WO1998045840A1 - Dispositif et procede d'enregistrement de donnees et dispositif et procede de commande de pile de disques - Google Patents
Dispositif et procede d'enregistrement de donnees et dispositif et procede de commande de pile de disques Download PDFInfo
- Publication number
- WO1998045840A1 WO1998045840A1 PCT/JP1998/001597 JP9801597W WO9845840A1 WO 1998045840 A1 WO1998045840 A1 WO 1998045840A1 JP 9801597 W JP9801597 W JP 9801597W WO 9845840 A1 WO9845840 A1 WO 9845840A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- data
- sub
- block
- blocks
- recording
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
- G11B20/00—Signal processing not specific to the method of recording or reproducing; Circuits therefor
- G11B20/10—Digital recording or reproducing
- G11B20/12—Formatting, e.g. arrangement of data block or words on the record carriers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/08—Error detection or correction by redundancy in data representation, e.g. by using checking codes
- G06F11/10—Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's
- G06F11/1076—Parity data used in redundant arrays of independent storages, e.g. in RAID systems
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
- G11B20/00—Signal processing not specific to the method of recording or reproducing; Circuits therefor
- G11B20/10—Digital recording or reproducing
- G11B20/18—Error detection or correction; Testing, e.g. of drop-outs
Definitions
- the present invention relates to a data recording apparatus and method for recording supplied source data on a plurality of disks, and recording / reproducing source data More particularly, the present invention relates to a disk array control apparatus and method for controlling a disk array for controlling a disk array, and in particular, by generating sub-blocks of different sizes according to a predetermined algorithm, always maintaining a real-time property and controlling a disk within a fixed time.
- the present invention relates to a data recording apparatus and method for ensuring that recording and reproduction processing is performed, and a disk array control apparatus and method.
- HDDs hard disks
- This disk array device incorporates a plurality of disk devices (HDDs) and speeds up reading or writing by moving each disk device in parallel.
- HDDs disk devices
- the data is divided to generate a plurality of divided data, and an error correction data is generated from the plurality of divided data. Then, the plurality of divided data and the error correction data are written to separate disk devices.
- a plurality of divided data and error correction data are simultaneously sent from a disk device storing a plurality of divided data constituting the data and an error correction data.
- the original data is constructed from the read and read divided data, and is transmitted as it is if there is no error.
- the data that stores the divided data cannot be read normally because of a damaged record area, etc., the other correctly read divided data and the error correction data are deleted. After restoring the correct data, send it.
- RAID-1 is to duplicate the data of the disk unit. Also called lard disc. In RAID-1, the exact same data is stored on two disk units. RAID-2, 3 divides input data into bit units or byte units and stores them in multiple disk units. RAID-2 uses a Hamming code, and MID-3 uses a parity as error correction data. RAID-4 and RAID-5 move data in sector units. In RAID-4, the harmony is stored on the same disk device, whereas in RAID-5, it is distributed over multiple disk devices.
- FIG. 64 shows an example of a configuration of a RAID-3 type disk array device
- FIG. 65 shows an example of a configuration of a RAID-5 type disk array device.
- the input data is divided into bytes, and each divided data is stored in a plurality of disk devices. Then, the parity as the correction data is stored in a predetermined disk device.
- the parity P 1-4 for the data of numbers 1 to 4 stored in a plurality of disk devices and the parity P 5-8 for the data of numbers 5 to 8 are stored.
- the input data is divided into sector units, and the data is interleaved and distributed to a plurality of disk devices.
- data A, E, and I are stored in the first disk device
- data B, F, and J are stored in the next disk device
- data C, G, and data I are stored in the next disk device.
- Parity PIL for L is stored.
- the data D, and the parity P EH for the data E to H are stored.
- the last disk device stores the parity P A-D for the data A to D and the data H, L.
- a conventional disk array device that uses such a RAID algorithm to record the divided data divided from one video frame on each disk in a distributed manner, for example, reduces the continuity of video data.
- it is configured to start recording data from the outer circumference (or inner circumference) of the disc.
- the divided data divided from the relatively first half frame is all recorded on the outer peripheral side of the disc, and the divided data divided from the relatively latter half frame are all recorded on the inner peripheral side of the disc. It will be recorded.
- the divided data obtained by the division also has a constant amount of data.
- both the outer track having the larger storage capacity and the inner track having the smaller storage capacity are recorded with the divided data having the same data amount.
- the time to access the split data on the inner track is longer than the time to access the split data on the inner track.
- this conventional disk array device has a problem that it cannot guarantee that any video frame can be read within a certain time. In other words, it is not possible to process video data overnight in real time.
- guaranteeing the upper limit that processing can be performed in such a short time in the worst case is called assurance of real-time performance.
- the ability to guarantee this real-time capability is a very important function. In other words, it is desired that the recording and reproducing process be performed overnight within a certain processing time.
- the present invention has been made in view of the above situation, and has a data recording apparatus that records supplied source data on a plurality of disks while guaranteeing real-time performance. It is an object of the present invention to provide a method and a disk array control apparatus and method for reproducing source data while ensuring real-time performance.
- the data recording apparatus is a data recording apparatus for recording the supplied source data on a plurality of disk media that can be randomly accessed.
- Data processing means for generating block data, and generating parity data for correcting errors of the plurality of sub-blocks from the plurality of sub-blocks; and a plurality of data generated by the data processing means.
- Recording means for recording the sub-block data and the parity data on different disk media, and a sub-block having a size corresponding to a position on the disk medium designated for the sub-block.
- control means for controlling the sub-port and the dividing process in the data processing means. It is.
- the data processing means generates the plurality of sub-ports and the parity so that the recording error on the outer circumference and the inner circumference of each disk is used uniformly instead of the control means.
- control means for controlling the processing can be mentioned.
- control means instead of the control means, the data processing means according to the position on the disc designated for the sub-block. Control means for controlling the data processing means so as to vary the ratio between the sub-block data and the parity data generated as described above.
- the data recording apparatus is a data recording apparatus for recording supplied source data on a plurality of disk media that can be randomly accessed, by dividing one block of the source data into a plurality of sub-block data.
- Data processing means for generating parity data and generating parity data for correcting errors in the plurality of sub-blocks from the plurality of sub-blocks, and data generated by the data processing means Recording means for recording the plurality of sub-block data and the parity data on different disks, respectively, and the sub-block in the data processing means in accordance with a position on the disk designated for the sub-block.
- Control means for controlling the data processing means so as to vary the number of divisions. .
- the data recording apparatus is a data recording apparatus for recording the supplied source video data on a plurality of randomly accessible disks, wherein the data of one frame of the source video data is stored in the data recording apparatus.
- Data processing means for generating a plurality of sub-blocks by dividing, and generating parity data for correcting errors of the plurality of sub-blocks from the plurality of sub-blocks; and Recording means for recording the generated plurality of subblocks and the parity data on different disks, respectively, and changing the number of divisions of the one-frame video data according to the signal system of the source video data.
- Control means for controlling the data processing means.
- the data recording apparatus is a data recording apparatus for recording the supplied source video data on a plurality of randomly accessible disks by dividing the source video data.
- Data processing means for generating parity data for correcting errors of the plurality of sub-blocks from the plurality of sub-blocks, and generating the parity data from the plurality of sub-blocks.
- Recording means for recording the plurality of sub-blocks and the parity data on different disks, respectively; and the plurality of sub-blocks and the plurality of sub-blocks so that the outer peripheral area and the inner peripheral area of each disk are used uniformly.
- Control means for changing the disk used to record the parity data for each frame. And it said that there were pictures.
- control means for controlling the data processing means so as to generate a plurality of different types of sub-blocks and parity data corresponding to the plurality of types.
- the data recording apparatus is characterized in that, in a data recording apparatus for recording supplied source data on a plurality of recording media that can be randomly accessed, one block of the source data is divided.
- a data recording apparatus for recording supplied source data on a plurality of recording media that can be randomly accessed, one block of the source data is divided.
- the above-described sub-blocks are theoretically recorded at positions where the head rotation waiting time is minimized.
- Block that generates a block map for formatting the recording area A plurality of sub-block data generated by dividing the source data based on the block map; and parity data for correcting errors of the plurality of sub-blocks from the plurality of sub-blocks.
- Data recording means for generating the subblock data and the parity data generated by the data processing means on different recording media, respectively, and a subblock having a size corresponding to the block map. And a control means for controlling the sub-block division processing in the data processing means so as to generate the sub-block.
- the data recording device in the data recording device that records the supplied source data on a plurality of recording media that can be randomly accessed, the data of one block of the source data is divided.
- the sub-blocks are recorded at a position where the head rotation waiting time is minimized, and the plurality of sub-blocks are recorded.
- the recording area of the above-mentioned disc should be such that the recording process of recording the disc on each disc and the reproducing process of reproducing the above-mentioned plurality of sub-blocks from each disc are performed within a certain time while ensuring real-time performance.
- the data of one block of the source data is divided.
- the sub-blocks are theoretically recorded so that the sub-blocks are recorded at positions where the head rotation waiting time is minimized.
- a plurality of sub-blocks having a size of a size are generated, and errors in the plurality of sub-blocks are corrected from the plurality of sub-blocks.
- a disk array control device in a disk array control device for controlling a disk array composed of a plurality of disks, generates a plurality of sub-blocks by dividing the source data.
- Data processing means for generating parity data for correcting errors of the plurality of sub-blocks from the plurality of sub-blocks, and the plurality of sub-block data and the parity data generated by the data processing means.
- Transfer means for transferring data to different disks; and a plurality of sub-packets and the parity by the data processing means so that the recording errors on the outer and inner circumferences of each of the disks are used uniformly. Control the generation process Control means.
- the disk array control method is a disk array control method for controlling a disk array composed of a plurality of disks, wherein the disk array control method is generated by dividing data of one block of the source data.
- the recording area of the disk is theoretically formed so that the above-mentioned sub-blocks are recorded at positions where the head rotation waiting time is minimized.
- This is a block map for matting, and the recording process of recording a plurality of sub-blocks on each disk and the reproduction process of reproducing the plurality of sub-blocks from each disk are fixed while guaranteeing real-time performance.
- FIG. 1 is a block diagram showing a configuration example of an editing system to which the present invention is applied.
- FIG. 2 is a block diagram showing a configuration of the editing device in FIG.
- FIG. 3 is a block diagram showing the configuration of the system control unit of FIG.
- FIG. 4 is a block diagram showing the configuration of the matrix switcher section of FIG.
- FIG. 5 is a block diagram showing the configuration of the video processing unit in FIG.
- FIG. 6 is a block diagram showing a configuration of the audio processing unit in FIG.
- FIG. 7 is a block diagram showing a configuration example of the disk recorder of FIG. 1.
- FIG. 8 is a block diagram showing a configuration example of the CPU block of FIG.
- FIG. 9 is a diagram showing an example of the ZBR tape of FIG. 8 in the case of the NTSC system.
- FIG. 10 is a diagram showing an example of the ZBR table of FIG. 8 in the case of the PAL system.
- FIGS. 11A and 11B are diagrams for explaining the relationship among cylinders, tracks, and sectors.
- FIG. 12 is a diagram illustrating the principle of ZBR recording.
- FIGS. 13 (A) and 13 (B) are diagrams showing examples of the physical address table of FIG.
- FIG. 14 is a diagram showing the format of the block map of FIG.
- FIG. 15 is a diagram showing an example of the block map of FIG.
- FIG. 16 is a flowchart illustrating the process of creating a map map.
- FIG. 17 is a diagram showing characteristics of seek time.
- FIG. 18 is a diagram illustrating the gap Sgap.
- FIG. 19 is a diagram for explaining the gap 0 gap and the skew 6> skew.
- FIG. 20 is a diagram for explaining skew between a plurality of cylinders.
- FIG. 21 is a diagram for explaining a relationship between a cylinder with a gap of 0 gap.
- FIG. 22 is a diagram for explaining the relationship of a gap 0 cylinder to a cylinder.
- FIG. 23 is a diagram for explaining the relationship of the gap 6> gap to the cylinder.
- FIG. 24 is a diagram illustrating scheduling.
- FIG. 25 is a diagram for explaining overhead.
- FIG. 26 is a flowchart for explaining the scheduling process of the controller in FIG.
- FIG. 27 is a flowchart illustrating the k-th data placement processing of the controller in FIG. 7.
- FIG. 28 is a flowchart illustrating a more detailed process of step S31 in FIG.
- FIG. 29 is a flowchart illustrating more detailed processing of step S32 of FIG. 27.
- FIG. 30 is a flowchart illustrating a more detailed process of step S51 in FIG.
- FIG. 31 is a flowchart illustrating a more detailed process in step S52 of FIG. 29.
- FIG. 32 is a flowchart for explaining the more detailed processing of step S33 in FIG.
- FIG. 33 is a diagram for explaining an arrangement of data on a hard disk realized by the processing of FIG.
- FIGS. 34 (A) and (B) are diagrams illustrating the generation of parity data.
- FIGS. 35 (A) and (B) are diagrams for explaining error correction by parity data.
- FIG. 36 is a block diagram showing a more detailed configuration example of the RAID controller and the DMA controller of FIG.
- FIGS. 37 (A) and (B) are diagrams illustrating the bit conversion processing of the RAID controller in FIG. 36.
- FIG. 38 is a view for explaining the operation of the FIFO in FIG.
- FIG. 39 is a diagram for explaining the pixel data of an image for one frame.
- FIG. 40 is a diagram illustrating a bucket transfer block.
- FIG. 41 is a diagram for explaining the DMA command.
- FIG. 42 is a diagram illustrating a state of data writing in the data buffer of FIG. 36 over time.
- FIG. 43 is a view for explaining the RAID area of one frame image.
