KR20100075727A - Device and method for migration of data recorded in recording medium - Google Patents

Abstract

PURPOSE: A transport unit of data recorded in a recording medium for improving access performance by performing data transfer is provided to improve access performance about groups of metadata by considering real access frequency. CONSTITUTION: An acquisition unit(64) acquires first information and second information about each data piece recorded in a first recording medium. The first information indicates the read frequency of the data from the first recording medium after recording data on the first recording medium. The second information indicates the times in which data is read from the first recording medium. A tracking unit(65) traces an access frequency about data. A decision unit determines the archival location of data in the second recording medium.

Description

A device and method for moving data recorded on a recording medium {DEVICE AND METHOD FOR MIGRATION OF DATA RECORDED IN RECORDING MEDIUM}

The present invention relates to an apparatus and a method for moving data recorded on a recording medium. In particular, the present invention relates to an apparatus and method for supporting the movement of data recorded on a recording medium to the recording medium or another recording medium.

For example, data recorded on one medium is often moved to another medium because the medium or the medium driving device reaches its end of life. In addition, data movement to other media may be performed for operational reasons, for example, to switch to a more cost effective medium.

In this data movement, frequently accessed data is preferably recorded in a position that is advantageous in terms of performance. This means that, for example, if a plurality of pieces of data are accessed as a single data group, such as a data object, the pieces of data recorded in the most unfavorable position in terms of performance will affect the overall access performance of the data group. Because I can give.

For the movement of data recorded on one medium to that medium or another medium, there are known techniques for improving performance in consideration of the frequency of access to data (see, for example, Patent Documents 1 to 3).

Patent Document 1 discloses that a frequently accessed logical volume is arranged in a RAID group in a long time power supply mode, and a less frequently accessed logical volume is a RAID group in a first short-time energization mode. And the least frequently accessed logical volume discloses a technique in which the rearrangement plan is arranged to be arranged in a RAID group in a second short-time power supply mode.

Patent document 2 stores an access frequency for each file including a plurality of blocks, extracts a plurality of files that are accessed at a frequency of a predetermined value associated with each other as a single group, and in each of the extracted files Disclosed is a technique for reading the data of each block from an optical disc and sequentially writing the read data pieces into successive blocks or adjacent blocks of the optical disc.

Patent document 3 discloses a technique for providing a memory to a cartridge of a tape cassette and recording information on the frequency of access to each partition in the memory. To copy a tape, information about the frequency of access is read and data pieces are written in descending order of access frequency from the beginning of the tape.

[Patent Document 1]

Japanese Patent Application No. 2007-164650

[Patent Document 2]

Japanese Patent Application No. 8-263335

[Patent Document 3]

Japanese Patent Application No. 11-31376

According to Patent Documents 1 to 3, the access frequency is recorded for a data group (e.g., logical volume, file, or partition) containing a plurality of data pieces. The data group may include data that has existed for a long time and only recently generated data. For data fragments created at different times, the actual access frequency cannot be obtained when determining the access frequency with reference to the time base. Therefore, the techniques disclosed in Patent Documents 1 to 3 have the disadvantage that it is impossible to shape the performance by considering the actual access frequency in data movement.

Accordingly, an object of the present invention is to improve performance by considering the actual access frequency when moving data without using the time base.

In order to achieve the above object, the present invention provides an apparatus for supporting movement of pieces of data recorded on a first recording medium to a second recording medium. The apparatus further comprises: for each of the pieces of data recorded on the first recording medium, first information indicating the number of times any data has been read from the first recording medium after the pieces of data have been recorded on the first recording medium and the An acquiring unit for acquiring second information indicating the number of times a piece of data has been read from the first recording medium; A tracking unit for tracking the frequency of access to the data pieces using the first information and the second information obtained by the acquiring unit; And a determining unit that determines a recording position of the data piece in the second recording medium based on the access frequency tracked by the tracking unit.

The apparatus may further include an updating unit for updating the first information for the piece of data and the second information for the specific data when specific data has been read from the first recording medium. In this case, the specific data may include main data and metadata that is additional information about the main data. The updating unit may determine that the specific data has been read when at least one of the main data and the metadata has been read from the first recording medium.

The first information indicates the number of times the first recording medium is mounted on the storage unit after the data piece has been recorded on the first recording medium, so that the first recording after the data piece has been recorded on the first recording medium It can indicate the number of times any data has been read from the medium.

The tracking unit can track the ratio of the number of times indicated by the second information and the number of times indicated by the first information as the access frequency. Alternatively, the tracking unit may track the access frequency using the second information only for data whose number of times indicated by the first information is within a predetermined range.

The second information is information representing a read history of the data piece, and the history may be represented by a plurality of bit values. Of the number of times M which data has been read from the first recording medium, the number of times the data piece has been read may be represented by N bit values of the plurality of bit values, where N is less than M.

The determining unit is further configured to determine the data in the second recording medium such that frequently accessed data is recorded at the first recording position in the second recording medium and less frequently accessed data is recorded at the second recording position in the second recording medium. The recording position of the piece can be determined, and the second recording position is a position where a time required for reading data from the second recording medium is longer than that at the first recording position. In this case, the second recording medium may be a tape medium in which data is read in the direction from the first end to the second end of the tape medium, wherein the second recording position is the first recording position at the second end. It may be closer to the second end than closer to. Alternatively, the second recording medium may be a tape medium in which data is read in the direction from the first end to the second end of the tape medium, wherein the tape medium divides the tape medium in at least one position in the longitudinal direction. Thereby having a plurality of regions formed; The first recording position is a position in a first region of the plurality of regions, the second recording position is a position in a second region of the plurality of regions, and the second region is the first region is the second end portion. Closer to the second end than closer to.

The present invention also provides an apparatus for supporting rearrangement of pieces of data in a recording medium. The apparatus includes, for each of the pieces of data recorded on the recording medium, first information indicating the number of times any data is read from the recording medium after the pieces of data have been recorded on the recording medium and the data pieces being recorded on the recording medium. An acquiring unit for acquiring second information indicating the number of times read from the; A tracking unit for tracking the frequency of access to the pieces of data using the first information and the second information acquired by the acquiring unit; And a determining unit that determines the recording position of the piece of data in the recording medium based on the access frequency tracked by the tracking unit.

The invention also includes an apparatus for moving pieces of data recorded on a first tape medium to a second tape medium. The apparatus is adapted for each piece of data recorded on the first tape medium and including metadata indicative of main data and information associated with the main data, which data is to be written after the piece of data is written to the first tape medium. An acquiring unit for acquiring first information indicating the number of times the first tape medium has been mounted in the tape drive for reading and second information indicating the number of times the piece of data has been read from the first tape medium; A tracking unit that tracks a ratio of the number of times indicated by the second information obtained by the acquiring unit and the number of times indicated by the first information as an access frequency for the data piece; And a recording unit for recording the piece of data on the second tape medium in descending order of the access frequency tracked by the tracking unit.

The present invention also provides a method for supporting movement of pieces of data recorded on a first recording medium to a second recording medium. The method includes for each of the pieces of data recorded on the first recording medium, first information indicating the number of times any data has been read from the first recording medium after the pieces of data have been recorded on the first recording medium and the Obtaining second information indicative of the number of times a piece of data has been read from the first recording medium; Tracking the frequency of access to the pieces of data using the obtained first information and the obtained second information; And determining a recording position of the piece of data in the second recording medium based on the tracked access frequency.

The invention also includes a program product for causing a computer to function as an apparatus for supporting movement of pieces of data recorded on a first recording medium to a second recording medium. The apparatus further comprises: for each of the pieces of data recorded on the first recording medium, first information indicating the number of times any data has been read from the first recording medium after the pieces of data have been recorded on the first recording medium and the An acquiring unit for acquiring second information indicating the number of times a piece of data has been read from the first recording medium; A tracking unit for tracking the frequency of access to the data pieces using the first information and the second information obtained by the acquiring unit; And a determining unit that determines a recording position of the data piece in the second recording medium based on the access frequency tracked by the tracking unit.

