (the last H is 1).
(Indicating hexadecimal representation).
Since the size of the transfer data is 16 blocks, 100
Data is stored from 00H to 11FFFH, which is the end of segment B. Here, it is assumed that the next write command has been issued following the previous write command. At this time, the HP is the start address a of the segment B.
However, by changing this to indicate the start address a2 of the next address, that is, the start address a2 of the segment A, that is, 12000H, the data sent from the host without delay in response to the next write command is emptied. Segment A. The host 72 intends to transfer data to the segment B, and is actually transferring data to the segment A.

When the data is transferred for 16 blocks, the segment B becomes full when viewed from the host 72 side, which means that the address a2 has been reached. Start address a1. As described above, data transfer is continuously performed while alternately setting the start address in the HP such as segment B → segment A → segment B →.

While the data is transferred from the host to the segments A and B, the data stored in the segment B is sequentially taken out from the start address a3 of the segment B indicated by the DP (disk pointer) and written to the disk 73. It is. When the data of the segment B becomes empty, the address a4 of the segment A is set in the DP, and the data stored in the segment A is written to the disk 73 this time. Hereinafter, as for the DP, similarly to the HP, the segment B → the segment A → the segment B →.
By repeating the switching setting as described above, data can be stored in the disk 73 without delaying data transfer.

As described above, according to the prior art, when a write command is issued sequentially from a host, that is, when a plurality of write blocks are continuously transmitted as in a write multiple command, data is continuously transmitted without delay. This is a technique suitable for sequential writing for the purpose of writing data to a disc.

[0009]

In the prior art described above, when a plurality of write blocks are continuously transmitted as in a write multiple command, that is, in the case of a sequential write, a burst transfer segment on a buffer memory is used. It is certainly suitable for efficiently writing data to a disk medium without delay using a read / write segment.

However, in the case where the transmitted data is a small size such as one block or two blocks and the write destination address is not continuous, that is, in the case of random write, every time a write command is issued from the host. Since writing to the disk is performed, a large amount of overhead for disk access such as head seek operation and disk rotation waiting time is generated. This is a problem that occurs when a random write command requesting access to discontinuous addresses frequently occurs.

The present invention has been made in consideration of the above circumstances, and has as its object that when data transferred from a host has a small block size, the data is held in divided segments and stored in physically continuous sectors to some extent. By performing disk access when the device is accessible,
It is an object of the present invention to provide a disk storage device capable of reducing the overhead by reducing the moving amount of the head and the number of disk accesses, and improving the speed of random write, and a segment cache control method in the device.

[0012]

SUMMARY OF THE INVENTION The present invention relates to a disk storage device having a buffer memory for temporarily storing data transferred between a disk medium on which recording and reproduction are performed by a head and a host. A segment cache buffer secured in a memory, which can be divided into a plurality of segments, and a segment management block generated for each segment in the segment cache buffer, and a disk medium for data stored in the segment An address field in which address information indicating the above write destination is set, a segment length field in which a segment length indicating the size of the segment is set, and address information in a positional relationship in which the head can be moved within a predetermined time. Information to classify A segment management block having a classification field for each adjacent address to be managed, a segment management table storage means for managing the segment cache buffer, and a host issuing a write command designating disk writing and a disk writing target. If the data size is less than or equal to a predetermined size, the data is stored in a segment of the same size allocated in the segment cache buffer, and a segment management block for managing the segment is stored in the segment cache buffer. The segment cache control means generated in the segment management table, and the segment management blocks generated in the segment management table and classified into the same group by the neighborhood address classification field, Based on the segment management block of the group meets the eye was writing condition, characterized by comprising a disk write control means for performing a disk write in groups.

That is, in the present invention, when write commands of discontinuous addresses of small block size are successively issued, that is, at the time of random write, the transfer data is stored in the segment cache buffer which can be divided into a plurality of segments. Segment management for temporarily storing the address information indicating the disk writing destination and classifying and managing those having a positional relationship on the disk medium such that continuous disk writing can be performed as the same group. Blocks are generated and registered in the segment management table for each segment, and when writing conditions are satisfied, disk writing is performed collectively according to the segment management blocks belonging to the corresponding group.

As described above, according to the present invention, when a write command of a discontinuous address having a small block size is continuously issued, the data write timing is adjusted using the segment management table, and the continuous write command is issued. Since the disk writing is performed in units of a group when data can be written in the group, the number of disk accesses can be reduced and overhead can be reduced.

