JP2006048871A - Apparatus and method for recording information, and program and recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an information recording apparatus for surely recording two or more stream data in a hard disk device. <P>SOLUTION: The information recording apparatus records stream data from two or more cameras on a continuous cluster area and packet data approximately in order of arrival in ascending order of cluster, and saves them as one file. Thereafter, the apparatus divides them into files for each of the stream data from each camera, and performs defrag processing in the files and/or releases the above continuous cluster areas by file transfer. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an information recording apparatus, and more particularly to an information recording apparatus and method for recording video information on a hard disk device.

  2. Description of the Related Art In recent years, hard disk devices have been widely used as recording media for information recording devices that record video information arriving via a network such as broadcast content and surveillance camera video. Since such video information is improved in image quality according to the demands of the times, high-speed technology is required to record the incoming high-quality video on the hard disk device without missing it. In order to record information on the hard disk device at a high speed, a sequential recording method of sequentially recording information in the ascending order (or descending order) may be employed. This is because the seek operation of the head can be minimized.

  Here, a recording method when video information (hereinafter referred to as stream data) from a plurality of video sources in the conventional information recording apparatus is simultaneously recorded on the hard disk device will be described with reference to FIG. FIG. 10 shows a software module for recording stream data on the hard disk device 1006.

  Processes 1001 to 1003 are modules for recording stream data from separate video sources on the hard disk device 1006, and recording the stream data to the hard disk device 1006 via the file system 1004 and the device driver 1005 according to a predetermined API. Execute the process. Recording processing in each process is executed independently, and stream data from each video source is recorded as a file by a separate file writing process.

Japanese Patent Laid-Open No. 5-35620

  Generally, in the file system, a device is devised so that a file written by one file writing process is arranged in a continuous sector area on the disk as much as possible.

  When a plurality of files are written to the disk by a plurality of writing processes as in the above conventional example, there is a problem that the head seek amount increases between the writing processes and the recording performance is lowered.

  There has been a need for a recording method capable of drawing out the recording performance of the hard disk device as much as possible with respect to a plurality of stream data that arrive at the same time and preventing the occurrence of stream data loss.

  The present invention has been made in order to solve the above-mentioned problems. File recording means for recording stream packet sequences from a plurality of video sources as a single file in a continuous area on a hard disk; and File dividing means for dividing the data into a plurality of files, and the file recording means includes one stream packet for one unit area and identification information for identifying the video source of the stream packet. Recording is performed in a state, and the file dividing means operates the unit area based on the recorded identification information.

  The present invention also provides a recording means for recording a plurality of stream data coming simultaneously as one file in a continuous area on the hard disk, a file dividing means for dividing the file into a plurality of files for each stream data, and the file In addition to the file division by the dividing means, there is a file rearrangement means for canceling the fragmented state of the divided files.

  According to the present invention, it is possible to draw out the recording performance of the hard disk device as much as possible from a plurality of stream data that arrive at the same time, thereby preventing the occurrence of stream data loss.

  At the same time, even when a hard disk device that records a plurality of stream data is connected to another computer device, a file operation for each stream data can be executed.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 shows a configuration of an information recording apparatus 100 in the present embodiment.

  The CPU 102 controls the entire apparatus by loading a control program stored in the FLASH ROM 103 onto the SDRAM 104 and executing it. The PCI bridge 106 serves to connect the host bus 105 and the PCI bus 107. The CPU 102 exchanges data with other devices on the network 110 via the network controller 108. Further, the CPU 102 reads / writes the hard disk 101 via the disk controller 109.

  Here, the hard disk 101 is detachably connected by a serial ATA interface. Therefore, it is possible to connect the hard disk 101 to another device and refer to the contents data.

  On the network 110, a camera 111, a camera 112, and a camera 113, which are video sources, are connected. Stream data from these cameras is received by the information recording apparatus 100 via the network 110 and recorded on the hard disk 101.