The present invention makes it possible to improve performance by taking into account the actual access frequency in data movement without using the time base.

Best Mode for Carrying Out the Invention The best method for carrying out the present invention (hereinafter referred to as "embodiment") will now be described in detail with reference to the accompanying drawings. Although the present invention is applicable to any type of recording medium, a tape medium (hereinafter simply referred to as "tape") will be described as an example. This is because it is nowadays difficult to store ever-increasing amounts of data in high speed storage devices, and it is important that the pieces of data stored in low speed storage devices such as tape drives are also arranged based on the access frequency.

First, a digital data storage device (hereinafter, simply referred to as a "storage device") to which the present invention is applied will be described.

1 illustrates an exemplary configuration of a storage device 100 to which the present embodiment is applied.

As illustrated, the storage device 100 includes a tape drive 10, a control mechanism 30, an accessor 40, and a cartridge slot 50.

The tape drive 10 includes a host interface (hereinafter referred to as "host I / F") 11, a buffer 12, a channel 13, a write head 14a, a read head 14b, and a motor 15. ). The tape drive 10 further includes a controller 16, a head position control system 17, and a motor driver 18. A tape cartridge 20 (hereinafter simply referred to as "cartridge") 20 that can be inserted and loaded into the tape drive 10 is also illustrated in FIG. 1. The cartridge 20 comprises a tape 23 wound on reels 21, 22. As the reels 21 and 22 rotate, the tape 23 moves longitudinally from the reels 21 to the reels 22 or from the reels 22 to the reels 21. Although the magnetic tape is described as an example of the tape 23, the tape 23 may be a tape medium other than the magnetic tape.

The cartridge 20 further includes a cartridge memory 24. For example, the cartridge memory 24 records information on how data is recorded on the tape 23. High-speed access to data can be realized by checking the index of the data recorded on the tape 23 and how to use the tape 23, for example via a contactless RF interface. An interface such as an RF interface that enables access to the cartridge memory 24 is illustrated in FIG. 1 as a cartridge memory interface (hereinafter referred to as "CM I / F") 19.

The host I / F 11 communicates with the host 200 via the control mechanism 30. For example, from the host 200, the host I / F 11 sends a command to write data to the tape 23 (first command), a command to move the tape 23 to a desired position (second command), And a command (third command) to read data from the tape 23. For example, the host I / F 11 uses SCSI as a communication standard. In SCSI, a first command corresponds to a write command, a second command corresponds to a locate command or a space command, and a third command corresponds to a read command. The host I / F 11 provides the host 200 with a response indicating whether the processing corresponding to this command was successful.

The buffer 12 is a memory that stores data to be written on the tape 23 and data read from the tape 23. For example, the buffer 12 is dynamic random access memory (DRAM). The buffer 12 has a plurality of buffer segments. Each buffer segment stores a data set that is a unit of reading and writing data to and from the tape 23.

The channel 13 is a communication channel used to transmit data to be recorded on the tape 23 to the recording head 14a and to receive data read from the tape 23 from the recording head 14b.

When the tape 23 moves in the longitudinal direction, the recording head 14a writes the information to the tape 23, and the readhead 14b reads the information from the tape 23. As shown in FIG.

The motor 15 rotates the reels 21 and 22. The motor 15 is represented by a single rectangle in FIG. 1, but it is preferred to provide two motors 15 on each reel 21, 22.

The controller 16 controls the overall operation of the tape drive 10. For example, in accordance with a command received by the host I / F 11, the controller 16 controls the writing or reading of data to and from the tape 23. The controller 16 also controls the head position control system 17 and the motor driver 18.

The head position control system 17 is a system that performs control so that the recording head 14a and the read head 14b can track one or more desired wraps. Here, the term "wrap" refers to a group of tracks on the tape 23. If a change in the lap is required, it is necessary to perform an electric switch between the recording head 14a and the read head 14b. This switching control is performed by the head position control system 17.

The motor driver 18 drives the motor 15. When two motors 15 are provided as described above, two motor drivers 18 are provided accordingly.

The control mechanism 30 is a mechanism for controlling the accessor 40 and the tape drive 10 in accordance with commands from the host 200. The control mechanism 30 instructs the accessor 40 to load the cartridge 20 into the tape drive 10 such that the data designated by the host 200 is read or written to the cartridge 20. In addition, the control mechanism 30 instructs the tape drive 10 to read or write data designated by the host 200 to the cartridge 20 loaded by the access 40.

Under the control of the control mechanism 30, the accessor 40 takes the cartridge 20 out of the cartridge slot 50 and loads the cartridge 20 into the tape drive 10.

The cartridge slot 50 is a place for loading a cartridge 20 in which a read operation and a write operation are not performed. Although the cartridge slot 50 is represented by a single rectangle in FIG. 1, there are actually a plurality of slots for loading a plurality of cartridges.

Although only one tape drive 10 is shown in FIG. 1, a plurality of tape drives 10 may be provided. In this case, in order to instruct the accessor 40 to load the cartridge 20 into one of the tape drives 10, the control mechanism 30 relates to the tape drive 10 that is the destination of the read command or the write command. The identification information is transmitted to the accessor 40. 1 illustrates a tape drive 10 in which only one cartridge 20 may be loaded, but the tape drive 10 may be a tape drive in which a plurality of cartridges 20 may be loaded.

The performance of access to data recorded on the tape 23 by the tape drive 10 greatly depends on the recording position on the tape 23. Therefore, moving the data to the tape 23 without considering the frequency of access to the data can reduce the speed of access to frequently accessed data. As a result, data access performance may be degraded.

Nowadays, the tape drive 10 is used as a storage device for picture data, video data and the like. Therefore, not only the data (main data) itself, but also metadata serving as additional information about the main data are becoming important as data to be stored for a long time operation. For example, the metadata for the photo data may provide information such as the place where the photo was taken, the date, the time, the photographer's name, the photographic equipment, the recording format, the file name, and the file size. Metadata is closely related to data and is therefore often treated with data. Therefore, when data is stored as archive data on the tape 23 and metadata for the data is stored on the same tape 23, access performance will be degraded unless the data and metadata are properly arranged.

Therefore, in this embodiment, the frequency of access to the pieces of data recorded on the tape 23 in association with each other is recorded and used to perform data movement. Thus, upon data movement, an arrangement suitable for access performance to groups of data and metadata is realized.

Specifically, the following configuration is adopted.

Data and metadata are associated with each other by dm-id. Each time data or metadata is accessed, an access history is written to a group of access counter variables (hereinafter simply referred to as "counters") corresponding to dm-id. Because of the characteristics of the tape, which is a sequential access recording medium, the access history is recorded in the counter group with reference to the number of times the tape 23 has been mounted for reading data without referring to the time axis. For portability of the tape 23, the counter group is written to a semiconductor memory attached to the tape 23.

For data movement from the recording medium A to the recording medium B after the access history is recorded as described above, the access frequency for the data and metadata associated with each other by dm-id in the recording medium A is It is calculated according to the policy based on the counter group. Next, for better access performance, the data and metadata pieces are rearranged on the recording medium B in descending order of the access frequency. The rearrangement method adopted here is to record the sets of associated data and metadata sequentially on the tape 23 from the beginning of the tape 23. Thus, access performance is improved in reading the sets of data and metadata sequentially from the beginning of the tape 23.

Hereinafter, a configuration for realizing the above operation will be described in detail.

First, the association between data and metadata will be described.

2 schematically illustrates how data is associated with metadata.

As illustrated in FIG. 2A, the host 200 first creates an association between data and metadata by assigning an identification number dm-id to the data and metadata. The host 200 then passes data to the control mechanism 30 through the data port and metadata to the control mechanism 30 through the metadata port.

