JP3991512B2 - Recording apparatus and method, and recording / reproducing apparatus - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a recording apparatus and method for recording and / or reproducing digital data using a magnetic tape housed in a cassette cartridge, and a recording / reproducing apparatus, and in particular, recording at a high speed after the cassette cartridge is inserted. The present invention relates to an apparatus and method, and a recording / reproducing apparatus.
[0002]
[Prior art]
Along with the increase in data capacity, there is an increasing need for a data recording device that records data on a magnetic tape stored in a cassette cartridge as means for storing a large amount of data. Conventionally, such a recording apparatus is often used for backup of a large-capacity hard disk, for example. In recent years, there are many examples used in an archive system constituted by a cart machine in which a large number of stored cartridges are automatically exchanged under the control of a host computer to perform recording and reproduction.
[0003]
On the other hand, for a tape device, quick cueing to the target tape position is an important performance index. In particular, computer streamers (hereinafter abbreviated as streamers) that handle computer data are required to speed up the cueing of files recorded on a tape, that is, to shorten the file access time.
[0004]
When the streamer is used for the above-mentioned data backup, the data recorded on the tape is read when, for example, a trouble occurs in the computer used, and high-speed access performance is often required. Absent. However, in the above-described archive system, reading of files on a tape frequently occurs. Therefore, when the streamer is used in the archive system, the high-speed access performance to the file requested to be read determines the performance of the entire system to which the streamer is connected. In particular, in the case of an archiving system combined with the above-described cart machine, it is strongly desired that the first file access be performed at high speed after the tape is inserted, that is, after the cassette cartridge is inserted into the recording / reproducing unit in the cart machine. Yes.
[0005]
In tape devices, since access is sequential, it is clear that the longer the distance from the currently stopped tape position to the target file position, the longer the access time to that file. In order to access any file recorded on the tape at high speed, for example, if the system has a function for statistically analyzing the distribution of files on the tape and the distribution of file access requests, it is accessed next. It is conceivable to feed the tape to a position with a high probability and eject the cassette cartridge.
[0006]
On the other hand, in the case of a general system that does not have such a statistical function, minimizing the average access time after inserting a tape is the same as minimizing the worst value of the access time. . This is the same as minimizing the distance from the tape insertion position to the farthest file.
[0007]
[Problems to be solved by the invention]
In conventional streamers, the cassette cartridge is often ejected and inserted in a state in which it is rewound to the beginning of the tape. In this case, when this cassette cartridge is next inserted into the apparatus, for example, when data is to be added, the tape must be sent to the end of the recording after the cartridge is inserted. There was a problem in that it was not possible to get access.
[0008]
In order to solve this problem, there is a streamer in which the cassette is discharged by sending the tape to the end of the user data recorded on the tape. In this case, for example, when data is recorded over the entire length of the tape, the cassette cartridge is ejected with the tape being sent to the end of the user data, that is, near the end of the tape. Therefore, when the cassette cartridge is next inserted into the apparatus and the head data is accessed, a long search from the end of the tape to the head must be performed.
[0009]
This method has a substantial effect in a system with many additional data such as a backup application. However, the worst value of the access time described above is the same as the example in which the cartridge is discharged after rewinding to the beginning of the tape. .
[0010]
In order to solve these problems, a method of discharging the cassette cartridge at a position in the middle of the tape can be considered. For example, as described above, statistics regarding the access of the tape are taken, a position having a high probability of being accessed next is obtained based on the statistical information, and the cassette cartridge is ejected after the tape is sent to that position. In this way, when the cartridge can be ejected at a position in the middle of the tape, it becomes possible to access at high speed the next time the cartridge is inserted.
[0011]
In order to eject the cartridge in the middle of the user data on the tape, it is necessary to provide a mechanism for protecting the user data from damage to the tape due to the ejection and insertion of the cartridge. In order to protect the user data from damage to the tape due to ejection at an intermediate position of the tape, an area where user data is not written is provided on the tape, and the cartridge is ejected by sending the tape to that area. Exist.
[0012]
However, this method has a problem in that an area in which user data is not written has to be created in advance in order to protect data when the tape is initialized.
[0013]
Further, when user data is written on the tape, there is a problem that it is necessary to provide a mechanism for continuing writing while avoiding an area where user data is not written in order to protect the data.
[0014]
SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide a recording apparatus and method, and a recording / reproducing apparatus which can easily access the data at the tape position when the cassette cartridge is ejected, and perform a high speed access after inserting the cassette cartridge. is there.
[0015]
[Means for Solving the Problems]
In order to solve the above-described problems, the present invention provides a recording device for recording user data on a magnetic tape, a memory means for storing input user data, and user data read from the memory means for recording on the magnetic tape. And when the user data is recorded by the recording means, the memory means repeats a plurality of times from the memory means at a designated position on the magnetic tape based on position information indicating the position on the magnetic tape recorded on the magnetic tape. Control means for reading the user data, recording the user data read a plurality of times on the magnetic tape, forming an area, and controlling the ejection of the magnetic tape at a predetermined position in the area. The recording apparatus characterized by the above.
[0016]
According to another aspect of the present invention, there is provided a recording method for recording user data on a magnetic tape, a step of storing input user data in a memory means, and a recording step of recording user data read from the memory means on a magnetic tape. When the user data is recorded in the recording step, the user data is repeatedly repeated from the memory means at the designated position on the magnetic tape based on the position information indicating the position on the magnetic tape recorded on the magnetic tape. And a step of controlling to record the user data read a plurality of times on the magnetic tape to form an area and to control the ejection of the magnetic tape at a predetermined position in the area. Is a recording method.