FIG. 2 shows the structure of stream data from the cameras 111 to 113.
Stream data from each camera is packetized and transmitted as shown in the figure. Each packet includes an IP header 201, a UDP header 202, a stream control header 203, and a payload 204 in which video information is stored. The stream control header 203 stores information such as a sequence number 205, a time stamp 206, and a video source identifier 207. A sequence number 205 indicates the order of packet transmission from each camera. The time stamp 206 indicates the time when the packet is transmitted in each camera. The video source identifier 207 is information for uniquely identifying the camera that transmitted the packet.

  Here, the size of the payload 204 is determined by performing negotiation between the cameras 111 to 113 and the information recording apparatus 100 before recording the stream data. The information processing apparatus 100 notifies the cameras 111 to 113 of a desired size, and the cameras 111 to 113 generate the payload 204 according to the size notified from the information processing apparatus 100.

  As a result, when the stream control header 203 and the payload 204 of the packet data are recorded on the hard disk 101 on the information recording apparatus 100 side, the total of the data management unit in the hard disk 101 and the stream control header 203 and the payload 204 The size can also be matched, and the load of recording processing of the received stream data can be reduced.

  In addition, since the packet data is not fragmented on the network, it is necessary to prevent the size from exceeding an MTU (Max Transfer Unit) that is a packet size that can be transmitted at once on the network.

FIG. 3 shows a stream data recording method in the hard disk 101.
The file system of this embodiment is configured based on the FAT32 file system. As a result, when the hard disk device 101 is connected to another device that supports the FAT32 file system, it becomes possible to operate (for example, copy or delete) a file in the disk by the other device.

  The recording area on the disc includes a FAT32 file system compatible management area 301 and a data area 302 in which actual data such as stream data coming from the cameras 111 to 113 is recorded. The management area includes data such as a FAT for managing a cluster (a block composed of a plurality of continuous sectors) in which a file is recorded by a cluster chain, and a directory entry storing directory and file management information. The FAT, root directory entry, and the like are arranged at fixed locations based on the FAT32 file system. As a result, other devices that support the FAT32 file system can access each file by following the metadata based on the information arranged in these fixed locations.

  A data area 302 in FIG. 3 shows a state in which stream data coming from the cameras 111 to 113 is recorded. The packet data from the camera 111 is A1, A2, A3, the packet data from the camera 112 is B1, B2, the packet data from the camera 113 is C1, C2, and each packet data is 305 to 311 with respect to the time axis 304. It is assumed that the information recording apparatus 100 is reached in order.

At this time, each packet data is recorded in the continuous cluster area 303 on the data area 302 in the order of arrival. As a result, high-speed recording with reduced head seek time becomes possible. Here, by arranging the continuous cluster area 303 on the outer peripheral side of the disk, a higher recording performance can be obtained.

  When recording packet data, a stream control header 203 is recorded along with the payload 204. As a result, it is possible to determine from which camera the payload 204 is data and further what number of packet data is from that camera. In this embodiment, since the stream control header 203 and the payload 204 are both fixed lengths, the recorded data can be interpreted by following each length as an offset. Also, the metadata in the management area 301 is updated so that the recorded packet data is stored as a file based on the FAT32 file system. At this time, the metadata cached in the SDRAM 104 is updated during the recording of the packet data so as not to cause a delay due to the head seek between the recording in the management area 301 and the recording in the data area 302. The contents are appropriately reflected in the management area.

FIG. 4 shows a recording method of the data area 302 in this embodiment.
In the information recording apparatus according to the present embodiment, the cluster size is 4 KB (a size corresponding to 8 sectors), and 401, 402, and 403 in the figure represent clusters. In this example, one packet data is recorded for one cluster, and packet data 404, 405, and 406 are recorded in clusters 401, 402, and 403, respectively.

  In this way, by adopting a method of recording one packet data per cluster, the processing burden in dividing a file recording a plurality of stream data from a plurality of cameras into a stream data file for each camera is reduced. It can be significantly reduced. That is, the FAT may be changed so that the stream data for each camera becomes a cluster chain, and a directory entry describing file information for each camera may be newly provided.