In the control mechanism 30, the information transmitted through the data port is treated as data, and the ID generating unit 31 generates d-id and assigns it to this data. Here, the d-id is an identification number for identifying data and is at least a value unique in an environment in which the storage device 100 operates. In addition, in the control mechanism 30, the information delivered through the metadata port is treated as metadata, and the ID generating unit 31 generates m-id and assigns it to this metadata. Here, the m-id is an identification number for identifying the metadata and is at least a value unique in an environment in which the storage device 100 operates. Even when data and metadata are temporarily transferred separately to the control mechanism 30, the relationship between the data and the metadata may be established by dm-id within the storage device 100.

Therefore, the relationship between the data based on the dm-id and the metadata is recorded by the control mechanism 30 as a dm-a table (hereinafter simply referred to as "dm-a") in the database (DB) 32. 2 (b) illustrates the format of dm-a. In the example of FIG. 2 (b), one piece of data is associated with one metadata, but the data and metadata may exist in a ratio of M and N, where M and N are natural numbers.

In addition, the recording positions of the data and the metadata in the storage device 100 are referred to as dm-l tables (hereinafter simply “dm-l”) by the control mechanism 30 for each of the data and the metadata. ) Is recorded. 2 (c) illustrates the format of dm-1. When a plurality of storage spaces are implemented in the storage device 100 and a plurality of copies of data and metadata exist in the plurality of storage spaces, as many entries as the number of copies are registered in dm-l. In Fig. 2 (c), entries are represented by position-1 and position-2.

Hereinafter, the association between data and metadata will be described using an example.

3 illustrates an example of an association between data and metadata.

Fig. 3 (a) shows a state in which data identified by d-id and metadata identified by m-id have a relationship identified by dm-id and are recorded on a recording medium having a medium name " A001 ". Specifically, each of the data fragments identified by d-id "d3" and the metadata fragments identified by m-id "m4", m-id "m6" and m-id "m8" are each dm-id ". each piece of data identified by d-id " d1 " and pieces of metadata identified by m-id " m1 " and m-id " m7 " The data fragment identified by d-id "d2" and the metadata fragment identified by m-id "m5" have a relationship identified and recorded by id "dm01" and identified by dm-id "dm02". And a data fragment identified by d-id "d5" and a metadata fragment identified by m-id "m9" have a relationship identified and recorded by dm-id "dm04". In the present specification, the combination between data and metadata associated with each other by dm-id is now referred to as "DM".

3 (b) illustrates how the identifiers d-id and m-id of the DM with dm-id “dm03” are listed in dm-a.

Since data and metadata are recorded on the tape 23 at the time of preparation for recording, the order of recording pieces of data and metadata on the tape 23 is not uniquely determined.

Therefore, even if the relationship between data and metadata is logically set by dm-id, the recording positions of pieces of data and metadata on the tape 23 can be physically distributed.

4 illustrates an example of pieces of data and metadata distributed and recorded on the tape 23. As illustrated, a plurality of wraps are present on the tape 23. In Fig. 4, each hollow arrow indicates the direction in which data and metadata are recorded on the tape 23, and each hollow arrow indicates that the running direction of the tape 23 is reversed.

First, metadata with m-id "m6", data with d-id "d3", metadata with m-id "m1", data with d-id "d1", m-id "m4" Metadata with is written to Lab # 0 in the forward direction. Next, the metadata with m-id "m5" and the data with d-id "d2" are recorded in lap # 1 in the reverse direction. Next, metadata with m-id "m7", metadata with m-id "m9", data with d-id "d5", and metadata with m-id "m8" are again wrapped forward. Recorded in # 2.

In other words, even when groups of pieces of data and metadata as illustrated in FIG. Can be arranged regardless of whether they are related. In this case, the DM with the dm-id cannot be read unless the positioning of the tape 23 is performed a plurality of times, so that the access performance is degraded by performing the positioning of the tape 23.

In this embodiment, the recorded data and metadata are rearranged based on the access frequency in movement.

Here, the following three counters are defined for each DM.

(1) Ram-count with read access

Ram-count is a counter defined for each dm-id. Ram-count holds the number of times the tape 23 has been mounted for data reading after pieces of data and metadata associated with each other by dm-id have been written to the tape 23. This embodiment is based on the assumption that whenever a data need to be read from the tape 23 becomes necessary, the tape 23 on which the data is recorded is mounted in the tape drive 10. Hereinafter, mounting the tape 23 for reading is referred to as "read mount operation".

In the definition of Ram-count described above, the dm-id with the largest Ram-count is to associate the metadata with the least recently recorded data on the tape 23. The Ram-count assigned to each dm-id shows which data and metadata are relatively new or old, and also shows the probability of accessing the data and metadata for reading. In order to prevent the wrapping of this counter, the counter needs to be defined to hold a large value. The counter resets to zero when wrapped.

(2) read frequency counter (Dm-readfreq)

Dm-readfreq is also a counter defined for each dm-id. Dm-readfreq holds a history indicating to some extent that data and metadata associated with each other are read by dm-id while the read mount operation is performed the number of times indicated by Ram-count. Dm-readfreq is updated when either data or metadata is read.

For example, the maximum value of Dm-count is 16, i.e., Dm-count holds only the history for the past 16 read mount operations. In this case, Dm-readfreq is represented by 16 bits when each bit of Dm-readfreq indicates whether data or metadata with dm-id has been read.

However, if one bit is allocated to one read mount operation, for example, when it is required to measure the access frequency from the history of past 256 read mount operations, it is necessary to prepare a 32 byte (256 bit) memory area.

Thus, a description will be given of a method of representing a plurality of read mounting operations in one bit.

For example, if eight read mount operations are represented by one bit, only four bytes (256 bits / 8 = 32 bits) of memory area are required. This method involves the use of the dm access counter (described below) as well as dm-readfreq. The dm access counter holds the number of times data with dm-id has been read from the past eight read mount operations.

However, for this method, the counter indicates 1 even when the dm-id is read only once in the past eight read mount operations. This means that data with dm-id has not been read once or eight times in the past eight read mount operations.

Next, a method of expressing a plurality of read mounting operations with a plurality of bits will be described.

For example, when eight read mount operations are represented by two bits, a memory area of eight bytes (= 256 bits / 8x2 = 32 bits x2) is required. This method also includes the use of Dm-readfreq, as well as the use of a dm access counter (described below) that holds the number of times data with dm-id has been read in the past eight read mount operations.

In this method, for example, the access frequency is expressed in 2 bits as follows:

"00": no access

"01": 1 to 2 times

"10": 3 to 5 times

"11": 6 to 8 times

These definitions may vary depending on the frequency calculation policy.

(3) Dm Access Counter (Dm-count)

Dm-count is a counter that holds the number of times the DM with dm-id has been read in M read-read operations, expressed as N (<M) bits as a unit in Dm-readfreq.

For example, when the past eight read mount operations are grouped together as a unit, if the data with dm-id is read in each of all the past eight read mount operations, Dm-count represents eight. If data with dm-id is read twice, Dm-count represents two.

After a unit of read mount operations such as eight read mount operations of the tape 23, Dm-count is reset to zero. Here, Ram-count is used to count 8 read mount operations grouped together as a unit. That is, Dm-count is reset to zero after the number of read mount operations reaches a multiple of eight.

Using the three counters described above, information as listed below can be obtained:

1. the frequency with which the DM with the target dm-id was read in past X read mount operations of the tape 23;

2. The number of times a DM with the target dm-id has been read;

3. The access probability calculated based on the number of times the DM with the target dm-id was read in past X read mount operations of the tape 23.

4. The propensity of access in X read-read operations, i.e., whether the data has been accessed recently or frequently in the past.