In the present invention, a physical positional relationship such that a head moving operation can be performed within a predetermined time, that is, a position where a disk access (head moving operation) can be optimally performed, instead of grouping in order of addresses. Since the related addresses are classified into the same group and the disk writing is performed in a group unit, overhead due to seek and skew can be reduced.

Here, the condition that the head moving operation for classifying as the same group can be performed within a predetermined time is that the write destination address is on the same track, on an adjacent track (within the same subzone), or on an adjacent track across the subzone. It is advisable to apply a physical positional relationship such as above. In addition, it is also possible to apply a physical positional relationship in which the disk rotation waiting time is kept within a predetermined time for grouping. By such grouping, overhead such as head transfer and cylinder transfer can be reduced.

Since a physical positional relationship such that a head movement operation can be performed within a predetermined time is used as a condition for classifying the same group, even continuous addresses satisfy this condition. By excluding those that do not belong to the same group, it is possible to prevent overhead from occurring when performing disk writing collectively for each group.

A disk which collectively performs the above-mentioned write conditions in a group unit by judging that the number of the segment management blocks classified into the same group in the classification field for each adjacent address exceeds a predetermined number. Writing can be performed efficiently.

In addition, the satisfaction of the above-mentioned writing condition is
By determining that the number of segment divisions in the segment cache buffer has reached a predetermined number,
It is possible to prevent an increase in the number of segment divisions and an overflow of the segment management table and the cache segment buffer.

In addition, the time elapsed since the generation of the most recently generated segment management block in the group is examined for each group classified in the classification field for each neighboring address, and the time exceeds a predetermined time. With that,
By determining that the write condition of the corresponding group is satisfied, the occurrence of an area that is not accessed at all depending on the grouping is determined by the LRU method (Least Recently
It can be prevented by applying a sed rule) method.

In addition, when it is detected that there is a group including a segment management block having address information having a positional relationship with which the head moving operation can be performed within a predetermined time from the current head position, writing is performed for the group. By determining that the condition is satisfied, it is possible to minimize the seek operation of the head. Note that a conventionally known sequential write may be applied to a write command of a continuous address that is not a small block size.

[0022]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below with reference to the drawings, taking an example in which the present invention is applied to a magnetic disk drive. FIG. 1 is a block diagram showing the overall configuration of a magnetic disk drive according to one embodiment of the present invention.

The configuration shown in FIG. 1 assumes a magnetic disk device in which two disks (magnetic recording media) 11-0 and 11-1 are stacked. A large number of concentric tracks are formed on both surfaces (both recording surfaces) of the disk 11-i (i = 0, 1). In the disk 11-i, the track (cylinder)
In order to increase the format efficiency of the disk 11-i by effectively using the area on the outer peripheral side of the disk where the physical circumference of the disk 11 becomes longer, as shown in FIG.
1-i recording surface in radial direction (consisting of multiple tracks)
A configuration in which the data is divided into a plurality of zones Z and the number of data sectors per cylinder (track) is different for each zone Z (the number of data sectors per outer zone becomes larger), that is, the data transfer speed (transfer rate) is different (the outer transfer zone) A CDR (Constant Density Recording) format is applied. In the zone Z, data is recorded at a constant recording density along the rotation direction of the disk 11-i.

In each track of the disk 11-i, servo areas SV in which servo information used for head seek / positioning and the like are recorded are arranged at equal intervals. A user area is provided between the servo areas SV, and a plurality of (generally 2 to 5) sectors (data sectors) as recording units are arranged in the user area. The servo areas SV are radially arranged on the disk 11-i radially across the tracks from the center.

The zone Z is further divided into a plurality of sub-zones SZ. Here, zone Z has three sub-zones S
Z. Each sub-zone SZ has, for example, three tracks as shown in FIG. Information is written in the data sector on the track along the track in the rotation direction of the disk 11-i.

In the same sub-zone SZ, logical addresses are continuous. Taking the sub-zone SZ on the outer peripheral side of the disks 11-0 and 11-1 shown in FIG. 3 as an example, the logical address is
Tracks a, from the outer circumference of the head 0 of the disk 11-0,
b, c, tracks d, e, f from the inner circumference of the head 1 of the same disk 11-0, tracks g, h, i from the outer circumference of the head 0 of the disk 11-1, and the head of the disk 11-1 Tracks j, k, and l from the inner peripheral side of the track 1 (in alphabetical order). A track m which is physically adjacent to the track c (in the inner circumferential direction) and which is located on the same recording surface (head 0) of the disk 11-0 as the track c on another subzone SZ is The logical address is not continuous with the track c but is continuous with the track 1 on the head 3 of the disk 11-1.