[0017]
According to another aspect of the present invention, there is provided a recording / reproducing apparatus for recording user data on a magnetic tape and reproducing the recorded user data, a memory means for storing the inputted user data, and a user data read from the memory means. Recording means for recording on the tape, reproducing means for reproducing the user data recorded on the magnetic tape, and position information indicating the position on the magnetic tape recorded on the magnetic tape when the user data is recorded by the recording means. Based on the specified position on the magnetic tape, the user data is repeatedly read from the memory means multiple times, and the user data read multiple times is recorded on the magnetic tape to form an area, and the magnetic tape is ejected. A recording / reproducing apparatus comprising: control means for performing control at a predetermined position in the area.
[0018]
As described above, according to the present invention, the user data read from the memory means is recorded on the magnetic tape based on the position information indicating the position on the magnetic tape recorded on the magnetic tape. It is controlled so that user data is repeatedly read out from the memory means at a designated position, and the user data read out multiple times is ejected in the area where the magnetic tape is recorded. Since it is controlled, it is possible to suppress the influence of damage to the tape when it is ejected in the middle of the magnetic tape.
[0019]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 shows an example of a connection form of each device that can be applied to an embodiment of the present invention. The host computer 1 and the tape streamer 3 to which the present invention is applied are connected using a SCSI (Small Computer System Interface) as an interface. That is, a SCSI board is mounted on the host computer 1 (not shown). A SCSI cable 2 is connected to the SCSI board, and the host computer 1 and the tape streamer 3 are connected by the SCSI cable 2. The tape streamer 3 has a CPU and a memory, and performs recording / reproduction of data using a magnetic tape stored in a cassette cartridge (hereinafter abbreviated as a cartridge) based on the control of the CPU.
[0020]
The SCSI board has a CPU, for example, and issues a SCSI command based on an instruction issued from the host computer 1. The SCSI board controls various phases of data transfer according to instructions from the host computer 1. Exchange based on the SCSI command is performed with the other party (the tape streamer 3 in this example) via the SCSI cable 2, and data is transferred between the host computer 1 and the tape streamer 3. Note that the interface between the tape streamer 3 and the host computer 1 is not limited to SCSI. As this interface, an interface of another system capable of communicating digital data, for example, IPI (Intelligent Peripheral Interface) can be applied.
[0021]
At the time of recording, a cartridge for recording is inserted into a cartridge insertion port (not shown) provided in the tape streamer 3. When the cartridge is inserted, a predetermined loading operation is performed. That is, the cartridge is set at a predetermined position, and the magnetic tape stored in the cartridge is pulled out and can be recorded. A command for instructing recording is transferred from the host computer 1 to the tape streamer 3 together with recording data to be recorded. In accordance with this command, recording data is recorded on the magnetic tape by the tape streamer 3. When the recording is completed, a predetermined unload operation is performed. That is, the magnetic tape is stored in the cartridge, the cartridge is moved, and is ejected from the cartridge insertion port. Even during reproduction, the operation of the cartridge is substantially the same as during recording.
[0022]
FIG. 2 shows an example of a data image on the magnetic tape in the tape streamer 3 that can be applied to the embodiment of the present invention. As shown in FIG. 2A, data is written in units of files on the tape. At the end of each file, a tape mark TM which is a file delimiter code is written. The tape mark TM is also referred to as a file mark FM. A new file is written after this mark TM. Each of the files consists of one or more blocks as shown in FIG. 2B. The size of each block may be different from each other as shown in FIG. 2B, or may be all equal.
[0023]
Although omitted in FIG. 2, a management table for managing data written on the tape is provided at the head of the entire tape. When this tape is loaded onto the tape streamer 3, the management table is read to refer to the arrangement of files on the tape.
[0024]
The tape streamer 3 performs recording on a magnetic tape by a helical scan method. Therefore, a helical track is formed on the magnetic tape. A track set is formed by four tracks, and data writing to the magnetic tape is performed in units of the track set. Each track set is given an ID for identifying the track set from each other. That is, 1 ID is assigned to every 4 tracks. In one track set, an area for subcode data is arranged at the head of the area where the data body is stored, and an ID is written. On the other hand, a block management table for managing the block configuration of the data body is arranged at the end of the area where the data body is stored.
[0025]
FIG. 3 shows a track pattern on the tape streamer 3. Longitudinal tracks are formed above and below the width direction of the tape, and helical tracks are formed between them. A control signal is recorded on the upper longitudinal track 26, and a time code is recorded on the lower longitudinal track 27. These control signals and time codes are recorded over the entire length of the magnetic tape, for example, when the magnetic tape is initialized. The time code indicates the position in the longitudinal direction of the tape. For example, the SMPTE time code is used. In one rotation of the drum 25, two helical tracks Ta and Tb are simultaneously formed by the heads Ra and Rb, and then two helical tracks Tc and Td are simultaneously formed by the heads Rc and Rd. Each helical track is formed by separating a first half portion and a second half portion, and a tracking pilot signal recording area 28 is provided in the middle portion.
[0026]
The SMPTE time code was developed for a video signal such as a VTR, and its minimum unit is a frame (1/30 second). As will be described later, in the data recorder, the four tracks Ta to Td shown in FIG. 3 are used as data units (referred to as track sets) that handle recordable data. For example, in the case where 16 tracks correspond to one frame of the video signal, a digit (value of 0, 1, 2, or 3) lower than the digit of the time code frame is provided, and the track set is used as a unit. It is necessary to use a time code (also referred to as ID). In the case of the SMPTE time code, since a user data area is prepared, such correction is possible.
[0027]
FIG. 4 is a block diagram showing an example of the configuration of the tape streamer 3 applicable to the embodiment of the present invention. As described above, the tape streamer 3 and the host computer 1 are connected via the terminal 100 by the SCSI cable 2 using the SCSI as an interface. Recording data and commands output from the host computer 1 are supplied to the SCSI I / F 101 via the terminal 100. In the SCSII / F 101, the SCSI command supplied from the terminal 100 is analyzed and executed, and the recording data to be transferred is separated from the recording data body and the accompanying data.