Typically, the frequency of access to data is determined with reference to the time base. However, if a removable recording medium such as a tape is ejected from the recording apparatus or there is no access to the tape in the recording apparatus, some "time stops". Therefore, in this embodiment, the frequency of access to the data is determined based on the number of times the tape has been mounted for data reading without considering the time base.

The factors that define the size of the three counters described above are:

1. The number of past read mounting operations to be targeted.

2. How many read mount operations are represented by how many bits.

For example, as described above, when past 256 read mount operations are used as a target to measure the frequency of access, the Ram-count is represented by 1 byte.

When eight read mount operations are grouped together in one unit and represented by two bits, dm-readfreq is represented by eight bytes and dm-count is represented by four bits.

5 illustrates three counters in the case described above.

Since three counters are provided for each dm-id, the dm-id is also illustrated in FIG.

Note that the number of bits allocated to dm-id is not specified above, so dm-id is simply represented as a rectangle.

On the other hand, since the number of bits assigned to each of the three counters is specified, one bit is associated with one narrow rectangle and represented by that rectangle. Specifically, Ram-count requiring an 8-bit memory area is represented by eight rectangles. Dm-count, which requires a 4-bit memory area, is represented by four rectangles, and four rectangles are represented by shading four of the eight rectangles. Shading indicates that 4 bits represented by 4 rectangles are not used. Dm-readfreq, which requires a 64-bit memory area, is represented by 64 rectangles. In particular, in Dm-readfreq, eight read mount operations are grouped together as one unit and represented by two bits. Thus, FIG. 5 shows the meaning represented by 2 bits of the right end of Dm-readfreq.

Next, the values of the three counters are stored for each dm-id in the dm access counter table.

6 illustrates an example of a dm access counter table.

In FIG. 6, the Ram-count column was performed after 215 read mount operations were recorded with a DM with dm-id "dm01", and after 24 read mount operations were recorded with a DM with dm-id "dm02". 13 read mount operations were performed after the DM with dm-id "dm03" was recorded, 156 read mount operations were performed after the DM with dm-id "dm04" was recorded, and 79 reads It shows that the mounting operation was performed after the DM with the dm-id "dm05" was recorded.

Also in FIG. 6, the Dm-count column shows that a DM with dm-id "dm01" and a DM with dm-id "dm02" have never been read in the past eight read mount operations, and have a dm-id "dm03". The DM has been read four times in the past eight read mount operations, the DM with the dm-id "dm04" has been read once in the past eight read mount operations, and the DM with the dm-id "dm05" has read the past eight times. It is read 7 times in the mounting operation. The number of times described above is reflected in the two bits of the right end of Dm-readfreq for each dm-id.

In the Dm-readfreq column of Fig. 6, the read of the DM is represented by 8 bits. Since the number of times the DM has been read in eight read mount operations is represented by 2 bits, the number of times the DM has been read in 25 to 32 read mount operations is shown in FIG. A series of dots "....." to the left of each bit string indicate that the bit values are omitted. For dm-id "dm02", where ram-count represents 24, 6 bits are sufficient to represent Dm-readfreq. Similarly, for dm-id " dm03 " where Ram-count represents 13, 4 bits are sufficient to represent Dm-readfreq. However, in these two cases, Dm-readfreq is represented by 8 bits filled with zeros on the left.

Next, the update of three counters is demonstrated with reference to an example.

FIG. 7 illustrates how each counter loaded in the cartridge 20 is updated when the tape tribe 10 receives a write command or a read command. Although not mentioned in the above description, the total read counter is also illustrated in FIG. 7. The total read counter is a counter for each tape 23 rather than each dm-id and is updated when the tape 23 is mounted for reading. In FIG. 7, Dm-readfreq is omitted. In FIG. 7, each of D1 to D3 represents data that is a generalized representation of DM.

As illustrated in FIG. 7A, when the write command for D1 is received, the total read counter indicates zero. Ram-count and Dm-count for D1 are generated, but their values are all zero.

Next, as illustrated in FIG. 7B, when the read command for D1 is received, the total read counter is updated to one. Ram-count and Dm-count for D1 are also updated to one.

Next, as illustrated in FIG. 7C, when a read command for D1 is received, the total read counter is updated to two. Ram-count and Dm-count for D1 are also updated to one.

As illustrated in D of FIG. 7, when a write command for D2 is received, the total read counter remains two. Ram-count and Dm-count for D1 also hold 2. Ram-count and Dm-count for D2 are generated, but their values are zero.

Next, as illustrated in E of FIG. 7, when a read command for D1 is received, the total read counter is updated to three. Ram-count for D1 is also updated to three. Dm-count for D1 is also updated to 3 because D1 has been read. On the other hand, the Ram-count for D2 is updated to 1, but the Dm-count for D2 remains 0, because D2 has not been read.

Next, as illustrated in F of FIG. 7, when a read command for D2 is received, the total read counter is updated to four. On the other hand, the Ram-count for D1 is updated to 4, but the Dm-count for D1 remains 3 because D1 has not been read. Ram-count for D2 is updated to 2. Dm-count for D2 is updated to 1 because D2 has been read.

Next, when a write command or a read command is received as illustrated in G, H, and I of FIG. 7, the Ram-count and Rd-count for D1, D2, D3, as well as the total read counter, are described above. It is updated in the same way as it was.

Next, the rearrangement performed based on the access frequency when the data and the metadata are moved to the other tape 23 will be described.

If data and metadata on the tape 23 are moved to another tape 23 (or the small tm tape 23 itself), the dm-id in descending order of access frequency shown in the dm access counter table of FIG. 6. The data and the metadata corresponding to are recorded on the tape 23 at advantageous positions in terms of performing access.

The most basic method of determining the access frequency is to calculate the Dm-count / total read counter (hereinafter, this determination method is referred to as the "basic method"). However, if this basic method is used when determining the frequency of access, the result depends on the time when the DM was recorded on the tape 23. For example, the DM recorded first on the tape 23 cannot be simply compared with the DM recorded on the tape 23 after the Ram-count reaches 100.

Therefore, this embodiment adopts the following methods of determining the access frequency.

(First decision method)

The first determination method is a method of determining the relative frequency of access to the DM by calculating Dm-count / Ram-count.

The first determination method will be described with reference to the example illustrated in FIG. 6. Here, it is assumed that zeros exist as bit values of the omitted portion "....." of Dm-readfreq.

If the default method is used and the total read counter indicates 250, then the access frequency for DM with "dm01" is 0.02 (= 5/250), the access frequency for DM with "dm02" is 0, and "dm03 The access frequency for DM with "is 0.012 (= 3/250), the access frequency for DM with" dm04 "is 0.016 (= 4/250), and the access frequency for DM with" dm05 "is 0.016 (= 4/250). That is, the descending order of access frequencies is dm01, dm05, dm04, dm03, and dm02. If the resulting values are the same, the more recently accessed DM is defined to be the more frequently accessed DM.

On the other hand, when the first determination method is used, the access frequency for the DM with "dm01" is 0.023 (= 5/215), the access frequency for the DM with "dm02" is 0, and "dm03" The access frequency for the DM with 0.231 (= 3/13), the access frequency for the DM with "dm04" is 0.026 (= 4/156), and the access frequency for the DM with "dm05" is 0.051 (= 4/79). That is, the descending order of access frequencies is dm03, dm05, dm04, dm01, and dm02. This order is different from the order obtained by the basic method.

(Second decision method)

The second determination method is a method in which the Ram-count retained for each DM uses the access frequency as a criterion for selecting a DM to be calculated.

For example, a recently recorded DM can be removed from DMs for which the frequency of access is to be calculated. Specifically, for example, DM having a Ram-count of 3 or less can be eliminated.