The technique of subdividing the zone Z (that is, the sub-zone SZ) and assigning the logical address is based on a limited address range (the same sub-zone SZ) on the disk (the disks 11-0 and 11-1 in this case). This is a conventional technique that has been proposed for optimizing a head moving operation or the like in random access within a device.

In this prior art, for example, when performing full sequential write on the disks 11-0 and 11-1 of FIG. 3, first, the data sector on the track a of the head 0 is targeted for the outermost sub-zone SZ. Data is written to Next, data is written to an adjacent track, that is, a data sector on track b. Hereinafter, for the same sub-zone SZ, c, d, e, f, g, h,
Data is written in alphabetical order, such as i, j, k, l. When the data writing to the data sector on the track 1 is completed, the data is written in the alphabetical order m, n, o, p... For another subzone SZ adjacent on the inner circumference side. That is, when writing of data to the data sector on the track arranged in the sub-zone SZ of one disk 11-i is completed, the data sector is adjacent in the radial direction (inner circumferential direction) on the recording surface of the same disk 11-i. Without moving the head to another sub-zone SZ, it moves in the head direction (perpendicular to the disk surface) and moves to the next head, and from that position on the track arranged in the same sub-zone SZ toward the outer periphery. Data is written to the data sector.

The same disk write operation is performed when a write command is continuously issued from the host and the addresses (write destination disk addresses) are continuous. That is, when a write command is issued from the host, the corresponding data is temporarily stored in the buffer memory 24. Thereafter, a plurality of write commands are successively issued, and their addresses are a, b, c, d, e,.
Is continuous (sequential), such as
After data transferred from the host for each command accumulates in the buffer memory 24, disk writing is performed continuously. As described above, when the write command is issued continuously from the host and the addresses are continuous, the disk writing is continuously performed after the data is accumulated in the buffer memory 24, so that the disk access occurs. The overhead involved is not a problem.

However, even if the host continuously issues a write command, if the address (the write destination disk address) is not continuous, the buffer memory (unlike the case where the addresses are continuous) is stored in the buffer memory. Since data is written in the disk 24 each time a write command is issued without storing data in the disk 24, that is, each time a write command is issued, overhead occurs. Hereinafter, the overhead at the time of accessing the disk will be described.

When accessing the disk 11-i, in order to access distant data sectors on the same track,
Waiting for rotation of the disk 11-i occurs. In addition, track seek and track skew (position shift) occur when accessing a distant track, and even when head switching is performed, head skew occurs. When accessing a non-sequential address, that is, when accessing a random address, an enormous overhead due to the seek and skew described above occurs.

The prior art to which the sub-zone SZ is applied is intended to improve the performance at the time of such random access. In order to shorten the seek time, the access direction within the sub-zone SZ is reduced as described above. In the head direction (head 0 → head 1 → head 2 → head 3)
As a result, the average seek distance in the cylinder direction (radial direction from the outer circumference to the inner circumference) is reduced. However, even in this prior art, when the head movement between the sub-zones SZ and the head switching are required at the same time (m → k, c → q), for example, when the addresses are m, k, c, q. ,
Frequent seek operation and skew of the head 12 and the track occur, so that the head movement amount and the disk rotation waiting time increase, and the overhead increases.

As described above, the prior art to which the sub-zone SZ is applied is an optimization technique using a logical address for a head moving operation or the like in random access within a limited address range (the same sub-zone). No consideration is given to the random access that is performed, that is, random access to the entire disk medium, and the overhead cannot be reduced. Even in the same sub-zone SV, a random access, for example, a data sector on the track a, a data sector on the track d, a data sector on the track a, and then a write to specify a disk write on the disk a If commands are issued continuously, the disk is written each time a write command is issued.
There is not much overhead. Also, the head movement amount and the disk rotation waiting time are not necessarily short.

Therefore, in this embodiment, as described later,
A temporary data holding operation in units of a write command to the segment cache buffer 240 and a disk writing operation performed when access to a physically continuous sector is possible to some extent using the segment management table 220 is performed. Memory 24 Disk 11
For random access of the entire area of -0, 11-1, optimization is performed by using physical addresses that can reduce the amount of head movement and the number of disk accesses.