[0028]
The recording data output from the SCSI I / F 101 is supplied to the buffer memory 103 via the memory controller 102 and temporarily stored in the buffer memory 103. The communication by the SCSI I / F 101 is bidirectional. For example, data output from the memory controller 102 is supplied to the SCSI I / F 101 and transferred to the host computer 1 via the terminal 100.
[0029]
FIG. 5 shows an exemplary configuration of the buffer memory 103. As an example is shown in FIG. 5B, data is stored in the buffer memory 103 in units of 1 ID. FIG. 5A is a configuration example of 1ID data. As described above, the subcode area and the user data area storing the data body are arranged from the head side, and the terminal side is the block management table area. The storage area of the buffer memory 103 is composed of a plurality of banks with a plurality of IDs as one bank. In this example, one bank consists of 40 IDs, and the buffer memory 103 as a whole has six banks. Writing to the magnetic tape 107 described later, reading of data from the magnetic tape 107, and retry processing are performed in units of one bank.
[0030]
The recording data stored in the buffer memory 103 is read at a predetermined timing for each ID under the control of the memory controller 102, and is supplied to an ECC (Error Correction Coding) encoder 104 via the memory controller 102. The recorded data is error-correction-encoded with a product code using, for example, a Reed-Solomon code, by the ECC encoder 104 in units of 1 ID. The recording data that has been subjected to error correction coding is subjected to timing control or the like by a system controller (system controller) 115 to be described later, and is supplied to the equalizer 105.
[0031]
The recording data supplied to the equalizer 105 is converted into a recording signal suitable for recording on the recording medium, and is supplied to the recording head provided in the rotary head 106. A recording signal is recorded on the magnetic tape 107 by the recording head on the rotary head 106.
[0032]
FIG. 6 shows an exemplary configuration of the rotary head 106. A recording head and a reproducing head are respectively arranged with two heads having different azimuths as a set. In the recording head REC, two sets of heads are arranged at positions facing each other, and two sets of heads are arranged at an angle of 90 ° with these heads. In other words, the recording heads REC are arranged in four sets by changing the angle by 90 °. Similarly, the reproduction head PB is also arranged in four sets with the angle changed by 90 °. The angle between the recording head REC and the reproducing head PB is shifted by 45 °.
[0033]
A signal recorded on the magnetic tape 107 is reproduced by a reproducing head provided in the rotary head 106, and the reproduced signal reproduced is supplied to the equalizer 108. The equalizer 108 performs predetermined processing on the supplied reproduction signal and converts it into digital data. The reproduction data output from the equalizer 108 is supplied to the ECC decoder 109.
[0034]
The ECC decoder 109 performs a decoding process of the error correction code applied at the time of recording on the supplied reproduction data. If there is an error in the data that exceeds the error correction capability of the error correction code as a result of the decoding process, for example, a flag indicating that error correction has not been performed is performed as an error detection result without performing error correction. Is output. The error detection result by the ECC decoder 109 is supplied to the system controller 115.
[0035]
Note that, as described above, the recording head REC and the reproducing head PB are both disposed on the rotary head 106. The track formed by recording by the recording head REC is traced by the reproducing head PB immediately after that, and the signal recorded immediately before is reproduced. The reproduced signal is supplied to the ECC decoder 109 via the equalizer 108, and the error correction code is decoded. By using the result of the decoding process, read after write can be performed to detect whether or not the recording has been performed correctly. A write retry process to be described later is realized by performing this read after write.
[0036]
The output of the ECC decoder 109 is temporarily stored in the buffer memory 103 via the memory controller 102. The reproduction data stored in the buffer memory 103 is read based on the control of the memory controller 102 in response to a request from the host computer 1, for example, and is led to the terminal 100 via the SCSI I / F 101.
[0037]
On the other hand, the travel of the magnetic tape 107 is controlled by the servo 113. That is, the magnetic tape 107 is driven to travel by the capstan motor 110 whose rotational speed is controlled by the servo 113. On the other hand, as described above, a control signal is recorded on the longitudinal track 26 on the magnetic tape 107, and this control signal is read by a fixed head (not shown). The output of the fixed head is supplied to the servo 113, the tape traveling speed is measured, and the rotational speed of the capstan motor 110 is controlled based on the measurement result.
[0038]
The sensor 111 is various sensors that detect the state of the magnetic tape 107 or a cassette cartridge in which the magnetic tape 107 is stored. Based on the detection result of the sensor 111, the loading state of the magnetic tape 107 can be determined. The output of the sensor 111 is supplied to the servo 113 and also to the syscon 115.
[0039]
The TCR circuit 114 is a circuit that decodes and generates a time code. Under the control of the syscon 115, for example, an SMPTE time code is generated by the TCR circuit 114. The generated time code is recorded on the longitudinal track 27 of the magnetic tape 107 by the fixed head 112 when the magnetic tape 107 is initialized, for example. The time code recorded on the longitudinal track 27 is read by the fixed head 112 and supplied to the TCR circuit 114. The TCR circuit 114 decodes this time code and converts it into an ID representing the corresponding track set. The ID obtained by the conversion is supplied to the syscon 115.
[0040]
The syscon 115 is composed of a microprocessor or the like, and controls recording on the magnetic tape 107 and reproduction from the magnetic tape 107, control of driving of the magnetic tape 107, and the like based on the ID supplied from the TCR circuit 114. The system controller 115 controls the output timing of data from the ECC encoder 104 based on the ID. Further, in the syscon 115, in accordance with an instruction from the host computer 1, control for causing the magnetic tape 107 to travel to a predetermined position so as to access target data is performed based on the ID.