Alternatively, it may be desired to measure how frequently the DM has been accessed during a particular time period, such as from the past x-y th read mounting operation. In this case, for example, the DM whose access frequency is to be calculated is limited to DMs having a Ram-count in the range of 100 to 70. Next, the number of times each DM has been accessed in the period from the 100th read mount operation to the 70th read mount operation is calculated from Dm-readfreq. By comparing the resulting times, the corresponding portions of the DM can be arranged in descending order of access frequency.

The range of Ram-counts considered in the first determination method described above is an example of a predetermined range for limiting targets for which the access frequency is to be calculated.

When the access frequency is calculated as described above, the tape drive 10 moves data from one tape to another based on the access frequency.

8 illustrates an example of how to arrange data and metadata on the tape 23 after data movement. As in the case of FIG. 4, a plurality of wraps are present on the tape 23. Further, each hollow arrow indicates the direction in which data and metadata are recorded on the tape 23, and each hollow arrow indicates that the running direction of the tape 23 is reversed. In this example, the descending order of access frequencies is as follows: dm02, dm03, dm01, dm04.

First, the metadata and data of the DM with the dm-id "dm02" are written to Lab # 0 in this order forward. Next, the metadata and data of the DM having the dm-id " dm03 " are written to Lab # 1 forward in this order. Next, the metadata of the DM having the dm-id " dm01 " is recorded in reverse in Lab # 1. Next, the metadata and data of the DM having the dm-id "dm04" are written to Lab # 2 in this order forward.

That is, after data movement, unlike in the case of FIG. 4, data and metadata are recorded based on the logical association illustrated in FIG. Therefore, when data and metadata are read sequentially from the tape 23, there is no need to perform unnecessary positioning of the tape 23 and improved access performance can be achieved.

In the data movement method described above, the data pieces are arranged in accordance with the access frequency calculated in this embodiment. Therefore, the time required to access data that is frequently accessed in practice can be reduced.

In addition, since data and metadata arranged in a distributed manner on the tape 23 before moving are arranged according to a predetermined rule (in this embodiment, sequentially), access performance to the data and metadata can be improved.

As a method of implementing a hierarchical storage device for storing a large amount of data, a storage system for storing digital data may adopt a method of using the tape 23 as a second or third storage device. This is based on the assumption that there is actually stored but frequently accessed data that is accessed. This embodiment also contributes to improved access performance for such data.

In the example described above, data and metadata distributedly arranged on a tape 23 (hereinafter referred to as "source tape 23") serving as a moving source are grouped together for each dm-id, It is sequentially stored on the tape 23 (hereinafter referred to as "destination tape 23") serving as a moving destination. However, the method of arranging on the destination tape 23 is not limited to this.

Now let's talk about another arrangement. In the arrangement method to be described herein, the data is rearranged on the tape 23 due to the characteristics of the tape medium. Specifically, in this arrangement method, frequently accessed data is arranged in an easily accessible position after the tape 23 is mounted.

Characteristics of the tape storage device include the following operations performed as a data readiness preparation.

1. Insert the cartridge into the storage unit.

2. Adjust the beginning of the tape 23 to the position of the head of the storage device, where the beginning is the position where the data to be read is stored.

That is, immediately after mounting of the cartridge, the head of the storage device is located at the beginning of the tape 23. Therefore, the access speed for data located near the beginning of the tape 23 is faster than the access speed for data located near the end of the tape 23. This is because it is necessary to move the tape 23 to perform positioning of the tape 23 when data is recorded near the end of the tape 23.

9 (a) illustrates a normal reading and writing of data for tape 23 based on an open linear tape-open (LTO) standard.

First, after reading or writing data to the right with respect to the lap # 1 as indicated by the arrow 201, the running direction of the tape 23 is reversed as indicated by the arrow 202. Next, after reading or writing data left to lap # 1 to the left, the running direction of the tape 23 is reversed as indicated by the arrow 203. Next, after reading or writing data to the right with respect to the lap # 2, the running direction of the tape 23 is reversed as indicated by the arrow 204. Finally, as indicated by arrow 205, data is read or written to left for lap # 55.

As such, in the LTO, the head reads or writes data to and from the tape 23 while reciprocating from the beginning to the end of the tape 23 on a plurality of data storage areas called wraps defined on the tape 23. do. When the data pieces are read or written sequentially to the tape 23, the head reciprocates on the lap as described above. However, the mechanical operation in which the head moves across the width of the tape 23 for the transition from one wrap to another is completed in a short time. Therefore, even in this LTO format, data located near the beginning of the tape 23 can be accessed faster than data located near the end of the tape 23.

Thus, data that is frequently accessed is arranged near the beginning of the tape 23 and data that is less frequently accessed is preferably arranged near the end of the tape 23. Here, the start of the tape 23 is an example of the first end, and the end of the tape 23 is an example of the second end.

Some tape storage devices implement a function called segmentation. The division is a method of dividing the tape 23 into a plurality of segments and sequentially recording data on the tape starting from the first segment. Here, the segment is an exemplary area formed by dividing the tape 23 at at least one position in the longitudinal direction.

9 (b) illustrates the reading and writing of data performed using partitioning.

First, after reading or writing data to the right for segment # 1 of lap # 0 as indicated by arrow 211, the running direction of tape 23 is reversed as indicated by arrow 212. FIG. Next, after reading or writing to the left for segment # 1 of lap # 1, the running direction of the tape 23 is reversed as indicated by arrow 213. Next, after reading or writing the data to the right for segment # 1 of lap # 2, the running direction of the tape 23 is reversed as indicated by arrow 214. Next, as indicated by arrow 215, data is read or written to left for segment # 1 of lap # 55.

Next, as indicated by arrow 220, the position of the head moves to segment # 2 of lap # 0. Next, after reading or writing data to the right for segment # 2 of lap # 0, as indicated by arrow 221, the running direction of tape 23 is reversed, as indicated by arrow 222. FIG. Next, after reading or writing to the left for segment # 2 of lap # 1, the running direction of the tape 23 is reversed as indicated by arrow 223. Next, after reading or writing data to the right of segment # 2 of lap # 2, the running direction of the tape 23 is reversed as indicated by arrow 224. Next, as indicated by arrow 225, the data is read or written to the left for segment # 2 of lap # 55.

As described above, the access speed for segment # 1 is faster than the access speed for segment # 2. When there are three or more segments, the access speed for the segment closer to the beginning of the tape 23 is generally faster than the access speed for the segment closer to the end of the tape 23.

Therefore, when data is moved to tape 23 using the characteristics described above, frequently accessed data and metadata are arranged in segments closer to the beginning of tape 23, and less frequently accessed data. And the metadata is preferably arranged in a segment closer to the end of the tape 23. Here, the segment closer to the beginning of the tape 23 is an example of the first end, and the segment closer to the end of the tape 23 is an example of the second end.

FIG. 10 is an example of data and metadata arranged on the tape 23 after data movement performed using division. As in the case of FIG. 4, a plurality of wraps are present on the tape 23. Further, each hollow arrow indicates the direction in which data and metadata are recorded on the tape 23, and each hollow arrow indicates that the running direction of the tape 23 is reversed. In this example, the descending order of access frequencies is as follows: dm02, dm03, dm01, dm04.

First, the metadata and data of the DM with the dm-id "dm02" are written to segment # 1 of lap # 0 in this order forward. Next, the rest of the data of the DM with the dm-id " dm02 " is recorded backward in segment # 1 of lap # 1. Next, the metadata and data of the DM having the dm-id "dm03" are recorded in this order forward in segment # 1 of the lab # 2. If there is an additional part of the data with dm-id "dm03", that additional part is recorded in segment # 1 (not shown) of lap # 3.

Next, the metadata and data of the DM having the dm-id " dm01 " are recorded in this order forward in segment # 2 of lap # 0. Next, the metadata and data of the DM having the dm-id " dm04 " are written in reverse order in this order to segment # 2 of lap # 1.