Referring again to FIG. 1, each disk 11-i
Each of the recording surfaces has a head (magnetic head) 12
Are provided facing the recording surface of the disk 11-i. Each head 12 is attached to a carriage (head moving mechanism) 13 as a rotary type actuator, and the disk 11-i is rotated according to the rotation of the carriage 13.
Move in the radial direction. Thus, the head 12 is sought and positioned on the target track. The disk 11-i is a spindle motor (SPM) 14
It rotates faster. The carriage 13 is driven by a voice coil motor (VCM) 15.

The spindle motor 14 and the voice coil motor 15 are provided with a motor driver circuit (motor driver I).
C) 16. Motor driver circuit 16
Drives the motor 14 by supplying a control current to the spindle motor 14 and drives the motor 15 by supplying a control current to the voice coil motor 15. The value of the control current (control amount) is determined by the calculation process of the CPU 20.

The head 12 seeks on the target track.
After the positioning, the rotation of the disk 11-i causes
Scan on that track. The head 12 sequentially reads the servo information of the servo areas SV arranged at equal intervals on the track by scanning. Also the head 12
Reads and writes data from and to a target data sector by scanning.

Each head 12 has a head amplifier circuit (head I) mounted on a flexible printed circuit board (FPC).
C) 17 is connected. The head amplifier circuit 17
It controls switching of the head 12 (according to control from the CPU 20), input / output of read / write signals to / from the head 12, and the like. The head amplifier circuit 17 amplifies an analog output (read signal of the head 12) read by the head 12, and a R / W (read / write) circuit (read / write).
The write data sent from the write IC (18) is subjected to predetermined signal processing and sent to the head (12).

The R / W circuit 18 is an AGC (automatic gain) for amplifying an analog output (read signal of the head 12) read from the disk 11-i by the head 12 and amplified by the head amplifier circuit 17 to a constant voltage. Control) function and a read signal amplified by the AGC function
A decoding function (read channel) for performing signal processing required for reproducing RZ code data, an encoding function (write channel) for performing signal processing required for recording data on the disk 11, and a servo from the read signal. A pulse function for pulsing the read signal and outputting it as pulsed read data to enable information extraction.

The servo circuit 19 extracts a cylinder code and the like in the servo information (necessary for seeking the head 12) from the read pulse output from the R / W circuit 18 and also outputs the same from the R / W circuit 18. The burst data (position error data) in the servo information (necessary for the positioning control of the head 12) is extracted from the read signal (amplified by the AGC function) and output to the CPU 20.

CPU (Central Processing Unit) 20
Is a control of each unit of the magnetic disk drive according to a control program stored in a ROM (Read Only Memory) 21, for example, a seek / position control of the head 12 based on a cylinder code and burst data extracted by the servo circuit 19, The read / write control is performed by the HDC 23 according to the read / write command.

In addition to the ROM 21, the CPU 20 has a CP
RAM (Random Acces) that provides a work area for U20
s Memory) 22 and HDC (disk controller)
23 are connected. The RAM 22 holds a management table (segment management table) 220 for the (segment cache buffer 240 secured in the buffer memory 24 described later).

The HDC (disk controller) 23
Protocol processing for command and data communication with a host connected via a host I / F (interface) 25, access control to the buffer memory 24, and access to the disk 11-i through the R / W circuit 18. It controls read / write control and the like. In HDC23,
For example, a buffer memory 24 constituted by a RAM is connected.

A part of the buffer memory 24 has a segment cache buffer 240 (for write data) for temporarily storing write data transferred from the host.
(Area of). The segment cache buffer 240 is used by being divided into a plurality of variable-length segments. Each segment is generated for each command from the host device, and is managed by a segment management table 220 provided in the RAM 22. In FIG. 1, a segment cache (for read data) for temporarily storing read data read from the disk 11-i is omitted. Further, the buffer memory 24 may be built in the HDC 23.

The host I / F 25 is connected to the host (host system) and the HDC 23, and receives commands (such as read commands and write commands) and write data from the host and transmits read data to the host. .

FIG. 4 shows the segment cache buffer 2
4 shows an example of segment division in 40 and an example of a data structure of a segment management table 220 for managing each segment in the cache buffer 240.

In the figure, the segment cache buffer 2
At 40, a state is shown in which six segments are secured (in response to six write commands instructing writing of a disk with a transfer block length equal to or less than the threshold value from the host, that is, a small block length).

The segment management table 220 is composed of a setting field (segment division number field) 221 for the number of segments (segment division number) forming a header portion of the table 220, and a segment management block 222 generated for each segment. Is done. The content of the segment division number field 221 is updated (incremented by one) each time the segment management block 222 is generated.

