JP2000222287A - Information recording and reproducing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To smoothly and efficiently transfer data with a host computer by transferring data from a cache memory to a recording medium at high speed. SOLUTION: A CPU 8 is composed of a command control part 21 for instructing completion to a command from the host computer, a control part 24 for storing cache managing data 22 and cache hit managing data 23 in a storage part 7, a cache hit discriminating part 25 for discriminating hit on the basis of the cache managing data 22, a sort function part 26 for storing the LBA of data stored by the cache managing data 22 or cache hit managing data 23 for the unit of managing data through the control part 24, and a write control part 27 for controlling an access control part for writing the non-written data in a cache memory onto an optical disk 2.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information recording / reproducing apparatus, and more particularly, to an information recording / reproducing apparatus characterized by a cache hit management part.

[0002]

2. Description of the Related Art In general, an information recording / reproducing apparatus is used as an external recording apparatus of a host computer in an information processing apparatus. However, a process of writing data to a recording medium depends on a data transfer processing speed with the host computer. In recent years, a cache memory is provided in the middle, and data is transferred through the cache memory to increase the speed of the data recording process.

In particular, in an optical disk apparatus using an optical disk as a recording medium, since it takes a lot of time to write data to the recording medium, it is common to provide a cache memory for temporarily storing recording data.

As a device provided with such a cache memory, for example, there is a disk cache device disclosed in Japanese Patent Application Laid-Open No. Hei 2-165248.
When a write command is issued, data from the host computer is written to the disk cache memory, and a write completion signal to the host computer is output to resume data processing. As a result, the unwritten data stored in the disk cache memory is written to the recording medium.

According to such a write-after type apparatus, the host computer recognizes the end of writing at the time of writing data to the disk cache memory and can resume data processing. Therefore, it is necessary to wait until the writing to the recording medium is completed. Therefore, the access time during the writing process can be reduced.

In a conventional apparatus, when there is a data read request (hereinafter referred to as a read command), if there is unwritten data in a recording medium in a cache memory, a read command from a host computer is issued. After waiting, all the unwritten data has been written to the recording medium, and then the requested data is read.

That is, in this case, the data stored in the cache memory becomes valid after being written to the recording medium, and is read from the cache memory when data corresponding to the read instruction exists in the cache memory (when a cache hit occurs). It is issued and transferred to the host computer.

[0008]

However, in the above-described information recording / reproducing apparatus having the conventional cache memory, when writing data from the cache memory to the recording medium, the seek for each LBA (logical block address) is performed efficiently. It is not performed well, and there is a problem that the time for actually writing data to the recording medium is not shortened.

Similarly, when writing data to a recording medium, if the address (LBA) of the data requested by the host computer is the same or duplicated in the cache memory, the same or duplicated data is repeated. Access to addresses is likely to occur.

[0010] If these data are left in the cache memory as much as possible and all the unwritten data in the cache memory are written and then these data are written to the recording medium, the number of seeks is drastically reduced and the speed is increased. Data transfer can be performed.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and enables high-speed writing from a cache memory to a recording medium, and smooth and efficient data transfer with a host computer. It is an object to provide an information recording / reproducing device.

[0012]

An information recording / reproducing apparatus according to the present invention comprises: a cache memory for temporarily storing data sent from a host computer; and a cache management data for storing cache management data for managing data in the cache memory. Storage means; cache hit determination means for determining a cache hit when data from the host computer is present in the cache memory based on the cache management data; and the cache determined to be a cache hit by the cache hit determination means A cache hit management data storage unit for storing cache hit management data for managing data in a memory; and a cache hit management data storage unit and / or a cache hit management data storage unit. Configured with a sorting means for sorting the BA in administration data units, and a write means for writing the unwritten data in the cache memory to the recording medium.

In the information recording / reproducing apparatus according to the present invention, when the writing means writes unwritten data in the cache memory to a recording medium, when the data managed by the cache management data is accessed again, Registered in the cache hit management data, after writing all the data in the cache management data to the recording medium, by writing the data registered in the cache hit management data to the recording medium after a predetermined time has elapsed,
The writing from the cache memory to the recording medium can be performed at high speed, and the data transfer with the host computer can be performed smoothly and efficiently.