[0041]
The cassette cartridge containing the magnetic tape 107 is inserted (loaded) and ejected (unloaded) from the tape streamer 3 under the control of the syscon 115. Information when the cartridge is discharged, for example, position information of the magnetic tape 107, is supplied to a main CPU 117 described later.
[0042]
The main CPU 117 is composed of a microprocessor or the like, and controls the entire tape streamer 3. The main CPU 117 and the syscon 115 are connected via a DP-RAM 116 that is a dual port RAM.
[0043]
The ID converted from the time code is supplied from the system controller 115 to the main CPU 117 via the DP-RAM 116. A command output from the host computer 1 is supplied to the main CPU 117 via the SCSI I / F 101 and the memory controller 102. The main CPU 117 instructs the system controller 115 to control each unit based on the supplied command. Further, the main CPU 117 issues an instruction for controlling the storage of data in the buffer memory 103 to the memory controller 102.
[0044]
A RAM (Random Access Memory) 118 is connected to the main CPU 117. Based on the above-described information when the cartridge is supplied from the syscon 115, the main CPU 117 generates header information. The generated header information is stored in the RAM 118.
[0045]
The I / F CPU 120 is composed of a microprocessor or the like, and controls each interface that controls communication between the tape streamer 3 and the outside. I / F
The CPU 120 is connected to the main CPU 117 via a DP-RAM 119 that is a dual port RAM. The DP-RAM 119 may be configured in common with the DP-RAM 116 described above. An SIO 121, an Ethernet IO 124, and a display CPU 125 are connected to the I / F CPU 120.
[0046]
The SIO 121 is a serial interface based on, for example, RS-422 or RS-232C, and communication of maintenance information is performed via the terminal 122. The Ether IO is an Ethernet interface and can communicate with an external device connected via the terminal 124. For example, the header information stored in the RAM 118 can be transmitted to another computer system connected to the outside via the I / F CPU 120 and the Ethernet IO 124.
[0047]
The display CPU 125 creates display data displayed on a display screen provided on an operation panel (not shown) of the tape streamer 3 based on data supplied from the main CPU 117 via the I / F CPU 120, for example. . At the same time, the display CPU 125 performs drive control of a display device such as an LCD (Liquid Crystal Display).
[0048]
Further, the display CPU 125 converts signals generated based on operations of various switches (not shown) provided on the operation panel into data that can be decoded by the main CPU 117. The state of the tape streamer 3 can be monitored on the operation panel, and the recording / reproducing process in the tape streamer 3 can be controlled.
[0049]
On the other hand, as shown in FIG. 7, a memory label 131 is attached to the cassette cartridge 130 on the back side. The memory label 131 is configured such that an antenna and a receiving circuit integrally formed in a film shape and a nonvolatile memory are formed in a sheet having adhesiveness on one side. A CPU can also be incorporated in the memory label 131. When a radio wave transmitted from the corresponding antenna is received, an electromotive force is generated by the antenna in the memory label 131, and by this electromotive force, data is written to the nonvolatile memory and data is read from the nonvolatile memory. Can be done without contact.
[0050]
A memory label I / F 126 that interfaces with the internal circuit of the memory label 131 attached to the cassette cartridge 130 via the antenna 127 is connected to the display CPU 125. Based on commands and data output from the main CPU 117 and supplied via the display CPU 125, a predetermined signal is generated at the memory label I / F 126. The generated signal is transmitted from the antenna 127. When this signal is received by the antenna in the memory label 131, data and commands are supplied to the internal circuit of the memory label 131, data is read from the nonvolatile memory of the memory label 131, and the memory label 131 is read. Data is written to the nonvolatile memory.
[0051]
For example, the header information stored in the RAM 18 is read by the main CPU 117 and supplied to the memory label I / F 126 together with a command for instructing writing of the memory label 131 to the nonvolatile memory. The header information and the write command are transmitted from the antenna 127 and received by the antenna in the memory label 131. In the memory label 131, a command and header information are extracted from the received signal, and the header information is written in a built-in nonvolatile memory according to the command.
[0052]
Next, a tape format for recording digital data will be described. First, the layout of the entire tape (for example, a tape in one cartridge) is shown in FIG. The entire tape is a physical volume. The recordable area between the physical tape beginning PBOT (Physical Beginning of Tape) and the terminal PEOT (Physical End of Tape), to which a leader tape is connected, is LBOT (Logical Beginning of Tape). And LEOT (Logical End of Tape). This is because at the beginning and end of the tape, the tape is easily damaged and the error rate is high. As an example, the invalid area between PBOT and LBOT is defined as 7.7 ± 0.5 m, and the invalid area between PEOT and LEOT is defined as greater than 10 m.
[0053]
A logical volume is arranged in one physical volume, and a VSIT (Volume Set Information Table) including management information for managing the logical volume is recorded at the head of the recording area. The head position of VSIT is set to the position of 0ID. ID (Identification) is an address corresponding to the position on the tape assigned to each of the four track sets. The ID is monotonically increased from the VSIT area to the DIT area of the last volume. The length of one VSIT is 1 ID.
[0054]
The logical volume includes a DIT (Directory Information Table), a UIT, and a user data area. The DIT has information for managing files in the logical volume. The length of one DIT is 40 ID. The UIT is optional and is user-specific information for managing files.
[0055]
In FIG. 8, the hatched area is a run-up area. The data track is servo-locked by the run-up area. An area with dots is a margin band. This position margin band prevents the valid data from being erased when VSIT and DIT are updated.
[0056]
VSIT is repeatedly recorded 10 times as shown in FIG. 9A in order to improve data reliability. Therefore, the VSIT area is 10 track sets (= 10 ID). A retry area of 90 track sets or more is secured after the VSIT area.