The segment management block 222 receives an instruction from the host (instructs writing to a disk having a small block length).
It is generated by the CPU 20 in response to the write command, and is used to set a write start address setting field (start address field) 222a of the write data stored in the corresponding segment on the disk 11, and a size (segment length) of the segment. A setting field (segment length field) 222b and a nearby address classification field 222c are included. The start address may be a physical address or a logical CHS (Cylind).
er Head Sector) address or LBA (Linear)
Block Address) address can be used.

Classification field by neighbor address 222c
Are those whose start addresses (on the disk 11-i) of the corresponding segments are close to each other (here, those on the same track or adjacent tracks), that is, the head 12
Are grouped with each other in a positional relationship in which the movement operation (disk access) can be optimally performed (with a minimum seek amount and a skew amount within a predetermined time), and the information, that is, the classified (grouped) data Types and their number,
And link information of the same classification.

Here, the “data type” is defined as follows: (1) Relationship on the same track (value 0) (2) Relationship on an adjacent track in the same subzone SZ (value 1) (3) Subzone SZ There are also three types of relations (value 2) on the adjacent track.

Next, the “number” indicates the number of the segment management blocks 222 determined to be of the same classification (group) at the time when the current segment management block 222 is generated.

Next, the “link information” includes the most recently generated (ie, the latest) segment management block 2 in the same category as the current segment management block 222.
No. 22 (block number) is used. If there is no corresponding item, that is, if there is no other item belonging to the same classification, the block number # 0 is used.

FIG. 5 shows a specific example of the contents of the segment management table 220. In this case, as shown in FIG. 5A, addresses u, v, and u of one block length (one sector size length) are used.
The contents of the segment management table 220 are shown in FIG. 5 (FIG. 5) in the case where commands for writing data to w, x, y, z (data sectors), that is, six write commands for discontinuous addresses are sequentially issued. This is shown in b).

In the example of FIG. 5, the addresses u, v, and y (data sectors) are on the same track, and are therefore classified into the same group. Therefore, the "type of data" is stored in the classification field 222c for each neighboring address in the three segment management blocks # 1, # 2, and # 5 (222) generated in response to the write commands corresponding to the addresses u, v, and y. Are set as values indicating “number”, 1, 2, and 3 are set as values indicating “number”, and # 0, # 1, and # 2 are set as “link information”, respectively.

Similarly, the addresses w and z (data sectors of) are on the same track, and the addresses w and z (data sectors of) and the addresses x (data sectors of) are on adjacent tracks across the sub-zone SZ. , The addresses w, x, and z are classified into the same group. Therefore, three segment management blocks # 3, # 3 generated in response to the write commands corresponding to addresses w, x, z
4, # 6 (222), 0, 2, 0 as values indicating "data type" and 1, 1 as values indicating "number" in the neighborhood address classification field 222c, respectively.
# 2, # 3 as "link information"
# 3 and # 4 are set.

Next, the operation of this embodiment will be described with reference to the flowchart of FIG. It is assumed that a command, for example, a write command has been issued from the host to the magnetic disk device of FIG. This write command is transmitted from the host I /
It is received by the HDC 23 via F25. HDC23
When receiving a command from the host, the CPU issues an interrupt to that effect to the CPU 20. The CPU 20 receives and interprets the received command (here, the write command) from the HDC 23 by this interrupt.

If the command received from the HDC 23 is a write command, the CPU 20 checks the number of write sectors specified by the command, and the block length (sector number) of data transferred from the host together with the command is equal to or smaller than a threshold (for example, 8 sectors or less)
That is, it is determined whether or not the command is a write command designating writing to a disk having a small block length (step S1).

If not, that is, if the block length exceeds eight sectors, the CPU 20 controls the HDC 23 to transfer the transfer data from the host to a buffer area different from the segment cache buffer 240 in the buffer memory 24 (see FIG. No. 7 buffer area corresponding to burst transfer segment B or read / write segment A)
A normal write process for writing to the disk is executed immediately, when the buffer area is full, or when sequential data of another address is transferred (step S2).

On the other hand, if the block length is 8 sectors or less, the CPU 20 controls the HDC 23 to secure a segment #i of a corresponding size in the segment cache buffer 240 of the buffer memory 24, and the segment #i To store the transfer data from the host (step S3).

Next, the CPU 20 updates the content (segment division number) of the segment division number field 221 of the segment management table 220 placed in the RAM 22 for managing the segment cache buffer 240.
That is, it is incremented by one (step S4).