- FIG. 44 is a diagram for explaining the configuration of the sub-block of the one-frame image shown in FIG.
- FIG. 45 is a view for explaining DMA commands at the time of reproduction when no error exists.
- FIG. 46 is a view for explaining the DMA command at the time of reproduction when an error exists.
- FIG. 47 is a block diagram showing a more detailed configuration of the RAID controller of FIG.
- FIG. 48 is a diagram illustrating parity generation processing in the RAID controller of FIG. 47.
- FIG. 49 shows the data correction process in the RAID controller in Figure 47.
- FIG. 50 is a diagram illustrating a configuration example of parity data and a subblock in the NTSC system.
- FIG. 51 is a diagram illustrating a configuration example of parity data and sub-blocks in the NTSC system.
- FIG. 52 is a diagram illustrating a configuration example of parity data and subblocks in the NTSC system.
- FIG. 53 is a diagram illustrating a configuration example of parity data and a sub-block in the NTSC system.
- FIG. 54 is a diagram illustrating a configuration example of parity data and sub-blocks in the NTSC system.
- FIG. 55 is a diagram illustrating a configuration example of parity data and subblocks in the NTSC system.
- FIG. 56 is a diagram illustrating a configuration example of parity data and subblocks in the case of the PAL system.
- FIG. 57 is a diagram illustrating a configuration example of parity data and subblocks in the case of the PAL system.
- FIG. 58 is a diagram illustrating a configuration example of parity data and subblocks in the case of the PAL system.
- FIG. 59 is a diagram illustrating a configuration example of parity data and subblocks in the case of the PAL system.
- FIG. 60 is a diagram illustrating a configuration example of parity decoding and sub-blocks in the case of the PAL system.
- FIG. 61 is a diagram illustrating a configuration example of parity data and sub-blocks in the case of the PAL system.
- FIG. 62 is a view for explaining the arrangement of sub-blocks of each frame on each disk in the data.
- FIG. 63 is a diagram illustrating the configuration of a stream.
- FIG. 64 illustrates a conventional RAID.
- FIG. 65 illustrates a conventional RAID. BEST MODE FOR CARRYING OUT THE INVENTION Preferred embodiments of a data recording apparatus and method and a disk array control apparatus and method according to the present invention will be described below.
- FIG. 1 shows a configuration example of an editing system to which the present invention is applied.
- video data is input to the editing apparatus 1 from a source video tape recorder (VTR) 9 or a directory server 7 and edited.
- the editing device 1 controls the disk recorder 2 to perform editing processing.
- the editing device 1 transmits video data and audio data to the disc recorder 2 via a 2-channel SDI (Serial Data Interface), and the disc recorder 2 transmits video data of 6 channels each. And 16 channels of audio data are supplied to the editing device 1 via SDI.
- the editing device 1 supplies video data and audio data to the main monitor 4, the on-air buffer 8, the source VTR 9, and the host personal computer 3 via SDI.
- the editing device 1 outputs an audio signal to the amplifier 5 and emits sound from the speaker 6.
- the host personal computer 3 controls the editing device 1 by sending a command to the editing device 1 using a communication protocol of the RS-422 standard.
- the editing device 1 also sends commands to the disk recorder 2, the daily server 7, the source VTR 9 and the like via the RS-422 to control each of them.
- the editing device 1 can be controlled by Ethernet via the host personal computer 3 or can control an external device.
- the disk recorder 2 is controlled by a host computer 3 via a SCSI (ANSI Small Computer System Interface).
- FIG. 2 shows an example of the internal configuration of the editing device 1.
- the editing device 1 is composed of a matrix switcher 21, a video processor 22, an audio processor 23, and a system controller 24, which are interconnected via a control bus 25. It is configured.
- the system control unit 24 controls a matrix switcher unit 21, a video processing unit 22, an audio processing unit 23, and the like via a control port bus 25.
- Device control CPU 1 22 2 1 to 1 2 2 that controls each device such as main CPU 12 1, source VTR 9, disk recorder 2 as oral call storage, and directory server 7 — Has three.
- the system control unit 24 further includes a communication CPU for performing interface processing for a GUI with the host personal computer 3 and transmitting and receiving a reference time record. It has 1 2 4.
- the matrix switcher 21 has an input line. This is a block for switching between the input and output lines.
- the matrix switch section 21 has 12 input lines 31-1 to 31-1-12 for receiving a video signal or an audio signal supplied based on the SDI format. I have.
- 12 output lines are arranged in a matrix, and each output line has an output processor 3 2 1 to 3 2 1 2 are arranged.
- the input lines 3 1-1 to 3 1-1 2 and the output processors 3 2-1 to 3 2-12 correspond to the 12 output lines, which are indicated by the Xs in the figure.
- the connection is made at the cross point as appropriate.
- the control block 34 of the matrix switcher 21 receives a command from the main CPU 121 of the system controller 24 via the control bus 25, and responds to this command. And it is made to control.
- This input line is supplied with video and audio signals from the directory server 7, the source VT R9, or the disk recorder 2, as shown in Fig. 1, as well as processed by the video processor 22. Video signals are also provided.
- the control program 34 converts a predetermined one of these inputs into one of the output processors 32-1 to 32-12 of the 12 output lines. Or supply.
- the output processors 32-1 to 32-12 perform a process of converting an input video signal into an SDI format.
- the output of the input processor 32-1 is taken as the output of the capture line and supplied to the host personal convenience unit 3.
- the output of the output processor 32-2 is used as the output of the preview line and supplied to the main monitor 4.
- combiners 33-1 to 33-10 are provided, and these combiners 33-1 to 33-10 are provided.
- — 10 indicates that the embedded audio signal supplied from the audio processing unit 23 is added to the video signal input from the corresponding input / output processor 32-3 to 32-12. Perform processing to superimpose on the I video signal.
- the outputs of the compliers 33-1 through 33-3-10 are supplied to the disk recorder 2 as a single call storage as an output of the program outline.
- the embedded audio signal means an audio signal inserted as ancillary data during a blanking period of the digital video data of the SDI standard.
- the video processing unit 22 is a block for performing an image conversion process on the video signal supplied from the matrix switch unit 21.
- the image conversion processing includes, for example, an animation effect for applying a special effect to a source video signal, a video signal having a special effect applied to a background video signal, and a background video signal. This refers to the processing of a transition effect that switches video to a foreground video signal, and the synthesis processing that mixes a first video signal and a second video signal with special effects.
- the video processing unit 22 includes a demultiplexer block 41 for extracting a key signal or a video signal (luminance signal and chroma signal) from an SDI format signal input from the matrix switch unit 21.
- Switcher block 42 for applying transition effects such as wipe to key signal or video signal, key signal Alternatively, the video signal from the special effect block 43, switcher block 42, special effect block 43, and matrix switch unit 21 for adding an animation effect such as three-dimensional image conversion to the video signal can be obtained. It comprises a mixer block 44 for mixing, and a control block 45 for controlling these demultiplexer blocks 41, switcher blocks 42, special effect blocks 43, and mixer blocks 44.
- the control block 45 receives a control signal from the system control section 24 via a control bus 25, and via a control line (not shown), a demultiplexer block 41, a switcher block 42, and a special Control the effect block 4 3 or the mixer sub-block 4 4.
- the demultiplexer block 41 has demultiplexer circuits 41 1 to 41 to 5. These demultiplexer circuits 41 to 1 to 41-5 are connected to the matrix switcher 21 to A circuit for extracting a video signal bucketed in conformity with the supplied SDI format. The circuit extracts the video signal recorded in the area of the pay mouth of each packet data. It extracts based on the synchronization signal and header information recorded at the beginning of each video signal recorded in serial form.
- the switcher block 42 includes a wipe signal generation circuit 52-1, which generates a wipe signal corresponding to the transition effect specified by the operation block in response to the control command from the control block 45. 5 2—2.
- the wipe signal generation circuit 52-1 supplies the generated wipe signal to the key signal processing circuit 51-1 and the video signal processing circuit 51-2, and the wipe signal generation circuit 52-2 provides the generated wipe signal.
- the lock signal is supplied to a key signal processing circuit 51-3 and a video signal processing circuit 51-4.
- the key signal processing circuit 51-1 processes the key signal supplied from the demultiplexer circuit 41-1 in accordance with the wipe signal supplied from the wipe signal generation circuit 52-1, or generates a new signal. Key signals are generated.
- the video signal processing circuit 51-2 is configured to process the video signal supplied from the demultiplexer circuit 41-2 in accordance with the wipe signal supplied from the wipe signal generation circuit 52-1. .
- the key signal processing circuit 5 1-3 processes the key signal supplied from the demultiplexer circuit 4 1-3 in accordance with the wipe signal supplied from the dip signal generation circuit 5 2-2. Or generate a new key signal.
- the video signal processing circuit 5 1-4 processes the video signal supplied from the demultiplexer circuit 4 1-4 in accordance with the wipe signal supplied from the wipe signal generation circuit 5 2-2. .
- One signal or video signal is respectively stored.
- the three-dimensional address generation circuit 63 receives a special image conversion control command from the control block 45 and generates a conversion address for converting a key signal or a video signal into a three-dimensional image, respectively.
- Frame memories 6 1— 1 and 6 1—2 are Readout of key signal or video signal is performed by 3D address generation circuit
- Each of the 6-in and 6-in-2 circuits performs a process for spatially interpolating the pixels of the key signal or video signal supplied from the frame memory 61-1 or 61-2. This is performed based on the conversion address from the three-dimensional address generation circuit 63.
- the mixing circuit 7 1-1 is connected to the interpolator 6 2-2 based on the image transformed key signal supplied from the interpolator 6 2-1.
- the supplied deformed video signal and the background video signal supplied from the demultiplexer circuits 41-5 are combined.
- the mix circuit 71-2 converts the video signal output from the mix circuit 71-1 and the video signal processed based on the wipe signal in the video signal processing circuit 51-2
- the key signal processing circuit 51-1 is composed so as to correspond to a key signal processed based on a wipe signal output from the wipe signal.
- the video signals output from the mix circuits 7 1-1 and 7 1-2 are supplied to two of the 12 input lines of the matrix switcher 21. .
- the separation block 81 has separation nights 81-1 to 81-3, and these separation night blocks 81-1 to 81-3 are provided.
- Reference numerals 8 1-3 demultiplex an embedded audio signal from the SDI format signal supplied from the matrix switch 21. These signals are considered to be signals of the AES IEBU (Audio Engineering Society / European Broadcasting Union) format.
- the mixer block 83 includes an adder 92-1, which adjusts the output of the separators 81-1 through 81-1-3 to a predetermined level by using variable resistors 91-1 through 91-13, and then adds the outputs. It has an adder 92-2 that adjusts the output of the separation unit 8 1-1 to 8 1-3 to a predetermined level by means of the variable resistors 91-4 to 9 1-6 and then adds them. I have.
- the embedded circuit 85 converts the audio signals output from the adders 92-1 and 92-2 into an embedded audio signal so that the audio signal can be superimposed on the SDI format signal. I do.
- the output of the embedded circuit 85 is supplied to the compensators 33-1 to 33-10 of the matrix switcher 21 and the output processor 32-2-3 to 32-2.
- the data is superimposed on the video signal converted into the SDI format by 1 and is supplied to the disc recorder 2.
- the control block 86 controls the operation of the separation block 81, the mixer block 83, and the embedded circuit 85, and transmits control signals from the system control section 24 to the control bus 25. It is made to receive and control via
- the outputs of the adders 92-1 and 92-2 are output from the speed 6 via the amplifier 5 in FIG.
- FIG. 7 shows an example of the internal configuration of the disk recorder 2.
- This disk recorder 2 is an application of the hard disk management algorithm FARAD (Fast Random Access Disk) (trademark) for high-speed real-time random access, and uses a relatively small number of disks and multi-channel data. Has been made to be able to perform high-speed random access.
- FARAD Fest Random Access Disk
- SCSI controllers 202-1 through 202-2-16 as SPCs (SCSI Peripheral Controllers) control two hard disks, respectively.
- SCSI controller 202-1 controls the hard disk 201-1 and the hard disk 201-2
- SCSI controller 202-2-2 controls the hard disk 201-3 and the hard disk 201-2. It is designed to control 2 0 1—4.
- knocker blocks 203-3-1 to 203-3-8 are provided.
- One buffer block is designed to control two SCSI controllers.
- the buffer block 2033-1 controls the SCSI controllers 2022-1 and 2022-2.
- the knocker block 203-3 controls the SCSI controllers 202-23 and 202-4.
- the data buffer 2 1 2—1 to 2 1 2—8 are assigned a uniquely defined serial address by the controller 26 2 of the CPU block 261, and the data is determined by the address.
- the storage location of can be specified. Therefore, the plurality of data buffers 2 1 2-1 to 2 1 2-8 form one data buffer when viewed from the controller 2 62.
- the disk recorder 2 can be equipped with a total of 36 hard disks, but in this configuration example, 32 hard disks are provided.
- These hard disks 201-1 through 201-32 are provided with audio blocks 231-1 and 231-2 as blocks for recording and reproducing audio data, and record video data.
- Video blocks 271-1 through 271-6 are provided for playback.
- the audio blocks 23 1—1 and 23 1—2 have the same configuration, and each is configured to process eight channels of audio data, for a total of 16 channels. The audio has been made to be able to handle overnight.
- Each of the digital oplocks 23 1-i has an input / output (I / O) controller-242-i 2) and a DMA controller 24 1-i.
- the input / output controller 242-i receives the supply of eight channels of audio signals from any of the compensators 33-1 to 33-10 of the matrix switcher section 21 in FIG. 4 and processes the signals.