FIG. 6 shows how the FAT and directory entry are changed when a file in which a plurality of stream data is recorded is divided into files for each stream data.
Reference numeral 601 denotes a FAT before file division, which is stream data from the camera 111 (A1, A2, A3, A4), stream data from the camera 112 (B1, B2), and stream data from the camera 113 (C1, C2). ) Is composed of one file, that is, one cluster chain. Reference numeral 602 denotes a FAT that has been changed by file division, and the stream data from the camera 111, the stream data from the camera 112, and the stream data from the camera 113 are composed of separate files, that is, separate cluster chains. .

  This change is executed by changing the cluster chain while referring to the video source identifier 207 of the packet data recorded in each cluster and determining from which camera each cluster is the stream data.

  Reference numeral 603 denotes a directory entry before file division, a directory entry 606 of a file in which a plurality of stream data is recorded, a directory entry 604 of a directory in which the file is recorded (hereinafter referred to as a current directory), and a directory entry of a parent directory. 605 etc. Reference numeral 604 denotes a directory entry that has been changed by file division. Instead of a directory entry 606 of a file that records a plurality of stream data, a directory entry 608 of a stream data file from the camera 111 and a directory of a stream data file from the camera 112 An entry 609 and a directory entry 610 of a stream data file from the camera 113 are recorded.

  In the information in the directory entries 608, 609, and 610, the information set in the directory entry 606 is set as the information indicating the creation date and time. In the information indicating the file size, information indicating the size of each divided file is set. In the information indicating the head cluster, information indicating the head cluster of each divided file is set. In the information indicating the file name, a separate file name generated based on a predetermined rule for identifying each file is set.

  In this way, by adopting a method of recording one packet data per cluster, when dividing a file in which a plurality of stream data is recorded into files for each stream data, only metadata such as FAT and directory entry can be changed. There is no need to transfer the recorded packet data to another position.

  On the other hand, when the difference between the cluster size and the packet data size is large, there is a problem that the recording capacity efficiency is deteriorated. In this case, the recording method of FIG. 5 is suitable.

  In the recording method of FIG. 5, the cluster is further equally divided into a plurality of sub-clusters (1 KB in this example), and one packet data is recorded per one sub-cluster. Packet data A1 (504), A2 (505), C1 (506), and B1 (507) are sequentially recorded in each sub-cluster of the cluster 501. Similarly, packet data A3 (508), C2 (509), B2 (510), and A4 (511) are included in each sub-cluster of cluster 502, and packet data C3 (512) and B3 are included in each sub-cluster of cluster 503. (513), A5 (514), and C4 (515) are recorded in order.

  In this way, when one file in which a plurality of stream data is recorded is divided into files for each stream data, it is necessary to move the packet data so that the plurality of stream data are not mixed in one cluster. That is, only packet data 519 to 522 from the camera 111 is recorded in the cluster 516, only packet data 523 to 536 from the camera 112 is recorded in the cluster 517, and only packet data 527 to 530 from the camera 113 is recorded in the cluster 518. As described above, a packet data movement process is executed. Since the packet data movement process may be performed in units of sub-clusters obtained by equally dividing the cluster, it can be realized with a simple algorithm.

  If the FAT and directory entry changing process as described with reference to FIG. 6 is executed after the moving process is completed, one file in which a plurality of stream data is recorded can be divided into files for each stream data.

  As described above, a method of recording one packet data in one cluster shown in FIG. 4 (hereinafter referred to as a first recording method), or a sub-cluster obtained by equally dividing one cluster shown in FIG. By recording one packet data (hereinafter referred to as a second recording method), the stream data from a plurality of cameras is recorded as one file on the hard disk device, thereby reducing the head seek time and achieving high performance. Recording can be performed, and it is possible to prevent the stream data from being lost. A file in which a plurality of stream data is recorded can be divided into files for each stream data with a simple algorithm. Since these files conform to a general-purpose file system such as the FAT32 file system, even when a hard disk device taken out from the information recording device is connected to a computer, these files can be operated.

FIG. 7 shows a flowchart of the stream recording operation of the information recording apparatus 100 in the present embodiment.
In this embodiment, one writing process records a plurality of received stream data on the hard disk device 101 as one file. This operation is started in response to, for example, an operation indicating a recording start instruction performed by the user or a recording start instruction received from the camera 111 operating as a master camera.