Next, the CPU 20 creates, in the segment management table 220, a new segment management block 222 (#i) corresponding to the segment #i in which the transfer data is stored in step S3, and writes a new command to the field 222a by a write command. The write start address #i on the specified disk 11 is stored in the field 222b with the segment length #
i are registered (step S5).

Next, the CPU 20 searches for a start address (set in the field 222a) of the already registered (that is, previously created) segment management block 222 (step S6), and finds one of the searched start addresses. It is determined whether or not the start address of the segment management block 222 newly created in step S5 (that is, the current segment management block 222) is in a relation of a neighboring address (step S7). The determination as to whether or not there is a neighborhood address relationship is made based on whether or not they are on the same track or on an adjacent track.

If the result of determination in step S7 is that the address is a neighborhood address, the CPU 20 registers the information in the neighborhood address classification field 222c of the current segment management block 222 as the same group (step S7).
8). Here, as the “type of data”, when the start address of the current segment management block 222 is on the same track as the start address of the registered segment management block 222, the value 0 is set in the same sub-zone SZ. The value 1 if it is on an adjacent track of
If it is on an adjacent track across the sub-zone SZ, the value 2 is registered. In addition, as the “number”, the number of segment management blocks 222 in the same group including the current segment management block 222 (the maximum number of segment management blocks 222 in the same group is 1).
Is added). Further, as the “link information”, the number of the latest segment management block 222 among the segment management blocks 222 of the same group, that is, the largest block number of the segment management blocks 222 of the same group is registered.

On the other hand, if the result of determination in step S7 is that the address is not a nearby address, the CPU 20 registers information as a separate group in the nearby address classification field 222c of the current segment management block 222 (step S9). Here, as in the example of the segment management block # 1 in FIG.
The value 0 is registered as “number”, 1 is registered, and “link information” is registered as “# 0”.

After executing step S8 or S9, the CPU 20 determines whether or not any of the segment management blocks 222 in the segment management table 220 satisfies the write condition (step S10). Here, that the write condition is satisfied means that the number of the segment management blocks 222 classified as the same group exceeds a predetermined number.

If no write condition is satisfied, the CPU 20 waits for the next command from the host. On the other hand, if there is a group (group) that satisfies the write condition, the CPU 20 controls the HDC 23 to control the HDC 23 so that the number of segment management blocks in the segment management table 220 exceeds the predetermined number in the same group (that is, for each neighboring address). For the same group classified by the classification field 222c, the data of the segment in the segment cache buffer 240 corresponding to each segment management block 2222 belonging to the block is continuously written to the disk 11-i (step S1).
1). In other words, among the data of the same group classified by the neighboring address, the data of the segments of the group in which the number of the segment management blocks exceeds a certain number is continuously written to the disk irrespective of the order in which the corresponding commands are issued. Is performed.

After executing step S11, the CPU 20 deletes the segment management block 222 for which writing to the disk has been completed from the segment management table 220,
Segment division number field 2 in the table 220
The value of 21 (the number of segment divisions) is subtracted by the number of deleted segment management blocks (step S12). Then, the CPU 20 waits for the next command from the host.

As described above, in the present embodiment, when a write command instructing a small block size disk write to a discontinuous address is continuously issued, the data transferred from the host is transferred to the buffer memory. In the cache buffer 240 in the H.24, the data is held in segments secured in command units. On the other hand, when there is another sequential disk write request, data from the host is stored in the buffer memory 24 in a buffer area for sequential access that is secured separately from the cache buffer 240, and the normal Treat as a disk write (conventionally known).

Further, in the present embodiment, the data stored in the segment secured in the cache buffer 240 in units of commands is stored in the segment management block 22 generated in the segment management table 220 corresponding to the segment.
Using 0 (neighboring address classification field 222 therein), the disc is classified into groups in which continuous disk writing is possible. Then, for the data of the group that satisfies the predetermined write condition, the corresponding segment management table 220 (in the segment management block 22 in the corresponding segment management table 220).
According to 2), the disk writing is continuously performed. As a result, the amount of head movement and the number of disk accesses can be reduced, that is, overhead can be reduced, and the speed of random write can be improved.

Hereinafter, the reason will be described with reference to FIG. As described above, when accessing a disk, a disk rotation wait occurs in order to access distant data sectors on the same track. In addition, track seek or track skew occurs to access a distant track, and head skew occurs even when head switching is performed. That is, when accessing a random address, a huge overhead due to seek and skew occurs.