- the MA controller 241-i In addition to supplying the signals to the MA controller 241-i, it also processes the audio signals for up to eight channels supplied from the DMA controller 241-i, and processes the audio signals of the editing device 1 in FIG. 4. Output to the input line of the trix switcher section 21.
- the DMA controllers 24 1-i transmit the audio signals supplied from the input / output controllers 24 2-i to the buffer controllers 2 1 1-1 through 2 1 1 via the DMA bus 25 1.
- the data read out from these data buffers 2 1 2-1 to 2 1 2-8 is read out via the DMA bus 25 1, and is DMA-transferred to the input / output controller 242-i.
- the DMA controller 28 l—i is connected to the data sofa 2 1 2—1 via the DMA bus 251.
- the video data is DMA-transferred between 2 through 2-8 and the RAID controller 282-i.
- the RAID controller 282-i performs processing for error correction on video data supplied from the video processor 283-i to be recorded on the hard disks 201-1 to 201-32. Error correction processing on the data supplied from the hard disks 201-1 to 201-32 supplied from the DMA controller 28 l-i while being supplied to the DMA controller 28 l-i. And outputs it to the video processor 283-i. Details of the processing of the RAID controller 282-i will be described later.
- the video processor 283— ⁇ converts the video data supplied from the input / output controller 284—i to the clock rate required for processing by the RAID controller 282—i, and conversely, the RAID controller
- the controller 28 converts the reproduced data supplied from the 2-i to the clock rate to be processed by the input / output controller 284-i.
- the input / output controller 284-i receives the video data supplied from any of the compensators 33-1 to 33-10 of the matrix switch unit 21 of the editing device 1, switches the video data, and switches the video data.
- the data is supplied to any of the data processors 283-1 to 283-6, and conversely, the video data supplied from the video processor 283-i is supplied to the input line of the editing device 1.
- the CPU block 261 has a controller 262 and a RAM 263, and a SCSI controller 202-i, a buffer block 203-i, and a DMA via a control bus 252. Controllers 24 1—i, I / O controllers 24 2—i, DMA controllers 28 1—i, RAID controllers 28 2—i, Video processors 28 3—i, and inputs It is connected to the output controller 284—i, and is adapted to control them as appropriate.
- the controller 26 2 controls each block, and the RAM 26 3 stores a software program and table data necessary for the controller 26 2 to execute various processes. Have been.
- the controller 26 2 uses the designated format parameters—evening and, as shown in FIG. 8, as shown in FIG.
- a block map 304 is created, and the block map 304 is appropriately updated.
- the format parameter is composed of a size S of one block as a unit of recording and reproduction, a number n for dividing the data of one block, and an optimum skew value ⁇ skew. .
- one block as a recording / reproducing unit is defined as one video frame, and the number of divisions of this one block of data is defined as the NTSC standard video signal. In the case of, it is set to "4", and in the case of a PAL standard video signal, it is set to "5".
- the controller 262 controls the operation of the DMA controller 281-i and the RAID controller 282-i with reference to the block map 304. Command to execute the command.
- the controller 26 2 reproduces the video data or audio data buffered in the data buffer 2 1 2-i when recording or reproducing the data on the hard disk 201-i.
- SCAN process This scheduling process is executed so that the head movement amount is minimized. This scheduling process is called SCAN process. This SCAN processing will be described later.
- the controller 262 also generates a physical addressable table 302 and an allocation map 303 as shown in FIG. 8, and records them in the M266.
- Zone bit recording ( ⁇ ) Table 301 is a table that associates the cylinder address of the hard disk with the number of sectors used in one track at that cylinder position.
- FIG. 9 and FIG. 10 show examples of this ZBR table.
- FIG. 9 shows an example of the ZBR table when the video signal is a video signal of the NTSC standard. Indicates a ZBR table when the signal is a PAL standard video signal.
- the hard disk 201-i is composed of a plurality of media (disks), and each medium has a plurality of tracks. Each track is divided into a plurality of sectors, and each sector has a size of 512 bytes in this embodiment.
- An area constituted by tracks located at the same diameter from the center of each medium is called a cylinder.
- the cylinder is sequentially numbered sequentially from the outer circumference toward the inner circumference, and this is used as a cylinder address.
- Each hard disk 201-i is formatted by zone bit recording, which divides the radial recording area of the disk into a plurality of zones.
- the recording surface is divided into a plurality of zones according to the distance from the center of the disk, and the outer zone is formatted so that the number of sectors per track is larger than that of the inner zone.
- the cylinder addresses are defined by cylinder addresses from 00001 to 60000.
- the cylinder addresses are formed 00 tracks, and each of these tracks is divided into 12 zones in units of 500 tracks.
- 567 sectors are formed per track.
- the number of sectors per track is 3933 in the innermost 12th zone defined by cylinder addresses 550 1 to 600.
- the outermost first zone and the innermost 12th zone are used as a pair, and so on.
- the sixth and seventh zones are used as pairs, respectively.
- the disk capacity can be used most efficiently. This point will be described in detail later.
- the ZBR table 301 contains the track W
- the physical address table 302 shown in FIG. 8 associates the logical sector number of the hard disk 201-i (this logical sector number is also called a logical block address) with the physical address. It is a table for The physical address is an address composed of a cylinder number, a media number, and a sector number.
- FIGS. 13A and 13B show examples of the physical address table 302.
- FIG. The logical sector number L ki is a number continuously assigned to each sector of all media in each hard disk 201-i. As shown in FIG. 13 (A), this logical sector number L ki is The sector number L ki can be specified by the cylinder number CYK ki, media number MED ki, and sector number SEC ki.
- Fig. 13 (B) shows a specific example.
- the sector of logical sector number 2 is the sector of sector number 2 of media number 0 of cylinder number 0. Therefore, by referring to the physical address table 302, a logical sector number can be converted into a physical sector number, or conversely, a physical sector number can be converted into a logical sector number.
- the allocation map 303 of FIG. 8 is a map for managing the recorded area of all the media (disks) of the hard disk 201-i and the address of the unrecorded area. By referring to this allocation map 303, the unrecorded area of each disc can be read. Can be searched.
- the block map 304 is a map on a table showing where data on the disk is stored and what size data is stored.
- k To manage the arrangement position on the disk for each block data (recorded data or data to be played back) having a block number indicated by (a natural number not exceeding the total number of data). It is a map of.
- one block data indicated by the block number k is, for example, a video data of one frame (image of a predetermined size), and the block number k is a frame of video data of all channels. This is a number (ID number) uniquely specified for each day. Therefore, when there is a request for video data reproduction, the arrangement position on the disc can be specified by the block number k.
- the block data for one frame is composed of n pieces of sub-block data D kl to D kn and one parity correction bit D kp for error correction.
- the location of the parity data is the disk identification number (ID) D kp of the hard disk 201-i, the logical sector number L kp indicating the recording start position of the area where the parity data is recorded, and the parity data. It is defined by the number S kp of sectors indicating the size of the recorded area.
- the location of the sub-block with the number k is such that the disk IDD ki of the hard disk 201-i, the logical sector number L ki indicating the recording start position of the area where the sub-block is recorded, and the sub-block are recorded. It is defined by the number Ski of sectors indicating the size of the area.
- n the number of divisions of one frame of video data
- NTSC or PAL data amount
- Figure 15 shows a specific example of a block map 304 used to record a video of one frame of NTSC format of 144 pixels x 512 pixels divided into four sub-blocks.
- the parity data P1 of the frame indicated by block number (frame number) 1 is 567 sectors from the start logical sector of number 0 on the hard disk whose disk ID is 1. (The outermost area).
- the data S 1—1 of the first sub-block of the frame indicated by the block number 1 is obtained from the start logical section of the number 5996 0 0 on the hard disk indicated by the disk ID 2 by 3 9 Recorded in the area of 3 sectors (the innermost area).
- data ID S 1-2 of the second sub-block of this frame is indicated by a disk ID of 3. From the starting logical sector of number 0 on the hard disk, it is recorded in an area of 567 sectors (outermost area), and the third subblock data S1-3 has the disk ID The data is recorded in the area of the 393 sectors (the innermost area) from the starting logical sector of the number 5996 0 0 on the hard disk indicated by 4, and the data S of the fourth sub-block is recorded.
- 1—4 is the opening of number 0 on the hard disk whose disk ID is 5 It is recorded in the area of 567 sectors from the start logical sector (the innermost area).
- the size of one sub-block of data is basically one-fourth the size of one frame of data, but this one sub-block of data is not necessarily the size of one frame of video 1/4 pixel data It is not data composed of evenings.
- pixel data of this sub-block is used as pixel data of a continuous 1/4 area of one frame of video data, but if this is done, the effect of missing pixel data will be growing. Therefore, a single sub-block is constructed by collecting a plurality of packet transfer blocks that consist of pixel data in a predetermined area that is distributed but not in a continuous area in one frame of video data. You. Also, the size of each sub-block is not necessarily the same. This point will be described later in detail with reference to FIGS. 43 and 44.
- the process of creating the block map 304 is a process performed before recording the source video data on each hard disk. In other words, rather than physically formatting the disk, it would be appropriate to formally format the disk using the created block map 304.
- the controller 262 controls the size of one sub-block of the video data to be recorded and the average movement of the head when performing the SCAN schedule as the format parameters for generating the block map 304.
- the number of cylinders, the number of sectors in one track, and the number of media that make up the cylinder) are determined in advance (the size of the sub-block will be described later with reference to FIG. 27). Then, the position of each sub-block on the hard disk 201 is determined by the procedure of steps S1 to S5 shown in FIG.
- the average moving distance L a of the head is given by the following equation (1) based on the total number of cylinders L t of the hard disk 201 and the number N of accesses processed in one scan (see FIG. 24 (C)).
- the drive seek time T s (L) is a function of the seek distance L (number of cylinders), the value of which depends on the mechanical properties of the disk drive used.
- Figure 17 shows an example.
- the number of access requests to be processed collectively using the scan (SCAN) algorithm depends on the characteristics of the application that uses this disk recorder 2, the performance required therefor, and the available data buffer 212. Determine from the amount. The larger the number N of access requests to be scanned at a time, the better the random access performance of the hard disk is, but the adverse effect is that the required buffer size increases and the response time increases.
- step S1 of Fig. 16 the number of subblocks (Bc) existing in one cylinder is calculated.
- the total number of sectors in one cylinder is the number of sectors in the track multiplied by the number of media. Therefore, Be is obtained by dividing the total number of sections in one cylinder by the number of sectors required to store one subblock of data.
- Gap 0 gap This is the angle difference between the first sector and the last sector of the block as viewed from the center of the disk. For example, as shown in FIG. 18, the sub-block shown with a shadow has a sector “0” of track “1” at the beginning and a sector “6” of track “2” at the end. The gap is 5/12 of the circumference, ie 57 ⁇ / 6 radians.
- the skew ⁇ skew is obtained in step S3 based on the above data.
- the skew 0skew refers to the angle difference between the heads of adjacent sub-blocks on the hard disk 201 as viewed from the center of the disk in the circumferential direction.
- T d (L) (L-B c- ⁇ skev + ⁇ gap + 2m ⁇ ) / ⁇
- L is the seek distance and the unit is the number of cylinders
- Be is the number of sub-projects in one cylinder
- 0 skew is the skew and the unit is radian
- 0gap is the gap and the unit is the radian
- ⁇ is The rotation speed (radian / second) of the hard disk 201 and m is any integer such that T d (L) is positive.
- FIG. 20 explains the meaning of the above equation (2) on the hard disk 201.
- FIG. 20 it is assumed that the access of the sub-block “0” has just finished. At this time, the head is in the direction of angle A when viewed from the center. Now, access the same sub-block “0” again. Considering that the disc must rotate for the gap of 0 gap, a waiting time of ⁇ gap / ⁇ occurs.
- To access the beginning of the inner sub-block “n” only n tracks (sub-blocks) from sub-block “0”, skew n sub-blocks from the start position of sub-block “0”. We have to wait for the rotation of the hard disk 201 by the angle of the sum with (n ⁇ skew).
- n 0 skew / ⁇ Since the hard disk 201 is rotating, the beginning of the data reaches the position where the head exists at the time obtained by adding an integer multiple of the rotation period (Trot) to the time obtained in this way. I do.
- the movement of n sub-blocks corresponds to the movement of n / Bc cylinders in the number of cylinders, so if you draw a graph with the number of cylinders on the horizontal axis and the waiting time until the top arrives on the vertical axis, As shown in 21. As the skew 0 skew increases, the slope of the linear group in FIG. 21 increases.
- the circumferential position (angle) of the head viewed from the center is assumed to be constant regardless of the distance from the center. Actually, depending on the mechanism of the head, the position may not be exactly constant, but the effect is sufficiently small that it can usually be ignored.
- the above equation (2) gives the time required for the head of the sub-block to reach below the head in each cylinder. However, since the head has to move to the desired cylinder within this time, the waiting time is the time from the seek until the beginning of the first block appears.
- This is the overhead T d (L) considering both seek and rotation wait.
- Figure 22 shows an example, and the following equation (3) shows its definition. Actual overhead (waiting time) T d (L) is shown by a thick line in FIG. The function of the seek time T s (L) is shown by a dotted line. In the figure, Trot is one rotation cycle.
- T d (L) (L-B c-0 skew + ⁇ gap) / ⁇ ⁇ ⁇ ⁇ (3)
- S 3 — 2, S 3 — 3 and S 3 — 4 indicate that this straight line is always the seek time T
- skew ⁇ skew is selected so that it is higher (larger) than s (L) and almost touches the seek time T s (L).