That is, after data movement, unlike in the case of FIG. 4, data and metadata are recorded based on the logical associations illustrated in FIG. Therefore, when data and metadata are read sequentially from the tape 23, there is no need to perform unnecessary positioning of the tape 23 and improved access performance can be achieved.

Although only one destination tape 23 exists here, a plurality of tapes 23 can serve as a destination of data movement. In the latter case, segment # 1 of the first destination tape 23 is used first. After segment # 1 of the first destination tape 23 is exhausted, segment # 1 of the second destination tape 23 is used. After segment # 1 of the second destination tape 23 is exhausted, segment # 1 of the third destination tape 23 is used. This process is repeated until segment # 1 of all destination tapes 23 is exhausted. After segment # 1 of all destination tapes 23 are exhausted, segment # 2 of first destination tape 23 and subsequent segments of destination tape 23 may be used sequentially.

As such, by preparing the plurality of destination tapes 23 as described above, it is possible to provide more memory space and increase the amount of data and metadata that can be written to segment # 1 with high access performance. It is possible. This means that the amount of data and metadata that can be accessed in a short time is increased.

Also in the case of this method, by changing the number of destination tapes 23 and the manner of defining segments, it is possible to provide appropriate memory space depending on the frequency of accessing data on the destination tapes 23. For example, when only a small amount of frequently accessed data is present, the first segment may be defined to be located near the beginning of the tape 23. When there is a large amount of frequently accessed data, the number of destination tapes 23 can be increased, or the first segment can be largely defined if necessary. If the access frequency varies, it may be possible to increase the number of segments so that data pieces can be grouped together based on the access frequency and written to the appropriate segments.

In the data movement method described above, the data pieces are arranged in accordance with the access frequency calculated in this embodiment. Therefore, it is possible to improve access performance for individual pieces of data that are frequently accessed in practice.

As a method of implementing a hierarchical storage device for storing a large amount of data, a storage system for storing digital data may adopt a method of using the tape 23 as a second or third storage device. This is based on the assumption that there is data that is actually accessed, archived but frequently accessed. Under circumstances in which such archived data is accessed individually, the present embodiment also contributes to improved access performance for data that is frequently accessed in the past.

A data and metadata arrangement method based on access performance in reading data and metadata will be described. However, an arrangement method that is not based on access performance in reading data and metadata may be adopted.

Next, the tape drive 10 that performs the operation described above will be described in detail. Here, the dm access counter table is stored in the cartridge memory 24. 1 illustrates a tape drive 10 in which only one cartridge 20 can be loaded. However, in the example described below, a tape drive 10 is used in which two cartridges 20 can be loaded. In this case, since data can be moved from one cartridge 20 to another within one tape drive 10, there is no need to replace the cartridge 20.

11 is a block diagram illustrating an exemplary functional configuration of the controller 16 of the tape drive 10.

As illustrated, the controller 16 is referred to as the command processing unit 61, the buffer management unit 62, the channel input / output unit 63, the cartridge memory input / output unit (hereinafter referred to as "CM input / output unit"). (64), the frequency calculating unit 65, the movement control unit 66, the head position management unit 67, and the tape travel management unit 68.

The command processing unit 61 receives a command from the host I / F 11. Examples of the command include a write command to store data in the buffer 12, a synchronization command to write data stored in the buffer 12 to the tape 23, and a read command to read data from the tape 23. have.

When the command processing unit 61 receives the recording command, the buffer management unit 62 prepares data in the buffer 12. When the command processing unit 61 receives the synchronization command, the buffer management unit 62 reads data from the buffer 12 and outputs the data to the channel input / output unit 63. When the instruction processing unit 61 receives a read command, if no corresponding data exists in the buffer 12, the buffer management unit 62 instructs the channel input / output unit 63 to read the corresponding data. . If the corresponding data exists in the buffer 12, the buffer management unit 62 provides the corresponding data to the host 200 via the command processing unit 61.

The channel input / output unit 63 outputs the data read from the buffer 12 by the buffer management unit 62 to the channel 13. In addition, the channel input / output unit 63 outputs the data received from the channel 13 to the buffer management unit 62.

The CM input / output unit 64 writes the information to the dm access frequency counter table stored in the cartridge memory 24 via the CM I / F 19 and from the dm access counter table to the information via the CM IF 19. Read it. In this embodiment, the CM input / output unit 64 includes first information indicating the number of times any data has been read from the first recording medium after each data has been recorded on the first recording medium, and data from the first recording medium. Is provided as an example of the acquiring unit for acquiring the second information indicating the number of times that is read, and also as an example of the updating unit for updating the first information and the second information.

Based on the information read by the CM input / output unit 64, the frequency calculating unit 65 calculates the access frequency for each DM. In this embodiment, the frequency calculating unit 65 is provided as an example of the tracking unit that tracks the frequency of access to each data.

The movement control unit 66 controls the data movement based on the access frequency calculated by the frequency calculating unit 65. That is, the movement control unit 66 reads the pieces of data and metadata from the source tape 23 in the descending order of the access frequency to the buffer management unit 62 and transfers the pieces of data and metadata to the destination tape 23. Instruct them to record. As such, pieces of data and metadata are sequentially recorded on the destination tape 23 in descending order of access frequency. This means that the movement control unit 66 determines the location of the pieces of data and metadata respectively on the destination tape 23. That is, in this embodiment, the movement control unit 66 is provided as an example of the determination unit that determines the recording position of each data in the second recording medium.

The head position management unit 67 outputs a signal to the head position control system 17 to offset the positions of the recording head 14a and the read head 14b relative to the tape 23 in the width direction of the tape 23. do. In addition, the head position management unit 67 acquires information about the current positions of the recording head 14a and the reading head 14b in the width direction of the tape 23.

The tape travel management unit 68 outputs a signal for driving the tape 23 in the forward direction and a signal for driving the tape 23 in the reverse direction to the motor driver 18.

Next, the operation of the tape drive 10 in this embodiment will be described in detail.

First, a description will be given of how the tape drive 10 operates when a write command is transmitted from the host 200.

The host 200 transmits a recording command with dm-id that associates the data with the metadata to the storage device 100. In the storage device 100, the control mechanism 30 generates a d-id identifying data and an m-id identifying metadata, and registers the d-id as dm-a in association with the m-id. Alternatively, the control mechanism 30 determines the cartridge 20 to which data and metadata are to be recorded, registers d-id and m-id with the cartridge 20 as dm-l and accessor 40 Instructs the tape drive 10 to load the cartridge 20. The control mechanism 30 then sends a write command to the tape drive 10 along with the dm-id, d-id, m-id, data and metadata. Thus, the operation of the tape drive 10 is started.

12 is a flow chart illustrating exemplary operations performed here by the controller 16 of the tape drive 10.

In the controller 16, first, the command processing unit 61 receives a write command together with dm-id, d-id, m-id, data and metadata through the host I / F 11 (step S601). ). Next, the dm-id, d-id, m-id, data and metadata are delivered to the buffer management unit 62 and stored in the buffer 12 by the buffer management unit 62.

Next, the controller 16 writes data and metadata to the tape 23 in the cartridge 20 loaded in the tape drive 10 (step S602).

Specifically, in accordance with an instruction from the buffer management unit 62, the head position management unit 67 controls the head position control system 17 to adjust the position of the recording head 14a in the width direction of the tape 23. do. In addition, in response to an instruction from the buffer management unit 62, the tape travel management unit 68 controls the motor driver 18 to drive the tape 23. At the same time, the buffer management unit 62 reads data and metadata from the buffer 12 and transfers the data and metadata to the channel input / output unit 63, and the channel input / output unit 63 is connected to the channel 13. The data and the metadata are output to the recording head 14a via the &quot;

The channel input / output unit 63 receives information regarding the recording position of data and metadata in the longitudinal direction of the tape 23 from the servo head via the channel 13. The head position management unit 67 receives information about the recording position of data and metadata in the width direction of the tape 23 from the head position control system 17. The information about these recording positions is transmitted to the buffer management unit 62, and the buffer management unit 62 inputs / outputs d-id and m-id read out from the buffer 12 and information about these recording positions. Output to unit 64. Next, the CM input / output unit 64 transmits information on the correspondence relationship between the d-id and the recording position of the data and the information on the correspondence relation between the m-id and the recording position of the metadata to the cartridge memory 24. Record via I / F 19. Accordingly, data can be read using d-id as a key and metadata can be read using m-id as a key.