[0014]

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

FIGS. 1 to 11 relate to an embodiment of the present invention. FIG. 1 is a configuration diagram showing a configuration of an optical disk device.
2 is a configuration diagram showing the configuration of the CPU and the recording unit of FIG. 1, FIG. 3 is a first explanatory diagram showing the structure of the cache management data and cache hit management data of FIG. 2, and FIG. 4 is the cache management data of FIG. FIG. 5 is a first flowchart illustrating the operation of the optical disk device of FIG. 1, and FIG. 6 is a second flowchart illustrating the operation of the optical disk device of FIG. 7, FIG. 7 shows a specific example of the cache management data of FIG. 2, FIG. 8 shows a specific example of the cache hit management data of FIG. 2, and FIG. 9 shows a modification of the specific example of the cache management data of FIG. FIG. 10 is a diagram showing a specific example of cache hit management data when a cache hit occurs in the cache management data of FIG. 7, and FIG. 11 is a diagram showing cache management data of the cache management data of FIG. It is a diagram showing a specific example of the cache management data when Yuhitto occurs.

(Configuration) First, the configuration of a portion relating to data recording / reproducing operation of the information recording / reproducing apparatus used in the present embodiment will be described. FIG. 1 shows an example of a configuration of a main part of an information recording / reproducing apparatus. Here, an optical disk apparatus using an optical disk as a recording medium will be described as an example.

The optical disk device 1 is connected to a host computer 10 that requests recording and reproduction of information, and is a head 3 for writing and reading information to and from an optical disk 2 as a recording medium. An access control unit 4, a head driver 5 including a modulation / demodulation circuit for recording / reproducing data, a cache memory 6 as an intermediate storage unit for temporarily storing recording / reproducing data,
A storage unit 7 for storing track information and data amount information in a command from the host computer 10, a CPU 8 as control means for performing various controls in the apparatus, and exchanges commands and data with the host computer 10. An I / F (interface) unit 9 is provided.

The command sent from the host computer 10 includes, in addition to the instruction content, track information and data amount information on a recording medium corresponding to the target data. It is included.

The optical disk device 1 inputs a command from a host computer 10 via an I / F unit 9,
Track information and data amount information in the command are stored in the storage unit 7 composed of a semiconductor memory or the like. In the recording area of the optical disk 2, a plurality of tracks are formed spirally or concentrically from the inner circumference to the outer circumference.
When recording or reproducing information, a seek operation of moving the head 3 by a plurality of tracks in a direction orthogonal to the track is performed to position the head 3 on a target track.

When the command is a recording command (hereinafter, referred to as a write processing command), the recording data (hereinafter, writing data) sent together with the command from the host computer is transferred to the cache memory 6 and temporarily stored.
Then, under the control of the CPU 8, the head 3 seeks the head 3 to the target track requested by the command on the optical disk 2 by the access control unit 4, and after the seek is completed, the write data stored in the cache memory 6 is read by the head driver 5. A modulation circuit of the head driver 5 modulates a semiconductor laser (not shown) constituting the head 3 and irradiates a recording light beam onto a target track of the optical disk 2 to record information.

When the command is a reproduction command (hereinafter, referred to as a read processing command), the CPU 8 controls the head 3 to read the reproduction data (hereinafter, read data) requested by the command by the access control unit 4. 2. A seek operation is performed on the target track on the track 2 and a read light beam is applied to the track to read out information. The information is demodulated by the modulation circuit of the head driver 5. To the host computer 10.