- step S3-2 the controller 26 2 initializes the skew Sskew to 0, and in step S3-3, the seek time T s (L) and the overhead T d Compare the size of (L). If it is determined that the seek time T s (L) is smaller than the overhead T d (L), the process proceeds to step S3-4, where the controller 26 increments the skew 0skew by delta. To Then, returning to step S3-3, the seek time T s (L) and the magnitude of the overhead T d (L) are compared again.
- the seek time T s (L) is equal to the overhead T d
- Steps S3-3 and S3-4 are repeatedly executed until it is determined that the value is equal to or larger than (L).
- step S3-3 the seek time T s (L) is changed to the overhead T d
- Steps S4 and S5 determine the position of each sub-block on the disk over the entire area of the hard disk 201 using the skew 0 skew and the gap 0 gap obtained as described above.
- step S4 first, the address of the physical address (cylinder / media / sector) is initialized to (0/0/0).
- step S5 is a loop that repeats for all sub-blocks. Inside the loop, first, in step S5-1, the logical sector number is obtained from the physical address by referring to the physical address table 302 in step S5-1. In step S5-2, this and information such as the physical address are written in the process map 304.
- step S5-3 the physical address pointer P pa is updated in step S5-3 in preparation for the processing of the next sub-block.
- step S5-4 it is determined whether or not the processing in steps S5-1 to S5-3 has been performed for all sub-blocks. If not, the processing is performed for sub-blocks that have not been processed. Steps S5-1 through S5-3 are performed.
- the arrangement of the Nth sub-workpiece on the hard disk 201 is as follows:
- a block map 304 shown in FIGS. 14 and 15 can be created.
- the video data and parity data at that time are not recorded on the disc. That is, before actually recording the video data on the disc, the block map 304 shown in FIGS. 14 and 15 is created.
- the block map 304 is used to theoretically format the recording error on the disk so that the video data and the parity data are recorded in accordance with the previously created block map 304. It can be said that it is a night. In other words, it can be said that the block map 304 is data for reserving a recording area on the disk so that video data and parity data are recorded according to the block map 304.
- the parity data P 1 of the frame indicated by the block number (frame number) 1 is the starting logical sector of the number 0 on the hard disk indicated by the disk ID 1.
- the recording area on the disk is reserved so that it is recorded in the area for the 5607 sectors (outermost area), and the first sub-block of the frame indicated by the block number 1 is reserved.
- Data S 1—1 is recorded in the area of 39.3 sectors (the innermost area) from the starting logical sector of number 996 00 on the hard disk with disk ID 2
- the recording area on the disk is reserved, and so on, starting from the number 0 on the hard disk whose disk ID is indicated by 3 in the second sub-block of this frame.
- FIG. 23 shows the overhead time of the disk on which each sub-block is arranged as shown in FIG.
- the horizontal axis represents the seek distance, that is, the number L of cylinders that the head of the hard disk 201 crosses when moving, and the vertical axis represents the time Td (L) required at that time.
- the chain line represents the seek time T s (L)
- the solid line represents the overall overhead time T d (L). Since the overhead time is the sum of the seek time T s (L) and the rotation waiting time T rd, the difference between the solid line and the one-dot chain line is the rotation waiting time T rd.
- an access request occurs to any location on the hard disk 201.
- the location to be processed by one scan may be unevenly distributed as shown in (A) and (B) in Fig. 24, or evenly distributed as shown in (C) in Fig. 24.
- five random accesses and accompanying overhead occur to move between the six access requests.
- overhead shown by the thick line in FIG. 22 occurs.
- the worst sum of overhead for these five random accesses is When the overhead function T d (L) is convex upward as shown in Fig. 23, it is when all accesses are evenly distributed as shown in Fig. 24 (C). .
- step S3 shown in FIG. 16 one of the straight line groups waiting for rotation given by the above equation (2) is above the seek time T s (L) and skewed so as to be as close as possible. Was selected. As a result, T d (L) can be reduced near the distance La, and the worst overhead T max can be reduced.
- FIG. 25 schematically shows the overhead length in the case of using the above-described algorithm (by the FARAD (trademark) method).
- the FARAD trademark
- FIG. 25 schematically shows the overhead length in the case of using the above-described algorithm (by the FARAD (trademark) method).
- the seek time can be reduced, and the overhead can be shortened accordingly.
- the seek time can be shortened as compared with the conventional case, but the rotation waiting time cannot be shortened.
- the size of the sub-block is a given fixed value, but the size of the sub-block can be selected within a certain range.
- the position of the straight line can be more finely controlled so as to approach the seek time near the average movement distance La. This point will be described in detail later.
- the overhead due to the movement between sub-blocks is greatly improved.
- the time required to change tracks and the time required to move to an adjacent cylinder must be considered.
- the time required for changing a truck and moving to an adjacent cylinder is constant, so after this time, between trucks or between cylinders, just like a night arrives under the head
- By giving the skew in the sub-block it is possible to suppress the occurrence of a long rotation wait in the sub-block due to the change of the track and the movement of the cylinder.
- step SI1 the controller 262 controls the SCSI controller 202 to move the head of the hard disk 201 to the cylinder “# 0”.
- step SI12 the process proceeds to step S12 for performing the actual schedule.
- step S 12 in step S 12-1, the controller 26 2 provides an N-order access request buffer (not shown) in the order from the oldest (oldest) one in time. ) To read from.
- the access request buffer holds the access requests supplied from the host personal computer 3 in the order of input.
- One access request describes the number of the sub-block to be accessed and the start address of the data buffer 2 12 used for data transfer.
- the number N is a constant given from the host personal computer 3 in advance.
- step S 1 2-2 the controller 26 2 refers to the block map 304 for each of the N access requests, and determines the physical address (cylinder number, media number, and sector number) of the sub-block to be accessed. Number).
- step S12-3 these N access requests are rearranged in ascending order of cylinder number (in order from the outer peripheral side to the inner peripheral side). This operation realizes the schedule of the SCAN algorithm.
- step S12-4 these sorted access requests are sent to the hard disk 201 via the SCSI controller 202 in ascending order of the cylinder number, and the actual access and data Perform overnight transfer.
- step S12-5 wait for the end of the data transfer, and then issue the next access instruction.
- step S1 2—6 it is determined whether or not all the requests have been transferred. If there are any requests that have not been transferred, the process returns to step S1 2—4, and the same processing is performed. Execute. By repeating this N times, it is determined in step S12-6 that the processing of the N access requests has been completed.
- step S12-7 the host personal computer 3 is notified that the processing of N accesses has been completed, and a series of processing relating to the N access requests is completed.
- step S12-8 all access requests are processed, and it is determined whether the access request buffer is empty. If an access request remains, the controller 26 2 returns to step S 12-1, retrieves the next N access requests, and continues processing. If there are no N requests in the access request buffer, wait in this step until N requests have been accumulated, and once N requests have been accumulated, return to step S12-1. A similar process is performed.
- the controller 26 2 determines the size of the sub-block and the recording start position of the sub-block so that the gap S gap and the skew 0 skew are substantially constant over the entire area from the outside to the inside of the hard disk 201. By doing so, the real-time access can be further improved. Actually, the controller 2626 generates the block map 304 so that the gap 0 gap and the skew 0 skew are substantially constant over the entire area from the outside to the inside of the hard disk 201.
- FIGS. 27 to 32 The processing procedure when the controller 262 places the k-th sub-block on a predetermined track of a predetermined hard disk 201 will be described.
- the hard disk 201 storing the parity data is selected, and then the hard disk 201 storing the sub-block data sequentially from the inner circumference side and the sub-block data is sequentially stored from the outer circumference side.
- Hard disk 201 to be placed is selected.
- sub-blocks are numbered from 1 to n, for example, odd-numbered sub-blocks are sequentially arranged on the hard disk 201 from the inner side, and even-numbered sub-blocks are sequentially arranged on the hard disk 201 from the outer side. Deploy. Further, the hard disks 201 to be arranged according to the data are shifted and selected so that the sub-blocks are not concentrated on a small number of hard disks 201.
- FIG. 28 is a flowchart showing details of the process in step S31 in FIG.
- MOD is an operator for calculating a remainder obtained by dividing k ⁇ 1 by m (the total number of hard disks 201).
- the size of the inner sub-block can be made smaller than the size of the outer sub-block. Yes And the gap of each sub-project 0 gap can be kept constant.o
- the sub-blocks are arranged alternately between the sub-blocks arranged sequentially from the inner circumference and the sub-blocks arranged sequentially from the outer circumference. If they do, they do not have to be alternated. That is, the sub-blocks may be sequentially arranged on the hard disk 201 from the inner circumference side, and then the remaining sub-blocks may be sequentially arranged on the hard disk 201 from the outer circumference side. .
- the j-th hard disk 201 represented by the above equations (4) and (5) need not be selected. I do not care.
- the hard disk 201 in which the sub-blocks are arranged sequentially from the inner side is selected from the hard disks 201 in which the inner area is hardly used, and the hard disk 201 in which the sub-blocks are sequentially arranged from the outer side.
- the parity data is sequentially arranged from the outer peripheral side on the hard disk 201 selected in step S41.
- step S32 on each of the selected hard disks 201, a start logical sector for starting the disposition of parity data and sub-blocks is determined.
- FIG. 29 is a flowchart showing a processing procedure for determining the start logical sector.
- step S51 sequentially from the outer peripheral side With respect to the hard disk 201 in which the parity data is to be arranged, a starting logical sector address Lkp is obtained.
- FIG. 30 is a flowchart showing details of the process in step S51 in FIG.
- step S61 the outermost track in an unallocated area where no sub-block (parity data) is allocated is selected as a track for allocating a sub-block (in this case, overnight recording).
- the unallocated area can be known from the allocation map 303. This determines the cylinder number (CY Lki) and media number (ME Dki) of the physical sector address. Referring to the ZBR table 301, the number of sectors per track at this location (Tki) I understand.
- step S62 from the value of the cylinder number (CY Lki) and the optimum skew (0skew), the following formula (6) is used to select the non-placed area in the unplaced area selected in step S61.
- the beginning of the outermost track (cylinder number CY Lki) and the beginning of the physical outermost track (the track on which the data of the first sub-block is recorded) (the track with cylinder number 0) The number of cylinders between them is CYL ki).
- step S63 the sector number (S E Cki) is obtained by the following expression (7) from the angle 0 ki obtained in the expression (6) and the number of sectors per track (Tki).
- ROUNDUP means an operator that finds an integer rounded up.
- step S64 referring to the physical address table 302 (FIG. 13), the logical sector address (Lki) is obtained from the physical sector addresses (CYLki, MEDki, SE Cki) obtained in steps S61 to S63. ) (In this case, Lkp) and return
- FIG. 31 is a flowchart showing details of the process in step S52 of FIG.
- step s71 referring to the physical address table 302 (FIG. 13), an innermost track in an unallocated area is selected as a track on which a sub-block is allocated.
- the shilling number (CYLki) and the media number (MEDki) of the physical sector address are determined.
- the ZBR table 301 the number of sectors per track (Tki) at this location is known.
- step S72 the head of the innermost track in the unallocated area selected in step S71 and the physical innermost track (the data of the first sub-block is recorded.
- the angle 0ki between the head of the track and the track is determined, and the sector number (SE Cki) is determined in step S73.
- step S74 the logical sector address (Lki) is determined from the obtained physical sector addresses (CYLki, MEDki, SE Cki) and returned.
- the steps described above The processing in S72 to S74 is basically the same as the processing in steps S62 to S64 in FIG. 30 except that tracks are selected sequentially from the inner circumference side instead of the outer circumference side. Since the processing is similar, detailed description thereof is omitted here.
- the processing procedure here is the same as the case described above with reference to the flowchart in FIG. 30, and a description thereof will be omitted here.
- the process returns.
- the track on which the sub-block is arranged is selected from the outermost track in the unallocated area
- the track on which the sub-block is to be arranged is selected from the innermost track in the unallocated area, but it is not necessarily the outermost track or the innermost track. You do not need to select. For example, in the ZBR format, if the zone is the same, the number of sectors per track is the same, and the size and gap value of each sub-block do not change.
- the processing procedure for determining the size of the sub-block will be described later with reference to the flowchart of FIG.
- each sub-block including parity data
- T Tkl + Tk2 +, ⁇ , + Tkn.
- the size Skp of the parity data is given by the following equation (11) when the size of each sub-block is determined.
- Skp M AX (Ski, Sk2, ⁇ , Skn) ⁇ ⁇ ⁇ (1 1)
- MAX is the subblock size Ski to This is the operator that finds the largest Skn.
- Tkp is the number of sections of the track in which the sub-block S kp is located.
- FIG. 32 shows a flowchart for realizing the procedure for determining the size of the sub-block described above.
- step S82 the calculation represented by the above equation (8) is performed, and the sum of the sizes Ski of the sub-blocks obtained in step S81 is equal to the original size of one block of data. It is determined whether it is equal to S. If it is determined that the sum of the size Ski of each sub-block is not equal to the size S of the original one block, the process proceeds to step S83, where the sum of the sizes S ki of each sub-block is The size Ski of each sub-block is slightly increased or decreased so as to be equal to the size S of one block of one block.
- step S82 the process returns to step S82, and the processes of steps S82 and S83 are repeated until it is determined that the sum of the sizes S ki of the sub-blocks is equal to the size S of the original one block. .
- step S82 the process proceeds to step S84, where the size of the parity data S kp is required. That is, the largest size in the sub-block S ki is set as the parity data size S kp.
- Step S84 Upon completion of the process in Step S84, the process returns. As a result, the processing in step S33 of the flowchart shown in FIG. 23 ends, and all the processing ends.
- each hard disk 201 the size of the sub-block arranged on the inner peripheral side is reduced so that the gap S gap of each sub-block is constant, and the sub-block is arranged on the outer peripheral side. Increase the size of the sub-block.