Next, in the dm access counter table stored in the cartridge memory 24, the controller 16 generates an entry corresponding to dm-id (step S603).

Specifically, the buffer management unit 62 reads the dm-id from the buffer 12 and delivers it to the CM input / output unit 64, and the CM input / output unit 64 sends the CM I / F 19. The record containing the dm-id is written to the cartridge memory 24 through the recording medium.

Next, a description will be given of how the tape drive 10 operates when a read command is sent from the host 200. Here, whenever a request to read data and metadata is received from the host 200, a cartridge 20 in which the data and metadata are to be read is mounted in the tape drive 10.

The host 200 transmits a read command to the storage device 100 together with the dm-id. In the storage device 100, based on dm-a, the control mechanism 30 identifies the d-id and m-id associated with each other by dm-id. Also based on dm-l, control mechanism 30 identifies cartridge 20 associated with d-id and m-id. The control mechanism 30 instructs the accessor 40 to load the cartridge 20 identified in the tape drive 10. The mechanism 30 then sends a read command to the tape drive 10 along with the dm-id, d-id, and m-id. Thus, the operation of the tape drive 10 is started.

13 is a flowchart illustrating an example operation performed here by the controller 16 of the tape drive 10.

In the controller 16, first, the command processing unit 61 receives a read command together with dm-id, d-id, and m-id via the host I / F 11 (step S621). Next, dm-id, d-id, and m-id are passed to the buffer management unit 62.

Next, the controller 16 reads the data corresponding to the d-id and the metadata corresponding to the m-id from the tape 23 of the cartridge 20 loaded in the tape drive 10 (step S622).

Specifically, in accordance with an instruction from the buffer management unit 62, the head position management unit 67 controls the head position control system 17 to adjust the position of the read head 14b in the width direction of the tape 23. do. In addition, in response to an instruction from the buffer management unit 62, the tape travel management unit 68 controls the motor driver 18 to drive the tape 23.

At the same time, the channel input / output unit 63 acquires the data and metadata read by the read head 14b via the channel 13, and transfers the obtained data and metadata to the buffer management unit 62. The buffer management unit 62 stores data and metadata in the buffer 12. Next, the data and metadata stored in the buffer 12 are transmitted to the host 200.

The controller 16 then updates the dm access counter table stored in the cartridge memory 24. The update process begins when the CM input / output unit 64 reads the dm access counter table from the cartridge memory 24 through the CM I / F 19 into the memory area of the CM input / output unit 64.

In the update process, first, the CM input / output unit 64 reads one record from the dm access counter table (step S623). Next, the CM input / output unit 64 determines whether the Ram-count included in the read record is a multiple of eight (step S624). If Ram-count is a multiple of eight, this read mount operation is the beginning of a new set of eight read mount operations performed as a unit. Next, the CM input / output unit 64 assigns 0 to Dm-count to reset Dm-count, and adds "00" to the right end of Dm-readfreq (step S625). On the other hand, if Ram-count is not a multiple of 8, the process goes to the next step.

Next, the CM input / output unit 64 adds 1 to the Ram-count included in the read record (step S626). Next, the CM input / output unit 64 determines whether the dm-id included in the read record matches the dm-id specified by the read command (step S627). If the dm-id contained in the read record matches that specified by the read command, the CM input / output unit 64 adds 1 to Dm-count, and according to the addition result, two bits at the right end of Dm-readfreq. Is updated (step S628). For example, the relationship between the value of Dm-count and two bits of Dm-readfreq is as shown in FIG. In this case, if Dm-count obtained by adding 1 is 0, 2 bits of the right end of Dm-readfreq are updated to "00", and if Dm-count obtained by adding 1 is 1 or 2, Dm The two bits at the right end of -readfreq are updated to "01", and if the Dm-count obtained by adding 1 is any of 3 to 5, the two bits at the right end of Dm-readfreq are updated to "10". If the Dm-count obtained by adding 1, is any one of 6 to 8, the two bits of the right end of Dm-readfreq are updated to "11". Rules in which Dm-readfreq is updated according to the value of Dm-count may be coded directly in the program, or may be defined in an external table of the program so that the program can obtain the rules by referencing the table.

Next, the CM input / output unit 64 determines whether the record read from the dm access counter table is the last record (step S629). If the record is not the last record, steps S623 to S628 are repeated for the next record. If the record is the last record, the process ends.

Next, at some point when the dm access counter table is updated as described above, the tape drive 10 performs processing for data movement.

14 is a flowchart illustrating an example operation performed here by the controller 16 of the tape drive 10. The following description is based on the assumption that data and metadata pieces are arranged on the destination tape 23 in the same manner as illustrated in FIG. 10.

In the tape drive 10, first, the CM input / output unit 64 reads the dm access counter table from the cartridge memory 24 via the CM I / F 19, and stores the dm access counter table in the frequency calculation unit ( 65) (step S641).

The frequency calculating unit 65 calculates the access frequency for each dm-id using the first or second determination method described above, rearranges the records in the dm access counter table in descending order of access frequency, After the rearrangement, the dm access counter table is transferred to the movement control unit 66 (step S642).

Thus, the movement control unit 66 moves the data from one tape 23 to the other tape 23.

That is, the movement control unit 66 reads out one record from the dm access counter table after rearrangement, and transfers the read record to the buffer management unit 62 (step S643). The buffer management unit 62 reads out data and metadata corresponding to the dm-id included in the read record from the source tape 23 (step S644).

Specifically, in accordance with an instruction from the buffer management unit 62, the head position management unit 67 controls the head position control system 17 to adjust the position of the read head 14b in the width direction of the tape 23. do. In addition, in response to an instruction from the buffer management unit 62, the tape travel management unit 68 controls the motor driver 18 to drive the tape 23. At the same time, the channel input / output unit 63 acquires the data and metadata read by the read head 14b through the channel 13, and transfers the obtained data and metadata to the buffer management unit 62. The buffer management unit 62 stores data and metadata in the buffer 12. Note that the d-id and m-id used as keys for reading the data and metadata corresponding to the dm-id can be identified, for example, by referring to the control mechanism 30.

Next, the buffer management unit 62 determines whether data and metadata can be recorded in the current lap on the destination tape 23 (step S645). Specifically, the channel input / output unit 63 receives the information regarding the recording position of the data and the metadata in the longitudinal direction of the tape 23 from the servo head via the channel 13, and thus the buffer management unit 62. Makes this determination based on the information received by the channel input / output unit 63.

If the buffer management unit 62 determines that data and metadata can be recorded in the current lap, the data and metadata are recorded on the destination tape 23 (step S649).

Specifically, in accordance with an instruction from the buffer management unit 62, the head position management unit 67 adjusts the head position control system 17 to adjust the position of the recording head 14a in the width direction of the tape 23. To control. In addition, in response to an instruction from the buffer management unit 62, the tape travel management unit 68 controls the motor driver 18 to drive the tape 23. At the same time, the buffer management unit 62 reads data and metadata from the buffer 12, transfers the data and metadata to the channel input / output unit 63, and the channel input / output unit 63 stores the data and the metadata. The metadata is output to the recording head 14a via the channel 13.

On the other hand, if it is determined that data and metadata cannot be recorded in the current lap, it is determined whether the current lap is the last lap (step S646). Specifically, since the head position management unit 67 receives information about the recording position of data and metadata in the width direction of the tape 23 from the head position control system 17, the buffer management unit 62 is the head. This determination is made based on the information received by the position management unit 67.