As shown in FIG. 2, the CPU 8 includes a command control unit 21 for instructing completion of a command from the host computer 10 via the I / F unit 9 and a cache management unit for managing data in the cache memory 6. Data 22
A control unit 24 for storing, in the storage unit 7, cache hit management data 23 for managing data when request data from the host computer 10 exists in the cache memory 6 based on the cache management data 22; Host computer 10 based on
A cache hit determination unit 25 that determines a hit if the request data from the cache memory 6 exists, and the LBA of the data stored in the cache management data 22 or the cache hit management data 23 via the control unit 24. It comprises a sort function unit 26 for rearranging data into units and a write control unit 27 for controlling the access control unit 4 to write unwritten data in the cache memory 6 to the optical disc 2.

The CPU 8 receives a control signal from the host computer 10 via the I / F unit 9, and controls the timing of command transmission / reception according to the control signal. The access control unit 4 and the head driver 5 are controlled by the CPU 8, but their detailed configurations are well-known and will not be described.

Next, referring to FIGS. 3 and 4, the cache management data 22 and the cache hit management data 23 will be described.
The structure of will be described.

As shown in FIG. 3, the cache management data 22 and the cache hit management data 23 are managed by management numbers 1 to n, and each management number consists of, for example, eight block data. At this time, one block is managed by one sector size of the optical disk 2. Although each management number is made up of eight block data, the number is not limited to this.

As shown in FIG. 4, the management data managed by each management number is an unprocessed data indicating whether the data in the block has already been written or has not been written. The data flag includes a data flag, a hit flag indicating whether the data in the block has a cache hit, and a buffer address indicating at which address in the cache memory 6 each block is located.

The cache hit management data 23 has a variable size (capacity), which will be described below.

A cache hit occurs when a write instruction is issued again to data in the cache management data 22. Therefore, depending on the OS (Operating System) of the host computer 10, a write processing command is frequently issued to the same address area in a file format to be copied, and a cache hit is likely to occur.

Table 1 shows that files having a directory structure (data hierarchical structure) are actually stored in the host computer 10.
2 shows an example of a SCSI command when copying from the optical disc device 1 to the optical disc device 1.

[0030]

[Table 1] In Table 1, taking the command “004354” as an example, “CM: 2A” is a write command, and “00 02 41” is an LBA that starts writing.
And the subsequent “01” is the write data transfer length (Le
ngth).

A to D in Table 1 indicate commands that have been accessed twice or more at the same address. As can be seen from the example of Table 1, there are many write processing instructions to the same address area.

In particular, as the directory structure becomes more complicated, the number of directories to be written increases, and the number of times of rewriting the same address area increases, so that the number of cache hit targets increases.

For example, if the cache hit management data 23 is not so large in spite of the complexity of the directory structure, the cache hit management data 23 is released despite the possibility of a sufficient cache hit thereafter. In some cases, there is no other way. As a result of the release, there is a possibility that an access to the area managed in the cache hit management data 23 up to the previous time may occur.

However, by increasing the size of the cache hit management data 23, it is possible to manage a large amount of cache hit areas that may be written many times later in the cache hit management data 23, Since the data is written only once, the number of times of writing can be reduced and the speed can be increased.

On the other hand, when the directory structure is simple, if the size of the cache hit management data 23 is large, the cache hit management data 23 is not fully utilized and a free space remains, and the available cache management data 23 is not used. 22 is reduced in size.

However, in such a case, by reducing the size of the cache hit management data 23, more cache management data 22 can be used.

As described above, by making the size of the cache hit management data 23 variable, the speed at the time of writing can be increased and the cache can be efficiently used.

(Operation) The operation of the present embodiment thus configured will be described. First, referring to FIG.
A data processing method in the case where the PU 8 receives a write processing command from the host computer 10 of the present embodiment will be described.

As shown in FIG. 5, when the CPU 8 receives the write processing command sent from the host computer 10 in step S 1 and receives a write request from the host computer 10, the CPU 8 passes through the I / F unit 9. The write data in the write command sent is divided into management block sizes.

Next, it is confirmed from the cache management data 22 in the storage unit 7 whether the LBA of the write data divided into the management block size is already managed in the cache memory 6 in step S2.

If the result of step S2 is that the LBA is new data that has not yet been managed, it is checked in step S3 whether there is free space in the cache management data 22.