- FIG. 33 shows an example of the sub-blocks distributed to the hard disk 201 as described above.
- the number m of the hard disk 201 is 6, and the number n of subblocks is 4.
- the parity data of the first frame is stored on the hard disk 210-1 and the corresponding four sub-blocks are stored on the hard disk 20.
- Hard disk 2 0 It is arranged in a distributed manner from 1 2 to 2 0 1-5.
- Hard disk 2 0 It is arranged in a distributed manner from 1 2 to 2 0 1-5.
- the sub-blocks located on 201-5 are put on the outermost track, and the sub-blocks placed on the hard disks 201-1 and 201-4 are put on the innermost track.
- the parity data of the second frame is on the hard disk 201-2
- the sub-blocks are on the hard disks 201-1 to 201-16, respectively, from the outer circumference to the inner circumference From the inner circumference
- the parity data of the third frame is stored on the hard disk 201-3
- the sub-blocks are hard disks 201-4 through 210-1-6, and the hard disk 210-1.
- the outermost, innermost, outermost, innermost, and outermost parts are inserted in that order.
- the data of each frame is arranged, and the data of the last frame is, for example, almost in the middle of the cylinders of the hard disk 201--3 to 201-161 and the hard disk 210-1. Placed around.
- the time for actually reading or writing the parity data and the sub-block is fixed regardless of the storage location of the data on the hard disk 201. It can be.
- scanning is performed from the outer circumference to the inner circumference, but scanning may be performed from the inner circumference to the outer circumference. In this case, it is possible to set the optimum skew when accessing the data overnight while moving the head from the inner circumference to the outer circumference.
- the RAID controller 282 for generating parity data and using this to correct errors will be described.
- the sub-block The size is made larger than the size of the sub-block recorded on the inner circumference, and one block is divided into four sub-blocks # 1 to # 4.
- the size of the parity block is set to be equal to the size of the largest subblock among the divided subblocks.
- the j-th parity data P j is generated from the j-th data of each sub-block.
- Fig. 34 (A) when the value of j is small, the j-th data corresponding to all sub-blocks exists, so that the data of all sub-blocks # 1 to # 4 exist. From this, parity data P j is generated.
- j becomes larger than a certain value the j-th data does not exist in subblocks # 1 and # 3 as shown in Fig. 30 (B).
- parity data P j is generated based on the j-th data from sub-blocks # 2 and # 4 whose size is j or more.
- a predetermined data may be added to the sub-blocks # 1 and # 3, and the parity data Pj may be generated from the data of all the sub-blocks.
- the operation in correcting a data error is as follows. That is, for example, when the j-th data of a predetermined sub-block is incorrect, the error is basically corrected using the j-th data and the parity data of another sub-block.
- the i-th data Because the parity data is generated from the evening, the j-th data block of the parity data and the error-free sub-block # 1 to # 3 are used, and the erroneous sub-block # 4 J-th de-correction in the evening.
- parity data is generated from the j-th data of the sub-block whose size is j or more.
- FIG. 36 shows the configuration of a part that performs such processing.
- the RAID controller 282-1 of the video block 271-1 has delay elements 401-1 to 403-1 for delaying data by one packet transfer program. ing.
- the RAID controller 282-1 generates the parity data from the pixel data supplied from the video processor 283-1 via the 10-bit bus, and outputs the FIF0404 via the 32-bit bus. -1 and the pixel data is supplied to the FIF0405-1 via a 32-bit bus.
- the RAID controller 282-1 converts the parity data (reproduced data) supplied from the DMA controller 281-1-1 through the 64-bit bus and the FIFO 406-1 into the parity data (reproduced data).
- FIGS. 37 (A) and (B) show an example of conversion processing of 10-bit data and 32-bit data in the RAID controller 282-1.
- the RAID controller 2 282-1 is controlled by the video processor 283-1 in units of 10 bits as shown in Figure 37 (A).
- the data is supplied, three 10-bit data are collected, and a 1-bit dummy data is placed between the first and second 10-bit data. Evening is inserted, and a dummy data of 0 is inserted between the second and third 10-bit data overnight to make a total of 32 bits of data.
- the RAID controller 282-1 is the lower 2 bits of the data input with 10 bits. Ignore the bits and combine the four sets of 8-bit data into a 32-bit data set.
- the data is divided into 8 bits, and 2 bits of dummy data are added to the data of each 8 bits. And 10-bit data.
- the RAID controller 282-1 and the DMA controller 281-1 in FIG. 36 are connected to each other via FIF040s 4-1 to 406-1.
- the FIFO controllers 284—1 to 406—1 have a RAID controller 282—1 side with a 32-bit bus, and a DMA controller 281—1 side with a 64-bit bus. It consists of a bus. Since each FIFO is composed of 32 bits, for example, FIF0404-1 uses the 32-bit parity data input first as a 64-bit bus. For example, the data is output to the DMA controller 281-1-1 via the upper 32-bit bus, and the next input 32-bit parity data is transferred to the lower 32-bit bus of the 64-bit bus. It is supplied to the DMA controller 281-1 through the bit bus.
- the first 32 bits of pixel data supplied from the RAID controller 282-1 are converted to the MSB side of the 64 bits bus. It outputs to the DMA controller port 281-1-1 via a 32-bit bus, and then inputs the input 32-bit pixel data to the LSB side of the 64-bit bus. Output to the DMA controller 281-1-1 via the 32-bit bus.
- the pixel data (reproduced pixel data) supplied from the DMA controller 281-1-1 through the 64-bit bus is supplied to the RAID controller 282-1-1.
- 6-1 is composed of FIFOs 406A-1 and 406B-1, each having a capacity of 32 bits.
- the pixel data supplied via the 32-bit bus on the MSB side of the 64-bit bus from the DMA controller 281-1-1 is supplied to the FIF0406A-1 and the pixel data supplied on the LSB side is supplied to the FIF0406A-1.
- the pixel data supplied via the bit bus is supplied to the FIF0406B-1.
- the pixel data stored in the FIF04 06 A-1 is supplied to the RAID controller 282-1, via the 32-bit bus, and then stored in the FIF04 06 B-1, The obtained 32-bit pixel data is read out and supplied to the RAID controller 282-1-1 via the 32-bit bus.
- the DMA controller 281-1-1 is connected to a 64-bit DMA bus 251. Also, FIF047-1 is connected to a 32-bit control bus 252, and a command input via the control bus 252 is provided. Output to the RAID controller 282-1. Although not shown, the video blocks 271-2 to 271-6 have the same configuration as the video block 271-1.
- the board 4 21-1 is provided with three SCSI boards 43 1-1 to 43 1-3.
- the SCSI board 43-11-1 is provided with a data buffer 2112-1 composed of S-DRAM, and its input / output is controlled by the buffer controller 21-1-1. It has been made to be.
- To the sofa controller 2 1 1 1 1, two SCSI controllers 2 0 2 -1, 2 0 2-2 are connected via a 32-bit bus, and the SCSI controllers are connected.
- the first rail 202-1 controls the two hard disks 201-1 and 201-2. Further, the SCSI controller 202-2 controls two hard disks 201-3 and 201-4.
- the SCSI boards 431-1 and 431-3 have the same configuration as the SCSI board 431-1.
- the boards 4 2 1-2 and 4 2 1-3 have the same configuration as the board 4 2 1-1. Therefore, in this configuration example, one board 4 2 1 — i can control 12 hard disks 201, and three boards are provided. It has a function of controlling a total of 36 hard disks 201. However, 32 hard disks 201 are actually connected.
- one frame of image data is divided into, for example, four sub-blocks, and this sub-block further includes a packet transfer in units of 128 words of pixel data.
- Blocks are divided into blocks, and parity data generation and error correction using this are performed. Is performed.
- one frame of the image data of the NTSC system is composed of 144 ⁇ 512 pixels.
- One pixel is composed of 10 bits of luminance (Y) data and 10 bits of color difference (U or V) data.
- one word consists of 32 bits
- three 10-bit luminance data can be arranged in one word.
- the data of three pixels can be finally obtained by one word.
- the 10-bit luminance data and the 10-bit color difference data constitute a pixel-by-pixel data transfer
- one bucket transfer block (128 words)
- 1.5 pixels can be placed in one word, so that one pixel of 12-pixel data is placed.
- one pixel is composed of 10-bit luminance data, and 384 pixels are arranged in one bucket transfer block. Shall be.
- FIG. 41 shows an example of a DMA command supplied from the controller 26 2 to the MA controller 28 1 via the control bus 25 2.
- This example shows a RAID controller 282 shows an example of a command when the pixel data to be recorded is DMA-transferred from the data buffer 2 12 to the data buffer 2 12.
- this DMA command is composed of a command and an operand.
- the operand an address of a data buffer 212 for recording pixel data is stored. Is held.
- the command specifies the number of times to transfer the packet transfer block (the number of loops).
- a command for instructing various controls can be arranged.
- a RAID ratio is set in the command for the RAID controller 282.
- pixel data for one frame is AO, B0, CO, D0, A1, B1, C1, D1,...
- ⁇ It is assumed to be sequentially divided from upper left to lower right like '.
- the DMA controller 281 transmits the F This command is received by IF0501-1-1, and the data is written to the data buffer 212 as shown in Figure 42.
- addresses a0 to d0 of the data buffer 212 are defined in the operand corresponding to the four sub-programs.
- P0 is described in the operand as the address of the data buffer 2 12 to be recorded in the parity data corresponding to the four sub-blocks.
- a loop count n is described as a command.
- the packet transfer block A0 is written to the address a0 of the data buffer 212.
- the packet transfer block B0 is written to the address b0
- the packet transfer block C0 is written to the address c0
- the packet transfer block DO is written to the address d0.
- the parity data P 0 of the first one-bucket transfer work is stored in the address p 0 of the data buffer 2 12. Written.
- the packet transfer block A1 is written to the address a0 + 128 (key) of the data buffer 212, and the packet is transferred to the address b0 + 128.
- Block B1 is written
- packet transfer block C1 is written to address c0 + 128, and packet transfer block D1 is written to address d0 + 128.
- P 0 +128 “No, Retain Day P 1 is written.
- data AO, A 1, A 2,--A (n-1) are written in an area continuous from address a 0, and are written in an area continuous from address b 0.
- B 0, B 1, B 2,... B (n-1) are written.
- data C0, C1, C2,... -C (n-1) are written in an area continuous from address c0, and data is written in an area continuous from address d0.
- D 0, D 1, D 2,..., -D (n ⁇ 1) are written, and parity data P 0 to P (n ⁇ 1) are written in an area continuous from the address p 0.
- FIG. 43 shows an area R1 in which pixel data of one frame of image is transferred by a 4: 1 RAID and an area R2 in which a 2: 1 RAID is performed.
- the “n: m RA ID” used in the following description divides the source video data into n sub-blocks, and This indicates that the RAID algorithm generates m parity data from n sub-blocks. Therefore, “4: 1 R AID” is a RAID algorithm that generates one parity data from four sub-blocks.
- A0, B0, C0, D0, A1, B1, CI, D1,..., An, Bn, Cn, Dn are successive packet transfer blocks. Is transferred in a 4: 1 RAID. Then, the packet transfer blocks A (n + 1), C (n + 1), A (n + 2), C (n + 2), '', A r, : Transfer with 1 RAID.
- the above-described packet transfer blocks A i, B i, C i, and D i constitute first to fourth sub-blocks, respectively.
- the first sub-block is composed of AO to Ar packet transfer blocks
- the second sub-block is composed of B0 to Bn packet transfer blocks
- the third sub-block is composed of C0 to Cr packet transfer blocks
- the fourth sub-block is composed of DO to Dn packet transfer blocks.
- the knowledge day is composed of packet transfer blocks of P0 to Pr.
- the packet transfer block Ai constituting the first sub-block is not a continuous pixel on the image of one frame shown in FIG. (Of course, the pixels are continuous in one packet transfer block).
- the n + 1 packet transfer blocks of each subblock, together with the n + 1 parity data blocks, have a 4: 1 ratio.
- RAID is configured.
- the bucket transfer block from the bucket transfer block A n +1 of the first sub-block to the bucket transfer block Ar, and the bucket transfer block C n +1 of the third sub-block constitute a 2: 1 RAID.
- the packet transfer block in the 4: 1 RAID area is MA-transferred to the data buffer 2 12 in FIG. 36 by a DMA command as shown on the left side of FIG. 44. That is, the data A 1 of the first sub-block is stored at the address a 0 of the data buffer 2 12, and the bucket transfer block B 1 of the second sub-block is the data buffer 2 1
- the packet transfer block C1 stored in the address b0 of the second sub-block, the packet transfer block C1 of the third sub-block is stored in the address c0 of the data buffer 212, and the packet transfer block C4 of the fourth sub-block.
- D1 is stored at address d0 of data buffer 212.
- the overnight data is stored in the address p0 of the data buffer 212.
- a second packet transfer block A 2, B 2, C 2, D 2 and parity data P 2 are sequentially transferred.
- n + 1 transfers are performed for the bucket transfer block in the area of the 4: 1 RA ID.
- the packet transfer block in the 2: 1 RAID area is DMA-transferred to the data buffer 212 of FIG. 36 by the DMA command shown on the right side in FIG. That is, the packet transfer block A n + 1 of the first sub-block is stored in the address a 0 of the data buffer 2 12. It is memorized. Similarly, the packet transfer block of the third sub-block
- FIG. 45 and FIG. 46 show DMA commands for reading the bucket transfer block from the data buffer 211.
- FIG. 45 shows a case where there is no error in each packet transfer block
- FIG. 46 shows a case where there is an error.