[0057]
The DIT is repeatedly recorded seven times as shown in FIG. 9B in order to improve data reliability. As shown in FIG. 9C, the DIT is composed of six tables. The six tables are VIT (Volume Information Table), BST (Bad Spot Table), LIDT (Logical ID Table), FIT (File Information Table), UT (Update Table), and UIT (User Information Table) in order from the top. It is. VIT, BST, LIDT, and UT have a length of 1 ID, and FIT has a length of 20 ID. The remaining 16 ID area is reserved.
[0058]
Each table of the DIT will be described. The ID address of VIT is the head physical ID of the volume written in VSIT, and its logical ID is equal to the head physical ID of the volume written in VSIT. The VIT includes volume information such as a volume label, the start physical ID of the first data block in the physical volume, and the last physical ID.
[0059]
The BST ID address is VIT physical ID + 1, and its logical ID is VIT logical ID + 1. The BST has position information of data that is logically invalidated. Logically invalid data is data that should be treated as invalid because data having the same track set ID is written later. For example, as shown in FIG. 10, the shadow area A is logically invalid data. Logically invalid data is generated by the write retry operation and the accompanying write operation. If an error has occurred during writing by the above-described read after write, a write retry is automatically performed, an error location is output, and this is registered in the BST. During the read operation, an invalid area is designated by the BST. Logically invalid data is also referred to as a bad spot. The BST manages the start physical ID and end physical ID of up to 14592 bad spots.
[0060]
The ID address of LIDT is the physical ID + 2 of VIT, and the logical ID is the logical ID + 2 of VIT. LIDT is a data table for high-speed block space and locate operations. That is, the logical ID of each pointer of the first to 296th pointers, the physical ID, the file number, and the first block number in the block management table of ID data are included in the LIDT.
[0061]
The FIT ID address is VIT physical ID + 3, and its logical ID is VIT logical ID + 3. The FIT is composed of a plurality of pairs in which two types of data corresponding to the tape mark are paired. The tape mark is a file delimiter code (delimiter code). The Nth data pair corresponds to the Nth tape mark from the beginning of the volume. One data of the pair is the physical ID of the Nth tape mark. The other data is the absolute block number of the Nth tape mark. This value is the absolute block number of the last block with the same file number as the tape mark. Since the position of the tape mark can be accurately known from the physical ID of the tape mark and the absolute block number, the physical position on the tape can be accessed at high speed.
[0062]
The UT ID address is VIT physical ID + 39. The UT is information indicating whether the volume has been updated. Before the update, the word (4 bytes) indicating the update status in the UT is set to FFFFFFFFh (h means hexadecimal), and after the update, this is set to 00000000h.
[0063]
The UIT is optional and is, for example, an area of 100 ID. A data table accessible to the user, reserved for the user header.
[0064]
In this example, 1 ID is assigned to each track set including four helical tracks. The logical structure of the data block is defined for each track set. FIG. 11 shows a logical track set structure. The first 4 bytes of the logical track set is a format ID, which is FFFF0000h.
[0065]
The next 136 bytes (34 words) is an area for subcode data. The subcode data stores information on track set management. For example, identification codes such as the above-described tables (VSIT, VIT, BST, etc.), user data, and tape marks are included in the subcode.
[0066]
Further, the number of bytes obtained by subtracting the length of the block management table from the next 116884 bytes is the user data write area. When the track set is used for writing user data, dummy data is packed in the remaining area when the size of the user data does not reach a predetermined size. The track set format defined in the user data area includes a user data track set for writing user data, a tape mark (TM) track set for indicating a tape mark, and an EOD (End Of Data) track. There are four types: sets and dummy track sets. A subcode is defined for each track set format.
[0067]
A block management table area is provided after the user data area. The block management table has a maximum length of 4096 bytes. The last 4 bytes of the track set are used as the end code (0F0F0F0Fh) of the track set, and the preceding 12 bytes are reserved. The block management table manages the data block configuration of user data.
[0068]
The retry process of the tape streamer 3 is performed as follows. If an uncorrectable error is detected by the ECC decoder 109 as a result of performing signal processing while data is being recorded, when the recording of the block including the block in which the error is detected is completed, the same block is buffered. It is read again from the memory 103. The block read from the buffer memory 103 is error correction encoded by the ECC encoder 104, and is recorded on the magnetic tape 107 following the block including the data in which the error is detected. A retry process is performed by repeating the process of reading a block in which an error is detected from the buffer memory 103, error-correcting the read block, and recording the block on the magnetic tape 107 a predetermined number of times.
[0069]
The determination as to whether or not to retry is made, for example, in units of 40 IDs that are the bank unit of the buffer memory 103. Data for 40 IDs are stored in the buffer memory 103, and if there is an error in one ID of the stored 40 ID data, a retry process is performed for the 40 IDs.
[0070]
Next, creation of a discharge area for the magnetic tape 107 according to the gist of the present invention will be described. In the present invention, an area where the same user data is repeatedly written is created at a regular position on the magnetic tape 107, and this area is defined as a cartridge discharge area (hereinafter referred to as a discharge area). The process of repeatedly writing the same user data to create the discharge area is performed by the above-described retry process (write retry).
[0071]
That is, when the main CPU 117 detects a logical ID of a preset discharge area, the main CPU 117 instructs the memory controller 102 to repeatedly read data in the same bank, and the main CPU 117 The system controller 115 is instructed to perform a retry process. In accordance with these instructions, the control as described above is performed by the syscon 115 to drive the tape, and the same data is repeatedly written on the magnetic tape 107.
[0072]
As described above, according to the present invention, the discharge area is constructed during the process of writing the user data. Therefore, for example, when the magnetic tape 107 is initialized, there is no need to separately initialize the discharge area. Further, since the operation for creating the discharge area is common to the retry process for guaranteeing the error rate, a process for avoiding the discharge area is not necessary. Further, since the same user data is repeatedly written, even if there is a tape damage when the cartridge is ejected and inserted in the ejection area, the user data of the next block can be read even if the user data cannot be read.