If the current lap is not the last lap, the lap is switched to another lap (step S647), and data and metadata are recorded on the destination tape 23 (step S649). The switching of the lap is performed by the head position control system 17 under the control of the head position management unit 67.

If the current lap is the last lap, the lap is switched to lap # 1 of the next segment (step S648), and data and metadata are recorded on the destination tape 23 (step S649). The switching of the segments is performed by the motor driver 18 under the control of the tape travel management unit 68, and the switching of the lap is performed by the head position control system 17 under the control of the head position management unit 67.

Next, the movement control unit 66 determines whether the record read from the dm access counter table is the last record (step S650). If the record is not the last record, steps S643 to S649 are repeated for the next record. If the record is the last record, the process ends.

In the exemplary operation described above, data movement is performed in a tape drive 10 in which two cartridges 20 can be loaded simultaneously, thereby eliminating the need to replace one cartridge 20 with another. Alternatively, control of this data movement can be performed by the control mechanism 30. In this case, the control mechanism 30 can control the tape drive 10 to perform data movement, and the two tape drives 10 can each accommodate only one cartridge 20.

The present invention can be realized only in hardware, only in software, or in hardware and software. The invention can be implemented as a computer, a data processing system, and a computer system. The computer program can be stored and controlled in a computer readable medium. The medium may be an electronic medium, a magnetic medium, an optical medium, an electromagnetic medium, an infrared medium, or a semiconductor system (or apparatus or device), or a propagation medium. Examples of computer readable media include semiconductor or solid state memory, magnetic tape, removable computer diskettes, random access memory (RAM), read-only memory (ROM), rigid magnetic disks, and optical disks. Current examples of optical disks include compact-disk read-only memory (CD-ROM), compact disk-read / write (CD-R / W), and digital versatile disk (DVD).

Although the present invention has been described using the above embodiments, the technical scope of the present invention is not limited thereto. It will be apparent to those skilled in the art that various modifications and alternatives may be employed without departing from the spirit and scope of the invention.

1 is a block diagram illustrating a configuration of a storage device to which an embodiment of the present invention is applied.

2 (a) to 2 (c) is a view for explaining the association of dm-id, d-id, m-id in accordance with an embodiment of the present invention.

3 (a) and 3 (b) is a view showing an example of the association of dm-id, d-id, m-id in accordance with an embodiment of the present invention.

4 illustrates a state in which pieces of data and metadata are arranged distributed on a tape prior to data movement in one embodiment of the present invention.

5 is a diagram for explaining counters used in one embodiment of the present invention;

6 illustrates an example of a dm access counter table used in an embodiment of the present invention.

7 is a diagram for explaining updating of counters in one embodiment of the present invention;

8 illustrates a state in which pieces of data and metadata are arranged on a tape in descending order of access frequency after data movement in accordance with one embodiment of the present invention.

9 (a) and 9 (b) are diagrams for explaining easily accessible portions and not readily accessible portions of a tape;

FIG. 10 illustrates a modified state in which pieces of data and metadata are arranged on a tape in descending order of access frequency after data movement in one embodiment of the present invention. FIG.

11 is a block diagram illustrating an exemplary functional configuration of a controller in accordance with an embodiment of the present invention.

12 is a flowchart illustrating an exemplary operation performed by a controller when a write command is received in accordance with one embodiment of the present invention.

FIG. 13 is a flow chart illustrating exemplary operations performed by a controller when a read command is received in accordance with one embodiment of the present invention. FIG.

14 is a flow chart illustrating exemplary operations performed by a controller after moving data in accordance with one embodiment of the present invention.

* Explanation of symbols for the main parts of the drawings

10: tape drive

11: host I / F

12: buffer

13: channel

14a: recording head

14b: readhead

15: motor

16: controller

17: head position control system

18: motor driver

19: CM I / F

20: cartridge

30: control mechanism

40: Accessor

50: cartridge slot

100: storage device

200: host

Claims (10)

An apparatus for supporting movement of pieces of data recorded on a first recording medium to a second recording medium, the apparatus comprising: For each of the pieces of data recorded on the first recording medium, the first piece of information and the data pieces indicating the number of times any data is read from the first recording medium after the pieces of data have been recorded on the first recording medium. An acquisition unit for acquiring second information indicating the number of times of reading from the first recording medium; A tracking unit for tracking the frequency of access to the data pieces using the first information and the second information obtained by the acquiring unit; And And a determining unit for determining a recording position of the piece of data in the second recording medium based on the access frequency tracked by the tracking unit. According to claim 1, And an updating unit for updating the first information for the piece of data and the second information for the specific data when specific data has been read from the first recording medium. According to claim 1, The first information indicates the number of times the first recording medium is mounted on the storage unit after the data piece has been recorded on the first recording medium, so that the first recording after the data piece has been recorded on the first recording medium Device indicating the number of times certain data has been read from the medium. According to claim 1, And the tracking unit tracks the ratio of the number of times indicated by the second information and the number of times indicated by the first information as an access frequency. According to claim 1, And the tracking unit tracks the access frequency using the second information only for data whose number of times indicated by the first information is within a predetermined range. According to claim 1, The determining unit is further configured to determine the data in the second recording medium such that frequently accessed data is recorded at the first recording position in the second recording medium and less frequently accessed data is recorded at the second recording position in the second recording medium. Determine a recording position of the piece, wherein the second recording position is a position where a time required for reading data from the second recording medium is longer than at the first recording position. An apparatus for supporting rearrangement of pieces of data in a recording medium, For each of the pieces of data recorded on the recording medium, first information indicating the number of times any data has been read from the recording medium after the data pieces have been recorded on the recording medium and the data pieces read from the recording medium. An acquiring unit for acquiring second information indicating the number of times; A tracking unit for tracking the frequency of access to the data pieces using the first information and the second information obtained by the acquiring unit; And And a determining unit for determining the recording position of the piece of data in the recording medium based on the access frequency tracked by the tracking unit.  An apparatus for moving pieces of data recorded on a first tape medium to a second tape medium, For each of the pieces of data recorded on the first tape medium and including metadata indicative of the main data and information associated with the main data, for reading some data after the data piece has been written to the first tape medium An acquisition unit for acquiring first information indicating the number of times the first tape medium has been mounted in the tape drive and second information indicating the number of times the data piece has been read from the first tape medium; A tracking unit that tracks the ratio of the number of times indicated by the second information obtained by the acquiring unit and the number of times indicated by the first information as an access frequency for the data piece; And And a recording unit for writing the pieces of data to the second tape medium in descending order of the access frequencies tracked by the tracking unit. A method of supporting movement of pieces of data recorded on a first recording medium to a second recording medium, the method comprising: For each of the pieces of data recorded on the first recording medium, the first piece of information and the data pieces indicating the number of times any data has been read from the first recording medium after the pieces of data have been recorded on the first recording medium. Acquiring second information indicating the number of times of reading from the first recording medium; Tracking the frequency of access to the pieces of data using the obtained first information and the obtained second information; And Determining a recording position of the piece of data in the second recording medium based on the tracked access frequency. A program product for causing a computer to function as an apparatus for supporting movement of pieces of data recorded on a first recording medium to a second recording medium, The apparatus comprises: For each of the pieces of data recorded on the first recording medium, the first piece of information and the data pieces indicating the number of times any data has been read from the first recording medium after the pieces of data have been recorded on the first recording medium. An acquiring unit for acquiring second information indicating the number of times of reading from the first recording medium; A tracking unit for tracking the frequency of access to the data pieces using the first information and the second information obtained by the acquiring unit; And And a determining unit for determining a recording position of the piece of data in the second recording medium based on the access frequency tracked by the tracking unit.

Images