- each packet transfer block when there is no error in each packet transfer block, the addresses a0 to d0 of the data buffer 212 where each packet transfer block is stored, and The number of transfers is described. That is, in this case, one bucket transfer block of the first sub-block is read from the address a0 of the data buffer 2 12 and the address of the second sub-block is read from the address b0. One packet transfer block of the third sub-block is read from address cO, and one packet transfer block of the fourth sub-block is read from address dO. Is read. Such reading is performed for the number of loops.
- the address c0 in which the packet transfer block having the error is stored is stored. Instead, the address P0 where the parity data is stored is described. And the top As the command, the number of loops is specified, and a flag indicating that there is an error in the data of the third sub-block stored in the address CO is described. Therefore, in this case, the data of the third sub-block is not read out from the data buffer 212, but the data of the third sub-block is read out instead.
- FIG. 47 shows a configuration example of such a part that generates parity data in real time of the RAID controller 282 and uses this to perform error correction.
- the selector 451 selects one of the recording data (pixel data) supplied from the video processor 283 and the pixel data (reproduction data) supplied from the data buffer 221. Then, the data is supplied to other circuits (not shown) as input data, and is also supplied to the input A of the delay elements 401 to 4 and the selectors 454 and 455.
- the delay elements 410-1-4 delay the input data by one packet transfer block, and then output the delayed data to the subsequent delay elements 410-13.
- the delay element 401-3 also delays the input data by one packet transfer block, and then outputs it to the subsequent delay element 4101-2.
- the delay elements 4101-1 also delay the input data by one bucket transfer block, and then output to the subsequent delay element 410-1.
- the delay element 410-1 delays the input data by one packet transfer block and then supplies it to the input A of the selector 453.
- the delay elements 4 0 1 -1 to 4 0 1 -4 are shown in FIG. Although four are shown in the block diagram, they are actually composed of one delay element 401 in FIG. 36, and this delay element 401 is divided by the number of loops (in this case, , 4), it is possible to obtain the delay of the bucket transfer block for the number of loops.
- the selector 454 selects one of the input B to which the output of the delay element 402 is supplied and the input A to which the output of the selector 451 is supplied, or exclusive ORs the two inputs. And outputs it to the delay element 402. The output of the delay element 402 is supplied to the input A of the selector 452.
- the selector 455 also selects one of the input A to which the output of the selector 451 is supplied and the input B to which the output of the delay element 403 is supplied, or exclusive ORs the two.
- the arithmetic operation is performed and output to the delay element 403, and is output to a circuit (not shown) as parity data.
- the output of the delay element 403 is supplied to the input B of the selector 452.
- the selector 452 selects one of the inputs A and B and supplies it to the input B of the selector 453.
- the selector 453 selects one of the input A from the delay element 410 1 and the input B from the selector 452, and outputs the selection result to the video processor 283.
- the RAID controller 282 converts the pixel data into units of 10 bits. Do Data is converted from data overnight to data in units of 32 bits (1 word). Then, this data is collected for 128 words (one packet transfer block), and the recorded data is supplied to the input A of the select 451. Assume that the number of the first packet transfer block input is 0. The packet transfer block with the number 0 is selected by the selector 451, and is supplied to the input A of the selector 455 and is output as it is to an unillustrated circuit as input data.
- the selector 455 selects the input A as it is.
- the output of the selector 455 (packet transfer program of number 0) is supplied to the delay element 403 and held.
- the selector 4551 selects this packet transfer block and supplies it to the selector 45.
- the input B of the selector 455 is supplied with the packet transfer block of the number 0 held in the delay element 403, so that the selector 455 is supplied from the selector 451.
- the exclusive OR of the bucket transfer block of No. 1 and the bucket transfer block of No. 0 supplied from the delay element 403 is calculated, and this is supplied to the delay element 403 for holding.
- the selector 4555 causes the buckets of No. 0, No. 1, and No. 2 to be transmitted.
- the exclusive OR of the packet transfer blocks is calculated and output, and when the bucket transfer block of number 3 is input, the exclusive logic of the bucket transfer blocks of numbers 0 to 3 is input. The sum is calculated and output. For a 4: 1 RAID, this is the desired parity.
- parity data is sequentially generated and output in real time as shown in FIG.
- the input data selected and output from the selector 451 is input to the FIFO controller 405-1 from the RAID controller 282 via a 32-bit bus, and the 64 bits are input therefrom. It is supplied to the DMA controller 28 1 through the bus of the device.
- the parity data output from the selector 455 is supplied from the RAID controller 282 to the FIFO 404 through a 32-bit bus, from which a further 6 bits are output. Supplied to DMA controller 281 via 4-bit bus.
- the MA command as shown in FIG. 41 is input to the DMA controller 281 from the controller 262.
- the MA controller 281 in response to this DMA command, reads the pixel data or parity data stored in FIF045-1 or FIF044-11, and reads the DMA bus 2 5 1 is supplied to the buffer controller 2 1 1.
- the NOF controller 211 executes a process of writing data supplied via the DMA bus 251 to the data buffer 212. Thereby, for example, the pixel data and the parity data are stored in the data buffer 212 in the state shown in FIG.
- the SCSI controller 202 receives a command from the controller 262 via the control bus 252, and in response to this command, sends a control signal to the buffer controller 211. To reproduce the pixel data and the parity data stored in the data buffer 212. Capture it. Then, the SCSI controller 202 transmits the data from the data buffer 211 captured through the north controller 211 to the predetermined data of the corresponding predetermined hard disk 201. Write to track. In this way, the hard disk 201 divides the pixel data for one frame into four sub-blocks in the state shown in FIG. 33, for example, and a different c Recorded on one disk 201. Also, the parity data corresponding to the pixel data of the frame is recorded on a different hard disk 201.
- the controller 26 2 outputs a command to the SCSI controller 202 via the control bus 25 2, and outputs a command to the pixel data of a predetermined frame recorded on the hard disk 201. Play the corresponding Paritaide overnight.
- This reproduced data is written from the SCSI controller 202 to the buffer 212 via the buffer controller 211.
- the controller 262 sends the data written to the data buffer 212 to the DMA controller 281 via the control bus 252 in this manner. Instructs DMA transfer to controller 2 82.
- the DMA controller 28 1 reads out the data written to the data buffer 2 12 via the buffer controller 2 11 1, and the DMA bus 2 1 5 Receive transfer through one. This data is supplied to FIF0406A-1 and 406B-1 via a 64 bit bus and stored. The data written to FIF04 06 A-1 and 406 B-1 is transferred to the RAID controller 282 via a 32-bit bus. Supplied to
- the data supplied from the FIFO 406A-1 and 406B-1 is received by the input B of the selector 451.
- the packet transfer block with the number 0 is input to the selector 451, in this way, it is selected by the selector 451, and as shown in FIG. 4 and to inputs A of selectors 45 4 and 45 5, respectively.
- the selector 455 is controlled to select the input B at this time, the data supplied to the input A is not supplied to the delay element 403 .
- the selector 454 since the input A is controlled so as to be selected, the selector 454 selects the packet transfer block of the number 0 and the delay element in the subsequent stage. Supply to 402 and hold.
- the packet transfer block of No. 1 is input to the selector 451, and when it is selected, it is supplied to the delay elements 401-1-4 and held in the delay elements 401-1-4 until then.
- the bucket transfer block of No. 0 that has been transferred is transferred to the delay element 401-3 in the subsequent stage and held there.
- the selector 454 outputs the packet transfer block of the number 1 from the selector 451 supplied from the input A and the packet transfer block of the number 0 held in the delay element 402.
- the exclusive OR of is calculated, and this is supplied to the delay element 402 and stored.
- the packet transfer block of number 3 is input, but now, if an error occurs in the packet transfer block of number 3, the DMA controller 281, Number 3 packet Instead of the data transfer block, the parity data is selected, read out from the data sofa 212, and supplied to the RAID controller 282.
- This parity data is supplied from the selector 451 to the delay element 401-4 and also to the input A of the selector 454.
- the selector 454 4 calculates the exclusive OR of the data from the input A and the data from the input B, but the parity data is input to the input A, and the data is input to the delay element 402. Since the exclusive OR of the packet transfer block of No. 0 and the packet transfer block of No. 1 is held, the selector 454 eventually selects the packet transfer block of No. 0 and the packet of No. 1 The exclusive OR of the transfer block and parity data is calculated, and this is output to the delay element 402.
- the bucket transfer block of No. 3 is input from the selector 451, this is supplied to the delay elements 401--4, and the parity data held in the delay elements 401--4 until then is supplied. Is supplied to the subsequent delay element 401-3. Until that time, the packet transfer block of No. 1 held in the delay element 410-1-3 is supplied to the subsequent delay element 410-2 and held there. Then, the bucket transfer work of number 0, which has been held in the delay element 411-2, is supplied to and held by the delay element 411-1 in the subsequent stage.
- the selector 454 includes a packet transfer block of No. 3 supplied from the selector 451, a packet transfer block of No. 0 held in the delay element 402, and a packet transfer block of No. 1 , And the exclusive OR with the data obtained by calculating the exclusive OR of the data.
- the result of this operation is, in the end, data obtained by correcting the error in the packet transfer block of No. 2, which is stored in the delay element 402. Is done.
- the packet transfer block of No. 4 is input to the selector 451, this is selected and held in the delay elements 410-14.
- the bucket transmission block of No. 3 held by the delay elements 410-1 to 4 is supplied to the subsequent delay element 401-3.
- the parity data stored in the delay elements 401-1-3 is supplied to the subsequent delay elements 410-1-2 and stored therein.
- the bucket transfer block of number 1 held in the delay element 401-1-2 is supplied to the subsequent delay element 411-1 and held therein.
- the bucket transfer block of number 0, which was held in the delay element 410-1 before, is supplied to the input A of the selector 453, selected by the selector 4553, and selected by the video processor 28. Output to 3.
- the packet transfer block of No. 5 is input to the selector 451, the same processing as in the case described above is performed, and the packet of No. 1 previously held in the delay element 401-1-1 is performed.
- the bucket transfer block is selected by selector 453 and output.
- the packet transfer block of No. 6 is the timing input to the selector 451, and if an error is detected in the packet transfer block of No. 6, the selector 45 1 Then, instead of the packet transfer block of No. 6, parity data is supplied. As a result, the parity data is supplied to and held by the delay elements 401-4, and is also supplied to the delay element 403 via the selector 455 and held.
- the combination of the selector 455 and the delay element 402 and the combination of the selector 455 and the delay element 403 are (4 packets per frame). (For each transfer block). Therefore, when the packet transfer blocks of numbers 0 to 3 are supplied to the delay elements 4 and 2 via the selectors 4 54, the numbers 4 to 8 of the next one frame are The data of the packet transfer block is supplied to the delay element 403 via the selector 455. Then, as in the case of the selector 545, the selector 455, when a packet transfer block of a new number is input from the selector 451, is held in the delay element 403 until then. A new exclusive OR is calculated with the packet transfer program or the exclusive OR operation result up to that point, and the operation result is supplied to the delay element 403 and held.
- the selector 45 When the packet transfer blocks of No. 4, No. 5, and No. 6 (parity data) are input from the selector 451, the selector 45 The exclusive-OR of the packet data and the result of the exclusive-OR of the packet transfer blocks No. 4 and No. 5 held in the delay element 400 is calculated. Supply to 3 and hold. On the other hand, the delay element 402 holds the data (the packet transfer block of number 2) in which the error in the previous field has been corrected.
- the delay element 411-1-1 detects the parity input in place of the bucket transfer block of No. 2 including the error. Will output the data. If this is selected and output as it is by the selector 45 3, the result will be output. Therefore, in this timing, the selector 453 selects the data from the input B instead of the data from the input A. Can be replaced. The output of the selector 45 2 is supplied to the input B of the selector 45 3, and the selector 45 2 corrects the number 2 held by the delay element 402 supplied to the input A. The bucket transfer work that has already been completed is selected and output. As a result, the parity data is not output from the switch 453 and the corrected packet transfer block of the number 2 is output.
- the error of No. 6 starts from the delay element 401-1-1.
- the parity data input in place of the packet transfer block containing the data is supplied to the input A of the selector 453.At this time, the selector 453 is switched to the input B side. Since the selector 452 selects the packet transfer program in which the error of the number 6 held by the delay element 403 supplied to the input B is corrected, this is output from the selector 453. Is done.
- the recording operation of the disk recorder 2 shown in FIG. 7 is summarized as follows. It is assumed that the video data supplied from the editing apparatus 1 to the disc recorder 2 is NTSC video data.
- the controller 26 2 of the disk recorder 2 refers to the allocation map 303 on the RAM 26 3 to edit the editing device. 1
- the supplied video data searches for recordable empty areas. This free area search is performed for each of the sub-blocks obtained by dividing the video data for one frame into four, and for each of the parity data generated from these pixel data. At this point, the number of sectors in the Nori-Tide area and the number of sectors in the four sub-blocks have not yet been determined accurately, so this free area search is performed not on a sector-by-sector basis but on a track-by-track basis. Done.
- the controller 262 determines a recording position for recording the overnight data and the data of the four sub-blocks on the hard disk 201 based on the status of the searched empty area.
- This recording position is specified by the disk ID of the hard disk 201 and the starting logical sector number of the sector from which recording is started. Specifically, as shown in FIG. 15, in the parity data, the recording start position is the logical sector number 0 (the outermost track sector) of the hard disk 201-1 with the disk ID of 1. In the first sub-block, the recording start position is the logical sector number 59996 (the innermost track sector) of the hard disk 201-1-2 having the disk ID of 2.