[0073]
The method for creating the discharge area will be described in more detail. In this embodiment, the number of discharge areas created for the entire magnetic tape 107 is preset by a predetermined method. Then, the entire length of the magnetic tape 107 is equally divided based on the set number of discharge areas, and the equally divided positions are determined as discharge position candidates. Hereinafter, this ejection position candidate is expressed as an ejection position candidate ID using an ID on the magnetic tape 107. In the following, all IDs are assumed to be logical IDs. Not limited to this, the ID may be a physical ID.
[0074]
The number of discharge areas varies depending on the worst access speed of the entire system. For example, the number of ejection areas is preferably determined so that the ejection position candidate ID is at an appropriate position on the tape 107 based on the worst value of the access speed of the entire system.
[0075]
For example, the user can designate the number of discharge areas from the host computer 1 to the tape streamer 3. However, the present invention is not limited to this, and a switch for setting the number of discharge areas may be provided in the main body of the tape streamer 3 so that the user directly sets the number of discharge areas for the tape streamer 3.
[0076]
FIG. 12 is a flowchart of an example of a discharge area creation process. FIG. 13 is a diagram for explaining creation of a discharge area based on this flowchart. In FIG. 13A, the left end side is LBOT, the entire magnetic tape 107 is already divided into eight equal parts, and the position indicated by the downward arrow is set as a discharge position candidate. That is, the number of discharge areas is set to 7. Each discharge position candidate is represented by a discharge position candidate ID. In addition, the written user data is indicated by hatching, the first four discharge areas are created, and the user data is written between the fourth discharge area and the fifth discharge area. ing. Hereinafter, a process for creating the fourth discharge area will be described.
[0077]
The ID information set as each discharge position candidate ID is stored in the RAM 118. At this time, each ID information is stored in the RAM 118 together with a confirmed bit indicating whether or not the ID is confirmed as a discharge / insertion position ID by a process described later.
[0078]
Prior to the execution of the flowchart of FIG. 12, the size of the damage zone is determined in advance by the number of IDs. The damage zone is an area where the tape 107 may be damaged when the cartridge is discharged or inserted in the middle of the tape 107. The discharge area is created so as to cover the damage zone according to the set size of the damage zone. Here, it is assumed that the damage zone has a size of 8 ID and the discharge area has a size of 10 ID.
[0079]
First, the cartridge has already been inserted into the tape streamer 3, and writing of user data has also started. In step S10, user data for 40 IDs (referred to as user data UD1) is written, and in step S11, the current ID is confirmed. The time code recorded on the longitudinal track 27 of the magnetic tape 107 is read by the fixed head 112, and the read time code is converted to ID by the TCR circuit 114 and supplied to the syscon 115. This ID is supplied from the syscon 115 to the main CPU 117, and the ID is confirmed.
[0080]
In the next step 12, the main CPU 117 determines whether or not the ID confirmed in step S <b> 11 is a discharge position candidate ID. If it is determined that the ID is not the ejection position candidate ID, the process returns to step S 10, and the next user data is written on the magnetic tape 107.
[0081]
On the other hand, if it is determined in step S12 described above that the ID has become a discharge position candidate ID, the process proceeds to step S13. In step S13, a write retry process is performed. In step S13, the main CPU 117 issues a command for performing a write retry process to the system controller 115. Upon receiving this command, the syscon 115 performs a predetermined write retry process regardless of the result of error detection by the ECC decoder 109.
[0082]
For example, as shown in FIG. 13B, first, user data UD2, which is the next data of user data UD1, is written for one block (40 ID). Then, assuming that the data UD2 is an error, the data is repeatedly read from the same bank of the buffer memory 103, and the read data is written to the magnetic tape 107 and subjected to a write retry process (FIG. 13B). The shaded part in the middle).
[0083]
Along with the write retry process, the main CPU 117 counts the number of IDs subjected to the write retry process.
[0084]
In the next step S14, the main CPU 117 determines whether or not the number of IDs subjected to the write retry process exceeds the number of IDs set as the damage zone. If it is determined that it has not exceeded, the process returns to step S13, and the write retry process is continued. On the other hand, if it is determined that it has exceeded, it is determined that the creation of one discharge area is completed, the number of IDs subjected to the write retry process is reset, and the process is returned to step S10, and the next user data UD3. Is written.
[0085]
In step S14, if it is determined that the number of IDs subjected to the write retry process exceeds the number of IDs in the damage zone, one discharge area is created and determined. When the discharge area is determined, the discharge / insertion position for discharging or inserting the cartridge 130 in the middle of the tape 107 is determined, and the ID of the position is determined. The discharge / insertion position is preferably approximately the center of the discharge area, and the ID of this position is determined as the discharge / insertion position ID in the discharge area.
[0086]
The determined discharge insertion position ID is stored in the RAM 118. At this time, the discharge insertion position ID is overwritten with respect to the corresponding discharge position candidate ID stored in the RAM 118, for example, and the corresponding confirmation bit is changed to a value indicating that the discharge insertion position ID has been fixed. . For example, the fixed bit value that was originally [0] is changed to [1].
[0087]
FIG. 14 schematically shows the discharge area created in this way. The discharge area is an area for 10 blocks in which user data UD2 is written. In the figure, the hatched portion is the damage zone described above. The length of the damage zone is due to the mechanical elements of the tape streamer 3 used, and is, for example, about 10 cm to 50 cm on the magnetic tape 107. The discharge area is created to cover the damage zone. In addition, the discharge / insertion position where the cartridge 130 is discharged or inserted in the middle of the tape 107 is set at a substantially central portion of the discharge area.