The sub-zone SZ is a means for improving the performance at the time of such random access. By making the access direction the head direction within the same sub-zone SZ, the average in the cylinder direction (radial direction) is obtained. The seek distance is shortened. However, even with these improvements, if random access requests occur,
In the prior art, a disk write is performed each time, so that the overhead is not small.

Therefore, in the present embodiment, when a write command instructing writing of a small block size disk to discontinuous addresses is continuously issued, the disk writing is not immediately performed as in the prior art, that is, the write command is not issued. Each time is issued, the designated data (write data) is held in the cache buffer 240 in the buffer memory 24 without writing to the disk, and the overhead is small in consideration of the timing at the time of writing to the disk. The overhead caused when the data sectors on the track a access the next data sector on the same track a is, for example, the rotation wait only, and is the smallest. Therefore, in the present embodiment, addresses on the same track are classified as one group, and set in the near-address-specific classification field 222c in the segment management block 222. Thus, when the group satisfies the write condition, the disk can be continuously written to each address belonging to the group, thereby reducing overhead.

Similarly, addresses on adjacent tracks in the same sub-zone SZ are also classified as one group, and when the group satisfies the write condition, disk writing to each address belonging to the group is performed. By performing the operations continuously, the overhead can also be reduced.

In addition, for example, the start address of the data of the segment # 1 in FIG. 4 is on the track m, the start address of the next segment # 2 is on the track k, and the next segment #
Assuming that the start address of No. 3 is on track c, c and m are adjacent tracks, but are located at a completely separated distance on the logical address due to the sub-zone SZ. For this reason, in the present embodiment, c and m on adjacent tracks across the sub-zone SZ are also managed as the same group, and when the group satisfies the write condition, the data sector on the track c and the data By continuously writing the data to the data sectors, the overhead can also be reduced. Conversely, in FIG. 3, track l and track m are continuous on a logical address, but are physically (on a physical address).
Are completely separated from each other, and the overhead for moving from the track 1 to the track m increases.

In the above-described embodiment, the writing condition is such that the number of the segment management blocks 222 classified as the same group exceeds a predetermined number, that is, whether the writing condition is satisfied. Is determined for each group (and based on the number of segment management blocks of the group), but the present invention is not limited to this.

For example, it may be determined that the writing condition is satisfied when the number of segment divisions reaches a threshold value, for example, the maximum value of the number of segment management blocks 222 that can be generated in the segment management table 220. In this case, for all the segment management blocks 222 in the segment management table 220, disk writing is performed continuously for each group. here,
The start address (the start address (which is set) has a positional relationship that minimizes the seek amount and the skew amount with respect to the current position of the head 12 on the disk 11-i, that is, the positional relationship that allows the disk access (head moving operation) to be optimal. If the disk write for the group including the segment management block 222 in the start address field 222a) is performed first, the overhead can be further reduced. The following disk writing order may be determined based on the position of the preceding group on the disk 11-i. By writing data to the disk when the number of segment divisions reaches the predetermined number, the number of segment divisions increases and the segment management table 2
20 and the segment cache buffer 240 can be prevented from overflowing.

Alternatively, it is possible that the write condition is satisfied when the elapsed time (holding time) of the data held in the segment cache buffer 240 exceeds a predetermined time. Specifically, each segment management block 222 is provided with time information at the time of registration, and the oldest time among the times of the segment management blocks 222 that were last grouped according to the neighborhood address classification is checked. If the difference exceeds a predetermined time, it is determined that the write condition has been satisfied. Such a so-called L
By applying a management method based on the RU method (Least Recently Used Rule), even a group that does not have a disk access opportunity can be a target of disk writing.

Further, the disk 11-i of the current head 12
The presence of a group including the segment management block 222 of an address (a start address field 222a in which is set) in a positional relationship in which disk access (head moving operation) can be optimally performed from the upper position,
It is also possible that the write information is satisfied.

In the embodiments described above, the case where the present invention is applied to a magnetic disk device has been described. However, the present invention is also applicable to a disk storage device such as a magneto-optical disk device.

[0081]

As described above in detail, according to the present invention, when a write command instructing a disk write of data of a small block size to a discontinuous address is continuously issued, the transfer data from the host is transmitted. Is stored in the corresponding divided segment, and disk access is performed when physically continuous sectors can be accessed to some extent, so that the amount of head movement and the number of disk accesses are reduced to reduce overhead. Can be reduced, and the speed of random write can be improved.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a magnetic disk drive according to an embodiment of the present invention.

FIG. 2 is a conceptual diagram of a format of a disk 11-i (i = 1, 2) in FIG.