- the recording start position is logical sector number 0 (the outermost track sector) of the hard disk 201-3 whose disk 1D is 3. Further, in the third sub-block data, a logical sector number 59996 (the innermost track sector) of the hard disk 201-4 having a disk ID of 4 is set as a recording start position. In the fourth sub-block data, the recording start position is the logical sector number 0 (the outermost track section) of the hard disk 201-5 whose disk ID is 5.
- the controller 26 2 refers to the physical address table 302, and comprises a cylinder address, a media number, and a sector number from the starting logical sector number determined as the recording position for the data recording. Ask for a physical address.
- the controller 26 2 refers to the ZBR table 301 to determine which of the 12 zones includes the cylinder address obtained by referring to the physical address table 302. And the number of sectors used in one track accordingly. Further, the controller 262 determines the disk ID, the logical sector number, and the number of sectors of the parity data obtained by the above control, the disk IDs of the first to fourth sub-blocks, and the logical sector number. Generate a block map 304 based on, and the number of sectors. As described above, a block map corresponding to the video data designated as the first frame (the frame of the number 1) is created.
- this sequence program includes an operand including an address for indicating a storage position of the parity data and the video data on the data buffer 212, the parity data and the parity data. It consists of a command to specify the number of loops for packet transfer of video data and the RAID ratio.
- the controller 262 transfers the command part of this sequence program to the FIFO 407 via the control bus 252.
- the MID controller 282 refers to this command supplied via FIF0407, and detects the ratio of video data to parity data (RAID ratio) and the number of transfer loops (transfer loop period). I do.
- the RAID controller 282 receives this command, generates parity from the video data supplied from the video processor 283 at a RAID ratio of any ratio, and executes the processing. , How many times to repeat.
- the RAID controller 282 calculates parity data from four bucket transfer blocks (sub-blocks) when the RAID ratio is 4: 1, and transfers two packets when the RAID ratio is 2: 1. Parity data is calculated from a block (sub block).
- the calculated parity data is supplied to the FIFO 404 every one word, and the pixel data is supplied to the FIF 405 every word.
- FIG. 50 schematically shows the size of one sub-block and one parity block of the frame indicated by number 1 in FIG.
- the size of the first sub-block S1-1 is 3939 sectors
- the size of the second sub-block S1-2 is 567 sectors
- the size of the third subblock is
- the size of S 1-3 is 3939 sectors
- the size of the fourth sub-block S 1-4 is 567 sectors.
- the size of the parity data is set as 567 sectors corresponding to the largest size of 567 sectors among the sizes of the first to fourth sub-blocks.
- the size of the fourth sub-block and the parity data is the same as the size of the second sub-block, and the size of the third sub-block is the same as the size of the first sub-block.
- the pixel data of the four sub-blocks is not obtained by dividing continuous pixels constituting one frame of image into quarters but by one pixel. It is a collection of a predetermined number of 128-word (bucket transfer program) units of data distributed at predetermined positions on the image.
- parity data is sequentially generated for the input pixel data, so that the packet transfer processes of the four sub-blocks are sequentially input. Then, the data is sequentially output to the data buffer 212, and the parity data corresponding to the RAID ratio is generated one by one at a time in the RAID ratio, and is output to the data buffer 212.
- transfer is performed in response to a DMA command that specifies a 4: 1 RAID.
- the first through fourth addresses are stored in the second through fifth addresses of the data buffer 201 corresponding to the hard disks 201 through 2 through 201 through 5 having the disk IDs 2 through 5, respectively.
- the video data (503004 pixel data) of the subpacket's 393 packet transfer block is DMA-transferred and written.
- the parity of 393 bucket transfer blocks is transferred to the first address of the data buffer 2 1 2 corresponding to the hard disk 2 0 1—1 having the disk ID 1 by 3 9 3 times loop transfer.
- the data (503004 words of parity data) is stored.
- the DMA controller 281 discards a pixel data of 174 sectors corresponding to the range T2 of the second sub-block.
- a disk ID of 5 To the eighth address of the data buffer 211 corresponding to the drive disk 201-5 (the video data of 393 sectors of the fourth sub-program of the range T1—the address following the area where the evening was recorded).
- parity data (2,272,2-word parity data) of the 174 bucket transfer blocks is stored in the sixth address (range) corresponding to the hard disk 210-1, having a disk ID of 1. At the next address of the area where the parity data for 393 sectors of T1 was recorded).
- the video data of the frame with the number 1 in FIG. 15 is transferred to the data buffer 212 and stored.
- the video data of the numbers 2, 3,... ′ Are sequentially transferred to the data buffer 21 and are buffered.
- Fig. 50 shows the parity data per zone on the outermost track of the ZBR table shown in Fig. 9 per zone 1 track with 567 sectors, and 3 per sector on the innermost track.
- This is an example of a case where recording is performed on a track of zone 12 having three sectors. That is, for example, the parity data of 567 sectors is recorded on the track of the outermost zone 1 of the hard disk 201-1, and the data of the first sub-block of 393 sectors among the sub-blocks is recorded.
- the data is recorded on the innermost track of zone 12 on the hard disk 201, and the second sub-block data of the section 567 is recorded on the hard disk 201-3.
- the data of the third sub-block of 393 sectors is recorded on the track of zone 1 on the outer circumference, and is recorded on the track of zone 12 on the innermost circumference of the hard disk 201-4, and 567 sectors Of the fourth sub-block of the minute, zone 1 on the outermost circumference of hard disk 201--5 Recorded on the track.
- the outermost track and the innermost track are used in pairs. Therefore, for example, if the pair of Zone 1 and Zone 1 2 becomes full and becomes unusable, then after V-one 1, a zone with 544 sectors per track located on the outer circumference side After zone 2 and zone 1 2, zone 11 having 4 16 sectors per track located on the inner circumference side is used as a pair.
- FIG. 51 shows a configuration example of parity data and sub-block data in this case.
- the parity data of 544 sectors is recorded on the track of the second zone 2 from the outermost periphery of the hard disk 201-2, for example, and 4 16 sectors of the sub block are recorded.
- the data of the first sub-block of the minute is recorded on the second track from the inner circumference of the hard disk 201-3, which has 4 16 sectors per track, and the second 5
- the data of 4 sub-blocks of 4 sectors is recorded on the track of zone 2 of hard disk 201-4
- the data of the 3rd sub-block of 4 16 sectors is recorded on hard disk 201-5
- the data is recorded on the track of zone 11 of zone 4
- the data of the fourth sub-block of 544 sectors is recorded on the track of zone 2 of hard disk 201-6.
- a 4: 1 RAID is performed, and in a range T 2 of 128 sectors thereafter.
- a 2: 1 RAID is done.
- FIG. 52 to FIG. 55 use zone 3 and zone 10 as a pair, use zone 4 and zone 9 as a pair, use zone 5 and zone 8 as a pair, or use zone 6 and Zone 7 It shows an example in the case of using.
- FIGS. 50 to 55 show the configuration in the case of the NTSC system.
- the video data is in the PAL system, as shown in FIG. 56 to FIG.
- Data and sub-block sizes can be set.
- the number of pixels in one frame is 144 ⁇ 612, which is larger than in the case of the NTSC system.As shown in Fig. 10, the number of sections per track in each zone is specified.
- FIGS. 56 to 61 show the case where zone 1 and zone 12 in FIG. 10 are used as a pair, the case where zone 2 and zone 11 are used as a pair, and the case where zone 3 and zone 10 are used as a pair.
- zone 4 and zone 9 are used as a pair
- zone 5 and zone 8 are used as a pair
- zone 6 and zone 7 are used as a pair
- the parity of 561 sectors is recorded on the track of zone 1 on the outermost periphery of the first hard disk 201
- the 391 sections of the sub-program are recorded.
- the data of the first sub-block is located on the track of zone 12 on the innermost circumference of the second hard disk, and the data of the second sub-block of 561 sectors is allocated to the third hard disk.
- the data of the third sub-block of 391 sectors is recorded in the track of zone 1 of the third hard disk, and the data of the third sub-block of the fourth hard disk is recorded in the track of zone 4 of the fourth hard disk. Is recorded on the track in zone 1 of the fifth hard disk, and the data of the fifth sub-block of 391 sectors is recorded on the innermost circumference of the sixth hard disk. Recorded on tracks in Zones 1 and 2.
- the data in the range T1 for 391 sectors from the left side is transferred with a 5: 1 RAID
- the data for the 170th sector on the right side is transferred with a 2: 1 RAID. Is done.
- the size of one packet transfer block is 128 words. Therefore, in the range T1, the transfer loop is performed 391 times, and in the range T2, the transfer loop is performed 170 times.
- step S12-8 of the flowchart in FIG. 26 for example, when the parity data and the subblock data for one frame are stored in the data buffer 212, the one buffer is stored.
- the parity data for one frame and the sub-block data are not immediately transferred to the hard disk 201.
- the parity data and the sub-block data for 10 frames are transferred to the data buffer 2 1 2
- the order of the data transfer is scheduled by the controller 262 according to the SCAN algorithm.
- the number of hard disks 201 is six, and data is to be recorded as shown in FIG. 62, data to be transferred to the hard disk 201 with a disk ID of 3 is assumed.
- the data buffered at the third address indicating the location holding the evening is the second sub-block of frame number 1, which is S 1-2, the frame number. 2nd first sub-block data S2_1, frame number 3 parity data P3, frame number 5 fourth sub-block data S5-4, frame number 6 third sub-block Data S6-3, second sub-block data S7-2 of frame number 7, first sub-block data S8-1 of frame number 8, parity data P9 of frame number 9, and the like.
- the sub-block data S 1-2 is arranged on the outer peripheral side
- the sub-block data S 2-1 is arranged on the inner peripheral side
- the parity data P 3 is arranged on the outer peripheral side
- the sub-block data S 5-2 is arranged on the outer peripheral side. 4 is located on the outer side
- sub block data S 6-3 is located on the inner side
- sub block data S 7-2 is located on the outer side
- sub block data S 8-1 is located on the inner side.
- the parity data P9 is arranged on the outer circumference side. Therefore, these data are processed in order from S1-2, S2-1, P3, S5-4, S6-3, S7-2, S8-1, P9. If data is recorded in the hard disk, the head of the hard disk with disk ID 3 may move between the outer circumference and the inner circumference of the hard disk, causing slow access.
- the controller 262 refers to the created block map 304 (FIG. 15) with respect to the video data of the frame numbers 1 to 10 and, by referring to the created block map 304 (FIG. 15), processes these multiple data. Scheduling is performed such that data is transferred to the hard disk 201 in order from the data located on the outermost side. That is, referring to the block map 304 in FIG. 15 and rearranging in order from the data recorded on the outermost side, the data is represented by S 1 ⁇ 2, P 3, S 5 ⁇ 4 , S 7-2, P 9, S 8-1, S 6-3, S 2-1. Controller In this way, by scheduling the order of the data arrangement, the amount of head movement for recording these multiple data arrangements is minimized and the access speed is improved. Let it do.
- the operation is the reverse of the case of recording.
- the controller 26 refers to the block map 304 to start the logical sector of the frame of the number instructed to reproduce.
- the physical address corresponding to the logical sector is obtained by referring to the physical address table 302.
- the controller 62 requests the access to the physical address thus requested to the buffer controller 21 1 via the control bus 25 2.
- the sofa controller port 211 controls the SCSI controller 202 to reproduce data from the address specified by the physical address of the hard disk 201. .
- This playback data is supplied from the SCSI controller 202 to the buffer controller 211, and further written to the data buffer 212.
- the controller 262 waits until the reproduction request for 10 frames is collected, and when the reproduction request for 10 frames is received, the controller 26 2 sequentially starts from the outer peripheral side. Scheduling in an order that can minimize the amount of movement of the node. Then, according to this schedule, the buffer controller 212 is requested to reproduce 10 frames of image data.
- Controller 26 2 also refers to block map 304, A sequence program for causing the RAID controller 282 to perform DMA transfer of the data stored in the data buffer 212 is generated. This sequence program is supplied to the FIF0 501 of the DMA controller 281 via the control bus 252. In the sequence program (DMA command), as described above, the address of the data buffer 212 is specified in the operand, and the number of loops is specified in the command. In response to this command, the DMA controller 28 1 reads out the data written in the data buffer 21 2, receives it via the DMA bus 25 1, and further converts this data into RAID data. Transfer to controller 2 8 2.
- DMA command the sequence program (DMA command)
- the DMA controller 28 1 reads out the data written in the data buffer 21 2, receives it via the DMA bus 25 1, and further converts this data into RAID data. Transfer to controller 2 8 2.
- the RAID controller 282 sequentially corrects the input data in real time and sequentially as described above.
- the error-corrected video data output from the RAID controller 282 is supplied to the video processor 283, and its clock is changed to a lower-frequency clock. Then, Isseki de this output from the video processor 2 8 3, c Conclusions supplied from the input-output configuration Bok roller 2 8 4 Conclusions re Kkususui Tcha portion 2 1 of the editing device 1 through the SD I
- the leak switch unit 21 causes the video processor 32-2 to convert the video data input from a predetermined channel of the SDI into a video signal of the SDI format, and to the main monitor 4. Output and display. In this way, the user can view the image recorded on the hard disk 201.
- the video switcher 21 of the editing device 1 converts the video data captured from the disc recorder 2 into the video processor 2. Entered in 2.
- the demultiplexer circuit 4111 takes in the key signal included in the input SDI format video packet and outputs it to the key signal processing circuit 51-1 .
- the demultiplexer circuit 41-2 outputs a video signal component extracted from the input SDI format video bucket to the video signal processing circuit 51-12.
- the control block 45 of the video processing unit 22 sends the command from the host personal computer 3 via the control bus 25. And controls the wipe signal generation circuit 521 in response to this command.