As a result of step S3, if there is a free space in the cache management data 22, the process proceeds to step S4.
The write data divided by 1 is registered with reference to an address in the cache memory 6 with the data unprocessed flag enabled in the cache management data 22. After the registration, the actual write data is transferred to the specified address of the cache memory 6 and waits until writing from the cache memory 6 to the optical disk 2 is completed.

If there is no free space in the cache management data 22 as a result of step S3, the process waits for free space in the cache management data 22 in step S5.

The cache management data 22 can create new free management data by releasing the management data that has been written to the optical disk 2.

In step S2, if the LBA of the write data divided into the management block size in step S1 is found in the cache management data 22, it is determined in step S6 that a cache hit has occurred, and the hit flag of the valid block is set.

After setting the hit flag of the valid block, it is checked in step S7 whether the LBA of this valid block has already been registered in the cache hit management data 23.

If it is determined in step S7 that the block is already registered in the cache hit management data 23,
In step S8, the data of this LBA is transferred (overwritten) to the specified address of the cache memory 6 to make the unwritten flag valid.

If it is determined in step S7 that the block is not registered in the cache hit management data 23, it is checked in step S9 whether there is a free space in the cache hit management data 23. If there is a free space, the cache hit management data 23 is checked in step S10. 23 and transferred to the specified address of the cache memory 6 to make the unwritten flag valid.

If there is no free space in the cache hit management data 23 in step S9, one hit flag is cleared and released from the block already managed in the cache hit management data 23 in step S11. As a result, a new hit block is registered because a space is created in the cache hit management data 23.

Here, the cache hit management data 23
For example, as a method of clearing and releasing the hit flag in the cache hit management data 23, a method of preferentially releasing a block having a low hit rate of the block in the cache hit management data 23, or releasing the block in the order registered in the hit flag. And how to compare the size of data LBA in registered block.
Various methods are possible.

In this embodiment, the number of blocks released from the cache hit management data 23 is set to one. However, it is possible to release n fixed numbers, and to write the write data in the write command to the management block in step S1. It is possible to release an indefinite number such as the number of blocks when divided into sizes.

Next, a data processing method for transferring data from the cache memory 6 to the optical disk 2 will be described with reference to FIG.

As shown in FIG. 6, first, at step S21
It is checked from the cache management data 22 managing the cache memory whether the data unprocessed flag is valid (set).

If it is determined in step S21 that the data unprocessed flag is valid, the data in the management block in which the data unprocessed flag is valid is written to the optical disk 2 in steps S22 and S23. At this time, if the data unprocessed flag is valid for a plurality of management blocks, and if the data LBAs seem to be continuous between the management blocks in step S22, the data is collectively written to the optical disk 2.

Similarly, when the data unprocessed flag is valid for a plurality of management blocks, the order of writing to the optical disk 2 is optimized in step S22 by using the sorting function unit 26 so that the data is written from the vicinity of the data written immediately before. Process in the proper order.

In the processing by the sort function unit 26, various settings can be made, such as, for example, preferentially sorting the closest position from the current pickup position, the order of the data LBA, and the order of the block size. It is.

In step S23, the data of the management block is transferred to the optical disk 2, and when the writing is completed, the unprocessed flag of the processed management block is cleared in step S24, and the process ends.

In step S21, it is confirmed from the cache management data 22 managing the cache memory 6 whether the data unprocessed flag is valid. If there is no data unprocessed flag, the cache hit management data is determined in step S25. In the cache hit management data 23, it is checked whether there is any unprocessed block.

If it is determined in step S25 that there is no unprocessed block, the process ends to wait for the next command sent from the host computer 10.

If it is determined in step S25 that there is an unprocessed block, a predetermined time interval is waited in step S26, and then data similar to steps S22 to S23 is written to the optical disk 2 in steps S27 to S28. Note that the predetermined time of the time interval in step S25 is variable according to the frequency of the write command.

Then, in step S29, step S24
Similarly, the unprocessed flag of the management block for which processing has been completed is cleared, and the processing ends.