FIG. 3 is a conceptual diagram showing a relationship between a zone Z, a sub-zone SZ, and a track on each recording surface of disks 11-0 and 11-1 in FIG.

FIG. 4 is a segment cache buffer 240 in FIG. 1;
FIG. 4 is a diagram showing a state of segment division and a data structure example of a segment management table 220 for managing each segment in the cache buffer 240.

FIG. 5 is a diagram showing a specific example of the contents of a segment management table 220 in a case where six write commands for discontinuous addresses are issued in order.

FIG. 6 is an exemplary flowchart for explaining the operation at the time of receiving a write command in the embodiment.

FIG. 7 is a diagram for explaining a conventional technique suitable for sequential write for the purpose of continuously writing data to a disk without delay when a plurality of write blocks are continuously transmitted.

[Explanation of symbols]

11-0, 11-1, 11-i Disk (disk medium) 12 Head 18 R / W circuit 20 CPU (segment cache control means, disk write control means) 21 ROM 22 RAM 23 HDC ( Disk controller, disk write control means) 24 buffer memory 220 segment management table 221 segment division number field 222 segment management block 222a start address field 222b segment length field 222c near neighbor address classification field 240 cache buffer

Claims (7)

[Claims] 1. A disk storage device comprising: a buffer memory for temporarily storing data transferred between a disk medium on which recording and reproduction are performed by a head and a host; A segment cache buffer that can be divided into segments, and a segment management block generated for each segment in the segment cache buffer, and an address indicating a write destination of data stored in the segment on the disk medium An address field in which information is set, a segment length field in which a segment length indicating the size of the segment is set, and information for classifying address information in a positional relationship in which a head moving operation can be performed within a predetermined time as the same group. Address where is set A segment management table storing means for managing the segment cache buffer, comprising a segment management block having a classification field; and a host issuing a write command designating disk writing and transferring data to be written to the disk. When the data size is equal to or smaller than a predetermined size, the data is stored in a segment of the same size allocated in the segment cache buffer, and the segment management block for managing the segment is stored in the segment management table. A segment cache control means generated in the segment management table, and a predetermined one of the segment management blocks classified in the same group in the neighborhood address classification field generated in the segment management table. And a disk write control unit for performing disk write on a group basis based on a segment management block of a group satisfying the write condition. 2. The disk write control means determines that the write condition is satisfied when the number of segment management blocks classified into the same group in the neighborhood address classification field exceeds a predetermined number. The disk storage device according to claim 1, wherein 3. The disk storage device according to claim 1, wherein the disk write control unit determines that the write condition is satisfied when the number of segment divisions in the segment cache buffer reaches a predetermined number. 4. The disk write control unit checks, for each group classified by the neighborhood address classification field, an elapsed time from the generation of the segment management block most recently generated in the group. ,
2. The disk storage device according to claim 1, wherein it is determined that the write condition is satisfied for the corresponding group when a predetermined time has elapsed. 5. The disk writing control means includes a group including a segment management block having address information in a positional relationship from a current position on the disk medium of the head to a head moving operation within a predetermined time. 2. The disk storage device according to claim 1, wherein it is determined that the write condition is satisfied for the group. 6. The segment cache control means,
When the segment management block is generated, even if there is a previously generated segment management block having address information that is continuous with the corresponding address information, if there is no positional relationship in which the moving operation of the head can be performed within a predetermined time. 2. The disk storage device according to claim 1, wherein the processing is performed as a non-target of the group to which the generated segment management block belongs. 7. A segment cache buffer which can be divided into a plurality of segments is secured in a buffer memory for temporarily storing data transferred between a disk medium on which recording and reproduction are performed by a head and a host. A method of controlling segment cache in a disk storage device, wherein each time a write command designating disk writing is issued from the host, the size of data transferred from the host together with the command is equal to or smaller than a predetermined size. In the case where the data is equal to or smaller than a predetermined size, the data is stored in a segment of the same size allocated in the segment cache buffer, and every time data from a host is stored in the segment,
A segment management block for managing the segment, address information indicating a write destination on the disk medium of data stored in the segment,
A segment management table in which a segment management block in which the segment length indicating the size of the segment and the neighboring address classification information for classifying the address information in a positional relationship in which the head movement operation can be performed within a predetermined time as the same group is set. Among the segment management blocks generated in the segment management table and classified into the same group by the neighboring address classification information, to detect a segment management block of a group that satisfies a predetermined write condition. And writing data to the disk in units of the group based on the segment management block of the group.