- the wipe signal generation circuit 52-1 generates a wipe signal corresponding to the command from the control block 45, and supplies the wipe signal to the key signal processing circuit 51-1 and the video signal processing circuit 51-2. I do.
- the key signal processing circuit 5 1-1 and the video signal processing circuit 5 1-2 process the key signal and the video signal according to the wipe signal input from the wipe signal generation circuit 5 2-1, respectively. Output to the circuit 7 1-2.
- the demultiplexer 41-3 and the demultiplexer circuit 41-14 extract the key signal and video signal of the other channels, respectively. Output to the key signal processing circuit 51-3 and the video signal processing circuit 51-4.
- the wipe signal generation circuit 52-2 generates a wipe signal in response to a command from the control block 45, and outputs the same to the key signal processing circuit 51-13 and the video signal processing circuit 51-4.
- the key signal processing circuit 51-3 and the video signal processing circuit 51-4 are derived from the wipe signal generation circuit 522-2.
- the key signal and the video signal are processed according to the input wipe signal, and output to the frame memory 61-1 and the frame memory 61-2, respectively.
- the three-dimensional address generation circuit 63 receives a command from the host personal computer 3 via the control block 45, and stores three-dimensional address coordinates corresponding to this command in the frame memory 61-1 and the frame memory 6. Output to 1-2.
- the key signal and the video signal converted into three-dimensional coordinates are read from the frame memory 61-1 and the frame memory 61-2, respectively, and the interpolator 62-1 and the interpolator are read out. Supplied at 6 2—2 each.
- the input and output signals correspond to the three-dimensional coordinate addresses from the three-dimensional address generation circuit 63 for the input key signal or video signal, respectively. And performs interpolation processing, and outputs the result to the mix circuits 71-2.
- the mix circuit 71-1 is also supplied with a background video signal extracted from a predetermined input of the matrix switcher 21 by the demultiplexer circuit 41-5.
- the mix circuit 71-1 converts the video signal with special effects output from the interpolator 62-1, 62-2 and the background video signal output from the demultiplexer circuit 41-5.
- the signals are appropriately mixed, and the mixed video signals are output to the mix circuits 71-2.
- the mix circuit 7 1-2 outputs the output of the mix circuit 7 1-1 to the output of the key signal processing circuit 5 1 1 1 with a transition effect and the output of the video signal processing circuit 5 1-2. It mixes with the output as appropriate and outputs the mixed video signal.
- the video signal processed by the video processing unit 22 is appropriately selected, and the output processor 32 is output. — Supplied to any one of 3 to 3 2 — 12, and further supplied to disc recorder 2 again from one of compiners 3 3 1 to 3 3 — 10, to hard disk 201 Written.
- the control block 34 selects this as appropriate and supplies it to the audio processing unit 23.
- the audio processing section 23 separates the embedded audio signal from the input SDI signal into which the separations 81-1 to 81-3 are supplied, and supplies the separated signal to the mixer block 83.
- the control block 86 sets the variable resistor 91-1 to 9-11 as appropriate to a predetermined value. Adjust to As a result, the audio signal separated at 8 1 1 to 8 1-3 is adjusted to a predetermined level, and then supplied to the adder 9 2 1 1 or the adder 9 2-2 and added. Is done.
- the audio signal output from the adder 92-1 or the adder 92-2 is input to the embedded circuit 85, compressed on a time axis, and converted to an embedded audio signal. It is supplied to a predetermined one of the compensators 33-1 to 33-10 of the matrix switcher section 21. Compilers 33-1 through 33-10.0 superimpose the input audio signal on the vertical blanking interval of the video signal input from the output processor 32-3-3-3-12. And disk recorder To the hard disk 201 for recording.
- the audio signal output from the adder 92-1 or 92-2 of the mixer block 83 is output from the speaker 6 via the amplifier 5.
- the audio processing in the disk recorder 2 is not described, it is processed in the same manner as video data.
- the video data processed by the editing device 1 is supplied from the output processor 32-1 to the host personal computer 3 as appropriate.
- Stream A is composed of temporally continuous frames A1 to A5 and the like
- stream B is composed of temporally consecutive frames B1 to B5 and the like
- Ream C is composed of temporally continuous frames C1 to C5.
- frame The frames Bl to B5 or the frames CI to C5 are streams recorded sequentially in the order of the numbers indicated by the subscripts.
- the editing operator operates the host computer 3 and, for example, the frames A4, It is assumed that an edit list is generated so that each frame is reproduced in the order of frame Bl of stream B, frame A1 of stream A, and frames C3 and C1 of stream C.
- the host computer 1 controls the disk recorder 2 so that each frame specified in this order is reproduced in real time. Note that the streams of frames A4, B1, Al, C3, and C1 are not actually recorded on the hard disk 201, but are simply played back in the form of an edit list.
- the order of the frames is specified by the host personal computer 3 side.
- the controller 262 controls the access order of each frame specified in the order of the shortest seek distance based on the SCAN algorithm according to the address of each frame obtained from the block map 304.
- a plurality of sub-blocks constituting each frame of the multi-channel video data are stored at random positions on each disc, but each sub-block has a minimum head latency. Since the data is stored according to the FARAD algorithm, it is possible to read out a plurality of sub-blocks constituting each frame of the video data of the multi-channel while guaranteeing real-time performance.
- the parity data associated with the RA ID algorithm is also read at the same time, so the CPU 262 uses the parity data to read the plurality of sub-blocks with errors. Control the RA ID controller to make corrections.
- controller 262 restores the original frame data from the plurality of read sub-block data because each sub-block is divided by the RAD algorithm.
- each frame data is rearranged and output as stream 1.
- an editing list composed of frames B3, B2, C1, Al, and A2 is generated by the host computer 3, and the editing list is recorded from the computer 3 by a disk recorder.
- the stream 1 based on the edit list of the first video channel and the stream 2 based on the edit list of the second video channel simultaneously have real-time characteristics. It can be played back from the disc recorder 2 while guaranteeing. And the book According to the editing system of the present invention, it is possible to perform signal processing on the reproduced video data of a plurality of channels in real time.
- the number of channels of each block, bus, SDI, and the like in the above embodiment is merely an example, and can be changed as necessary. Also, the number of sub-blocks that divide one frame of data and the number of words of a packet transfer work that further divides each sub-work can be appropriately changed. In addition, the RAID ratio can be set to any value.
- video data of a plurality of channels is simultaneously reproduced from a disc based on the editing information, and the reproduced plurality of channels are reproduced. Since the video data of the channel is edited based on the editing information and the video data of at least one channel is generated, quick editing is possible.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Signal Processing For Digital Recording And Reproducing (AREA)
Description
Claims
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP98911232A EP0905693A4 (en) | 1997-04-07 | 1998-04-07 | DEVICE AND METHOD FOR RECORDING DATA AND DEVICE AND METHOD FOR DRIVE STACK |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP10394497 | 1997-04-07 | ||
JP9/103944 | 1997-04-07 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO1998045840A1 true WO1998045840A1 (fr) | 1998-10-15 |
Family
ID=14367561
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP1998/001597 WO1998045840A1 (fr) | 1997-04-07 | 1998-04-07 | Dispositif et procede d'enregistrement de donnees et dispositif et procede de commande de pile de disques |
Country Status (4)
Country | Link |
---|---|
EP (1) | EP0905693A4 (ja) |
KR (1) | KR20000016425A (ja) |
CN (1) | CN1228187A (ja) |
WO (1) | WO1998045840A1 (ja) |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001166887A (ja) | 1999-12-08 | 2001-06-22 | Sony Corp | データ記録再生装置及びデータ記録再生方法 |
US6891690B2 (en) * | 2002-11-20 | 2005-05-10 | International Business Machines Corporation | On-drive integrated sector format raid error correction code system and method |
US7814128B2 (en) * | 2003-05-30 | 2010-10-12 | Symantec Operating Corporation | Multi-volume file support |
US7562230B2 (en) * | 2003-10-14 | 2009-07-14 | Intel Corporation | Data security |
US7711965B2 (en) | 2004-10-20 | 2010-05-04 | Intel Corporation | Data security |
KR100694300B1 (ko) * | 2005-12-08 | 2007-03-14 | 한국전자통신연구원 | 블록 분할 분산 패리티 디스크 어레이에서의 데이터 배치및 제어 방법 |
US20150222705A1 (en) * | 2012-09-06 | 2015-08-06 | Pi-Coral, Inc. | Large-scale data storage and delivery system |
CN104168253A (zh) * | 2013-05-17 | 2014-11-26 | 环达电脑(上海)有限公司 | 保护网络上传信息的方法及储存控制系统 |
GB2519815A (en) * | 2013-10-31 | 2015-05-06 | Ibm | Writing data cross storage devices in an erasure-coded system |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0738854A (ja) * | 1993-07-20 | 1995-02-07 | Matsushita Electric Ind Co Ltd | ディスク記録再生方法およびディスク記録再生装置 |
JPH07235151A (ja) * | 1994-02-21 | 1995-09-05 | Sony Corp | 映像送出装置 |
JPH09200673A (ja) * | 1996-01-23 | 1997-07-31 | Sony Corp | データ記録再生装置 |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5457703A (en) * | 1990-11-21 | 1995-10-10 | Hitachi, Ltd. | Array disk system and control method thereof |
JPH08203209A (ja) * | 1995-01-26 | 1996-08-09 | Alps Electric Co Ltd | ディスク媒体及びディスク媒体への書き込み方式 |
US5742443A (en) * | 1996-05-31 | 1998-04-21 | Industrial Technology Research Institute | Method and apparatus for data placement of continuous media to utilize bandwidth efficiency |
EP0834878A2 (en) * | 1996-10-04 | 1998-04-08 | Sony Corporation | Method and device for controlling access to a disc storage device |
-
1998
- 1998-04-07 WO PCT/JP1998/001597 patent/WO1998045840A1/ja not_active Application Discontinuation
- 1998-04-07 KR KR1019980710009A patent/KR20000016425A/ko not_active Application Discontinuation
- 1998-04-07 CN CN98800769A patent/CN1228187A/zh active Pending
- 1998-04-07 EP EP98911232A patent/EP0905693A4/en not_active Withdrawn
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0738854A (ja) * | 1993-07-20 | 1995-02-07 | Matsushita Electric Ind Co Ltd | ディスク記録再生方法およびディスク記録再生装置 |
JPH07235151A (ja) * | 1994-02-21 | 1995-09-05 | Sony Corp | 映像送出装置 |
JPH09200673A (ja) * | 1996-01-23 | 1997-07-31 | Sony Corp | データ記録再生装置 |
Non-Patent Citations (1)
Title |
---|
See also references of EP0905693A4 * |
Also Published As
Publication number | Publication date |
---|---|
EP0905693A4 (en) | 2001-02-14 |
KR20000016425A (ko) | 2000-03-25 |
EP0905693A1 (en) | 1999-03-31 |
CN1228187A (zh) | 1999-09-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP4126713B2 (ja) | 画像再生装置および画像再生方法 | |
JPH11296975A (ja) | 編集装置および方法、並びに提供媒体 | |
WO1998045845A1 (fr) | Systeme et procede d'edition | |
US5991496A (en) | Recording/reproducing apparatus and method thereof | |
WO1998045840A1 (fr) | Dispositif et procede d'enregistrement de donnees et dispositif et procede de commande de pile de disques | |
JP4499193B2 (ja) | 記録再生装置及び記録再生方法 | |
JP4265019B2 (ja) | データ蓄積装置および方法、ならびに、データ送出装置および方法 | |
JPH10334596A (ja) | ディスク装置 | |
JPH10283126A (ja) | データ記録装置および方法、並びにデータ再生装置および方法 | |
JPH10285554A (ja) | 誤り訂正データ生成装置および方法、誤り訂正装置および方法、並びにデータ記録装置および方法 | |
JPH10285553A (ja) | データ記録装置および方法 | |
JPH10285524A (ja) | 編集装置および方法 | |
JPH11306671A (ja) | 編集装置および方法、並びに提供媒体 | |
JPH10289204A (ja) | 調停装置および方法 | |
JPH11296971A (ja) | 編集装置および方法、並びに提供媒体 | |
JPH11306670A (ja) | 編集装置および方法、並びに提供媒体 | |
JPH11296972A (ja) | 編集装置および方法、並びに提供媒体 | |
JPH11296973A (ja) | 編集装置および方法、並びに提供媒体 | |
JP2009055503A (ja) | 再生方法および装置 | |
JPH11306668A (ja) | 編集装置および方法、並びに提供媒体 | |
JP4206240B2 (ja) | 情報記録媒体、同時記録の方法、同時再生の方法、情報記録装置および情報再生装置 | |
JPH11296974A (ja) | 編集装置および方法、並びに提供媒体 | |
JP4131032B2 (ja) | データ再生装置及び方法 | |
JPH11306669A (ja) | 情報処理装置および方法、並びに提供媒体 | |
MXPA98010248A (en) | Device and method for recording data and device and method for controlling disk array |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 98800769.X Country of ref document: CN |
|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): CN JP KR MX US |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE |
|
WWE | Wipo information: entry into national phase |
Ref document number: PA/a/1998/010248 Country of ref document: MX |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1998911232 Country of ref document: EP Ref document number: 1019980710009 Country of ref document: KR |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWP | Wipo information: published in national office |
Ref document number: 1998911232 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 09194937 Country of ref document: US |
|
WWP | Wipo information: published in national office |
Ref document number: 1019980710009 Country of ref document: KR |
|
WWW | Wipo information: withdrawn in national office |
Ref document number: 1998911232 Country of ref document: EP |
|
WWW | Wipo information: withdrawn in national office |
Ref document number: 1019980710009 Country of ref document: KR |