Next, specific data examples of the cache management data 22 and the cache hit management data 23 will be described.

When the host computer 10 sends a write processing instruction to the CPU 8, the cache management data 22 and the cache hit management data 23 are assumed to be as shown in FIGS. 7 and 8, for example. here,
The hatched portions indicate blocks having management data.
As shown in FIG. 8, in this case, since no cache hit has yet occurred, all of the cache hit management data is empty.

More specifically, as shown in FIG. 7, in the cache management data 22, first, LBA1 to Length
3 is managed in blocks 1 to 4 of control number 1,
The data of LBA248 to Length3 is the management number 2
Is managed in blocks 0 to 3 of LBA498 to Len.
The data of gth6 is managed by the blocks 2 to 7 of the management number 3, and when a new write processing instruction is transmitted and the data is the data of LBA99 to Length4, the data is sorted in the order of LBA by the sort function unit 26. , LBA1 to Length3 data is the management number 1
Is managed by blocks 1-4 of LBA99 to Leng
The data of th4 is managed by the blocks 3 to 6 of the management number 2, the data of LBA248 to Length3 is managed by the blocks 0 to 3 of the management number 3, and the LBA498 to Le
The case where the data of ngth6 is managed by the blocks 2 to 7 of the management number 4 is shown.

Note that in FIG.
Although the sorting in the order of BA shows an example of sorting from a small LBA to a large LBA, sorting from a large LBA to a small LBA may be performed.

For example, from LBA1 to Length1
When data of 0 is written, as shown in FIG.
LBA1 to LBA7 are managed by blocks 1 to 7 of management number 1, and LBA8 to LB are managed by blocks 0 to 2 of management number 2.
A10 is managed.

Cache management data 2 after sorting in FIG.
Assuming that a new write processing instruction is transmitted from LBA 97 at Length 4 in the state of 2, LBAs 97 and 9
8 has already been assigned to the blocks 2 and 3 of the management number 2 and is managed by this block. On the other hand, LBA9
9, 100 is a cache hit because data is already managed in blocks 4 and 5 of management number 2,
Hit flag becomes effective.

Therefore, the management number 2 of the cache management data 22 in which the block in which the hit flag is valid is shifted to the management number 1 of the cache hit management data 23 as shown in FIG. 10, and is shown in FIG. like,
In the cache management data 22, the management number in which the block in which the hit flag is valid exists is released, and the management numbers are sorted. In this way, the cache and the cache hit are managed.

(Effect) As described above, in the present embodiment, the predetermined time according to the time interval is variable with the frequency of the command from the host computer 10 and the size of the cache hit management data 23 is variable. From the optical disk 2 to the optical disk 2 at high speed.

When the data registered in the management data is written to the optical disk 6, the management data is once written to the optical disk 2 by using the sort function unit 26 to change the ascending order of the LBA. By giving priority to the data processing in the vicinity of the last data LBA and shortening the seek distance, the write time can be reduced.

[Appendix] (Appendix 1) A cache memory for temporarily storing data sent from the host computer, a cache management data storage unit for storing cache management data for managing data in the cache memory, and the cache management A cache hit determining unit that determines a cache hit when data from the host computer is present in the cache memory based on data; and manages the data in the cache memory that is determined to be a cache hit by the cache hit determining unit. A cache hit management data storage means for storing cache hit management data; and an LBA of data stored in the cache management data storage means and / or the cache hit management data storage means as a management data unit. Sorting means for rearranging, and writing means for writing unwritten data in the cache memory to a recording medium, wherein the writing means uses the cache management data when writing the unwritten data in the cache memory to a recording medium. When the managed data is accessed again, it is registered in the cache hit management data,
An information recording / reproducing apparatus, wherein after all data in the cache management data is written to the recording medium, after a lapse of a predetermined time, data registered in the cache hit management data is written to the recording medium.

(Additional Item 2) The information recording / reproducing apparatus according to additional item 1, wherein the predetermined time is variable according to a frequency of a write command from the host computer.