[0088]
The determined discharge insertion position ID is stored in the RAM 118 as header information. The header information stored in the RAM 118 is transferred to and stored in the nonvolatile memory of the memory label 131 attached to the cartridge 130. Data contents stored in the nonvolatile memory of the memory label 131 will be described later. The header information stored in the RAM 118 is finally written to the VSIT of the magnetic tape 107 or sent to the host computer 1. The header information stored in the RAM 118 may be transferred to a higher-order system connected by, for example, Ethernet.
[0089]
When the cartridge 130 is ejected, for example, the main CPU 117 obtains the current tape position ID information and reads the ejection insertion position ID stored in the RAM 118 as header information. The main CPU 117 compares the ID information of the current tape position with the discharge / insertion position ID, and causes the syscon 115 to move the magnetic tape 107 to the position of the discharge / insertion position ID closest to the current tape position. Instructions are given. The syscon 115 controls the servo 113 based on this instruction, drives the magnetic tape 107 to move to the instructed ID, and then ejects the cartridge 130.
[0090]
The header information is not limited to this example. For example, the header information may be supplied to the Ethernet IO 123 via the I / F CPU 120 and transferred to the outside from the terminal 124 via Ethernet. The header information is supplied to, for example, a higher system via the Ethernet, and is used as management information for each cartridge 130.
[0091]
When reading data from the magnetic tape 107, the user data UD2 is read after the user data UD1. At this time, the user data read out first, that is, the first user data UD2 in the discharge area is usually used. However, for example, when the first user data UD2 in the discharge area cannot be read due to tape damage due to the discharge position deviating before the discharge insertion position ID at the previous discharge of the cartridge, the next user data UD2 is Used.
[0092]
This will be described with reference to FIG. In the example of FIG. 15, among the user data UD2 # 1 to UD2 # 10 for 10 blocks in the discharge area, the data UD2 # 1 to UD2 # 5 and the data UD2 # 7 and # 8 can be read for reasons such as tape damage. It disappears (the shaded area in FIG. 15). In this example, when a request for reading the user data UD2 is received from the host computer 1, the data of the user data UD2 # 6 that is not damaged is returned to the host computer 1. Since the same user data is repeatedly written in the ejection area, the user data can be protected from tape damage due to ejection and insertion operations.
[0093]
When the cartridge 130 is ejected, the tape streamer 3 can write the header information, in this case, the above-described VSIT and DIT, on the magnetic tape 107. Whether to write VSIT and DIT on the tape 107 is instructed by the host computer 1. However, the present invention is not limited to this. For example, it is also possible to instruct the VSIT and DIT to be written on the tape 107 from a higher system connected by Ethernet.
[0094]
An extra write retry process is performed when creating the discharge area so that the header information composed of VSIT and DIT can be written after the discharge area. That is, as shown in FIG. 16A as an example, the user data UD2 is repeatedly recorded on the discharge area 200 created by repeatedly writing the user data UD2 by the write retry process in the above-described process, An area 201 is created. When the header information composed of VSIT and DIT is written on the tape 107, when the cartridge 130 is ejected, the header information composed of VSIT and DIT is written to the user data UD2 written in the area 201 as shown in FIG. 16B. Overwritten.
[0095]
FIG. 17 shows an example of items relating to ejection (unloading) of the cartridge 130 among the data stored in the nonvolatile memory of the memory label 131. The nonvolatile memory of the memory label 131 is assigned an address, and stores various information relating to unloading, that is, ejection of the cartridge 130 to which the memory label 131 is attached. In the 4 bytes from the address [0x000000024] to [0x000000024B], the previously ejected position of the cartridge 130 to which the memory label 131 is attached is stored as the physical ID of the magnetic tape 107.
[0096]
Furthermore, in this example, the determined discharge insertion position IDs described above are provided for eight locations as “halfway unload position” in 32 bytes from the address [0x000000260] to [0x00000027F] of the nonvolatile memory of the memory label 131. , Can be memorized. That is, in this cartridge 130, eight discharge areas are provided for the entire length of the magnetic tape 107. The nonvolatile memory of the memory label 131 uses the ID information stored in the RAM 118 as a confirmed discharge / insertion position ID, which is a value indicating that the ejection / insertion position ID is confirmed, for example, [1]. Stored at the above address.
[0097]
The contents of the RAM 118 may be stored in the nonvolatile memory of the memory label 131 in order from the ID whose confirmation bit is set to [1]. For example, when the cartridge 130 is ejected, the contents of the RAM 118 are collectively stored in the nonvolatile memory of the memory label 131. You may make it memorize | store in memory. For example, the contents of the RAM 118 can be transferred to an external system connected via Ethernet, or can be transferred to the host computer 1.
[0098]
【The invention's effect】
As described above, according to the present invention, since the discharge area is established during the process of writing user data, it is necessary to initialize the discharge area separately, for example, when initializing the magnetic tape. There is no effect.
[0099]
Further, according to the present invention, since the discharge area creation operation is common to the retry process for guaranteeing the error rate, there is an effect that it is not necessary to perform a process for avoiding the discharge area.
[0100]
Further, according to the present invention, since the same user data is repeatedly written in the discharge area, even if there is tape damage when the cartridge is discharged and inserted in the discharge area, the user data cannot be read out. There is an effect that the user data of this block can be read.
[Brief description of the drawings]
FIG. 1 shows an example of a connection form of devices applicable to an embodiment of the present invention.
FIG. 2 is a schematic diagram illustrating an example of a data image on a magnetic tape in a tape streamer.
FIG. 3 is a schematic diagram illustrating a track pattern on a tape of a tape streamer.
FIG. 4 is a block diagram showing a configuration of an example of a tape streamer applicable to the present invention.
FIG. 5 is a schematic diagram illustrating a configuration of an example of a buffer memory.
FIG. 6 is a schematic diagram illustrating a configuration of an example of a rotary head.
FIG. 7 is an external view of an example of a cartridge.