(Additional Item 3) The information recording / reproducing apparatus according to additional item 1, wherein the cache hit management data has a variable capacity according to a data hierarchical structure sent from the host computer.

(Additional Item 4) When writing the data registered in the cache management data and / or the management data of the cache hit management data to the recording medium, the sort unit changes the ascending order of the LBA of the data. 2. The information recording / reproducing apparatus according to claim 1, wherein writing is performed from the vicinity of the data written immediately before.

[0075]

As described above, according to the information recording / reproducing apparatus of the present invention, when the writing means writes the unwritten data in the cache memory to the recording medium, the writing means rewrites the data managed by the cache management data. When there is an access, the data is registered in the cache hit management data, and after writing all the data in the cache management data to the recording medium, the data registered in the cache hit management data is written to the recording medium after a predetermined time has elapsed. This has the effect that data can be transferred to a recording medium at high speed, and data can be smoothly and efficiently transferred to and from a host computer.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing a configuration of an optical disk device according to an embodiment of the present invention;

FIG. 2 is a configuration diagram showing a configuration of a CPU and a recording unit in FIG. 1;

FIG. 3 is a first explanatory diagram showing a structure of cache management data and cache hit management data of FIG. 2;

FIG. 4 is a second explanatory diagram showing the structure of the cache management data and the cache hit management data of FIG. 2;

FIG. 5 is a first flowchart for explaining the operation of the optical disk device of FIG. 1;

FIG. 6 is a second flowchart for explaining the operation of the optical disk device of FIG. 1;

FIG. 7 is a diagram showing a specific example of the cache management data of FIG. 2;

FIG. 8 is a diagram showing a specific example of cache hit management data of FIG. 2;

FIG. 9 is a diagram showing a modification of the specific example of the cache management data of FIG. 2;

FIG. 10 is a diagram showing a specific example of cache hit management data when a cache hit occurs in the cache management data of FIG. 7;

11 is a diagram showing a specific example of cache management data when a cache hit occurs in the cache management data of FIG. 7;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Optical disk apparatus 2 ... Optical disk 3 ... Head 4 ... Access control part 5 ... Head driver 6 ... Cache memory 7 ... Storage part 8 ... CPU 9 ... I / F part 10 ... Host computer 21 ... Command control part 22 ... Cache management data 23 cache hit management data 24 control unit 25 cache hit determination unit 26 sort function unit 27 write control unit

Continued on the front page (72) Inventor Toshihiro Ogata 2-43-2 Hatagaya, Shibuya-ku, Tokyo Inside Olympus Optical Industrial Co., Ltd. (72) Inventor Yuji Horie 2-43-2 Hatagaya, Shibuya-ku, Tokyo Olympus Optical F term in Industrial Co., Ltd. (reference) 5B005 JJ12 MM11 NN12 NN31 UU04 5B065 BA03 CE12 CH01

Claims (3)

[Claims] A cache memory for temporarily storing data sent from a host computer; a cache management data storage unit for storing cache management data for managing data in the cache memory; and the host based on the cache management data. A cache hit determining unit that determines a cache hit when data from a computer exists in the cache memory; and a cache hit management data that manages data in the cache memory determined to be a cache hit by the cache hit determining unit. A cache hit management data storage unit, and a sort for rearranging LBAs of data stored in the cache management data storage unit and / or the cache hit management data storage unit in management data units. And write means for writing unwritten data in the cache memory to a recording medium, wherein the writing means is managed by the cache management data when writing the unwritten data in the cache memory to a recording medium. If the data is accessed again, it is registered in the cache hit management data, and after writing all the data in the cache management data to the recording medium, the data registered in the cache hit management data is passed after a predetermined time has passed. An information recording / reproducing device for writing on a recording medium. 2. The information recording / reproducing apparatus according to claim 1, wherein the predetermined time is variable according to a frequency of a write command from the host computer. 3. The information recording / reproducing apparatus according to claim 1, wherein the cache hit management data has a variable capacity according to a data hierarchical structure sent from the host computer.