FIG. 8 is a schematic diagram illustrating an example layout of the entire tape.
FIG. 9 is a schematic diagram for explaining VSIT and DIT.
FIG. 10 is a schematic diagram for explaining a bad spot.
FIG. 11 is a schematic diagram illustrating an example of a logical track set structure.
FIG. 12 is a flowchart illustrating an example of a discharge area creation process.
FIG. 13 is a schematic diagram for explaining discharge area creation processing;
FIG. 14 is a schematic diagram schematically showing a created discharge area.
FIG. 15 is a schematic diagram for explaining reading of user data when damage occurs;
FIG. 16 is a schematic diagram for explaining creation of a discharge area when writing VSIT and DIT when a cartridge is discharged;
FIG. 17 is a schematic diagram illustrating an example of items related to ejection of a cartridge among data stored in a memory label.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Host computer, 3 ... Tape streamer, 102 ... Memory controller, 103 ... Buffer memory, 104 ... ECC encoder, 106 ... Rotary head, 107 ... Magnetic tape, 109 ... ECC decoder, 112 ... Fixed head, 113 ... Servo, 114 ... TCR circuit, 115 ... System controller, 117 ... Main CPU, 118 ... RAM, 123 ... Ether IO, 126 ... Memory label I / F, 130 ... Cassette cartridge, 131 ... Memory label

Claims (15)

In a recording device for recording user data on a magnetic tape,
Memory means for storing input user data;
Recording means for recording user data read from the memory means on a magnetic tape;
When recording the user data by the recording means, a plurality of times from the memory means at a specified position on the magnetic tape based on position information indicating the position on the magnetic tape recorded on the magnetic tape. Control means for repeatedly reading user data, recording the user data read repeatedly several times on the magnetic tape to form an area, and controlling the ejection of the magnetic tape at a predetermined position in the area A recording apparatus comprising: The recording apparatus according to claim 1,
The recording apparatus, wherein the designated positions are provided at a plurality of locations on the magnetic tape. The recording apparatus according to claim 1.
Error correction encoding means for error correcting encoding user data recorded by the recording means;
Immediately after the user data is recorded by the recording means, the recorded user data is reproduced, and error detection means for detecting an error by decoding an error correction code of the reproduced user data;
When an error is detected by the error detection result by the error detection unit, the user data in which the error is detected is read from the memory unit, encoded by the error correction encoding unit, and encoded by the recording unit. Retry means for performing a retry process by repeating recording a plurality of times, and
A recording apparatus, wherein the area is formed by the retry means. The recording apparatus according to claim 1,
The recording apparatus according to claim 1, wherein the control means controls to eject the magnetic tape at a substantially central position in the area. The recording apparatus according to claim 1,
A recording apparatus further comprising position information storage means for storing position information indicating the area controlled so that the magnetic tape is ejected by the control means. The recording apparatus according to claim 1,
Position information indicating the area controlled so that the magnetic tape is ejected by the control means is stored in a non-volatile memory means provided in a cassette cartridge in which the magnetic tape is stored. Recording device. The recording apparatus according to claim 1,
The recording apparatus according to claim 1, wherein the control means forms the area in excess for recording predetermined header information. In a recording method for recording user data on a magnetic tape, the step of storing input user data in a memory means;
A recording step of recording user data read from the memory means on a magnetic tape;
When the user data is recorded in the recording step, a plurality of data are recorded from the memory means at a specified position on the magnetic tape based on position information indicating the position on the magnetic tape recorded on the magnetic tape. Control to read user data repeatedly, record the user data read multiple times on the magnetic tape to form an area, and control the ejection of the magnetic tape at a predetermined position in the area A recording method comprising the steps of: In a recording / reproducing apparatus for recording user data on a magnetic tape and reproducing the recorded user data,
Memory means for storing input user data;
Recording means for recording user data read from the memory means on a magnetic tape;
Reproducing means for reproducing user data recorded on the magnetic tape;
When recording the user data by the recording means, a plurality of times from the memory means at a specified position on the magnetic tape based on position information indicating the position on the magnetic tape recorded on the magnetic tape. Control means for repeatedly reading user data, recording the user data read repeatedly several times on the magnetic tape to form an area, and controlling the ejection of the magnetic tape at a predetermined position in the area And a recording / reproducing apparatus. The recording / reproducing apparatus according to claim 9.
The recording / reproducing apparatus according to claim 1, wherein the control means controls to eject the magnetic tape at a substantially central position in the area. The recording / reproducing apparatus according to claim 9.
The recording / reproducing apparatus, wherein the designated positions are provided at a plurality of locations on the magnetic tape. The recording / reproducing apparatus according to claim 9.
Error correction encoding means for error correcting encoding user data recorded by the recording means;
Immediately after the user data is recorded by the recording means, the recorded user data is reproduced, and error detection means for detecting an error by decoding an error correction code of the reproduced user data;
When an error is detected by the error detection result by the error detection unit, the user data in which the error is detected is read from the memory unit, encoded by the error correction encoding unit, and encoded by the recording unit. Retry means for performing a retry process by repeating recording a plurality of times, and
A recording / reproducing apparatus, wherein the area is formed by the retry means. The recording / reproducing apparatus according to claim 9.
The upper SL control means recording and reproducing apparatus characterized by further comprising position information storage means for the position information indicating the area to be controlled as emissions of the magnetic tape is performed is stored. The recording / reproducing apparatus according to claim 9.
Characterized in that the position information indicating the region in which the upper Symbol control means the discharge of the magnetic tape is controlled to be performed is stored in the nonvolatile memory means provided in the cassette cartridge in which the magnetic tape is accommodated A recording / reproducing apparatus. The recording / reproducing apparatus according to claim 9.
The recording / reproducing apparatus is characterized in that the control means forms the area in excess for recording predetermined header information.

Images