  In step S701, the information recording apparatus 100 negotiates with the cameras 111 to 113, and selects either the first recording method or the second recording method. The information recording apparatus 100 obtains the size of the packet data by negotiation with the cameras 111 to 113, and selects a suitable method based on the result of comparison with the cluster size of the own device. For example, when the size of the packet data is larger than half of the cluster size, the first recording method may be adopted, and otherwise, the second recording method may be selected. In addition, when the information recording apparatus 100 has a plurality of partitions having different cluster sizes or has the ability to dynamically change the cluster size, the cluster size is set so that the difference between the cluster size and the packet data size is as small as possible. After determining, the first recording method may be selected.

  In step S702, on the basis of the determined recording method, stream data from a plurality of cameras are recorded in a continuous cluster area in the ascending order of packet data, and stored as one file.

  In step S703, a file in which stream data from a plurality of cameras is recorded is divided into files for each stream data from each camera by changing metadata. Here, the file division processing in step S703 may be scheduled so as to be executed in a time zone in which the stream recording operation does not occur. In that case, the subsequent steps S705 and S706 are scheduled to be executed after step S703.

  Through the above processing, a file in which a plurality of stream data is recorded is divided into files for each stream data. As shown by 602 in FIG. 6, a continuous cluster area in a state where clusters are fragmented in each file. 303 is recorded.

  Therefore, in step S704, defragmentation processing is executed in the continuous cluster region 303.

FIG. 8 shows a state after the inside of the continuous cluster area 303 is defragmented.
In the example of 602 in FIG. 6, clusters (A1, A2) in which stream data from the camera 111 is recorded, clusters (B1, B2) in which stream data from the camera 112 is recorded, and clusters in which stream data from the camera 113 is recorded. (C1, C2) were arranged in a nested state. In the example of FIG. 8, clusters in which stream data from each camera are recorded are arranged in separate continuous areas. Thereby, the head seek time when reading each file is suppressed, and the performance of file read is improved.

  In step S705, in order to record the next multiple stream data, the file in the continuous cluster area 303 is moved to another free area to release the continuous cluster area 303 and make it unused.

FIG. 9 is a diagram showing a state in which a file in the continuous cluster area 303 is moved. 900 shows a state before the file is moved, and 905 shows a state after the file is moved.
Three files 901, 902, and 903 in the continuous cluster area 303 are moved to empty areas 904, 905, and 906, respectively, and the continuous cluster area 303 is changed to an unused state again.
Here, when the free area 901 is less than the capacity capable of accommodating the file 901, it is possible to secure the capacity capable of accommodating the file 901 by further moving the file recorded in the vicinity of the free area 901. Good. As described above, by leaving the continuous cluster area 303 in an unused state by moving a file, the area provided on the outer peripheral side of the disk and having high recording performance can be used for the recording of the next plurality of stream data.

  As described above, according to the present embodiment, stream data from a plurality of cameras are recorded on a continuous cluster area in approximately ascending order packet data and stored as one file. In addition, it is possible to perform recording with high performance while suppressing the head seek time, and it is possible to prevent the stream data from being lost.

  In addition, after recording stream data from multiple cameras, it is divided into files for each stream data from each camera, so file operations for each stream data can be performed even if the hard disk device is connected to another computer device. It becomes.

  Since one packet data is recorded for one cluster, only the metadata needs to be changed when the file is divided later, so that the load of the dividing process can be reduced.

  For each sub-cluster obtained by equally dividing one cluster, packet data including information identifying the transmission source is recorded one by one, so metadata for identifying the transmission source of each packet data in the cluster is created. Even without this, it becomes possible to divide the data into files for each stream data later.

  Further, according to the present embodiment, the file defragmentation process is executed together with the file division process, so that the head seek time when reading each file is suppressed, and the file read performance is improved. Thereby, the necessity of executing the defragmentation process of the entire hard disk device 101 is reduced, and the availability of the information processing apparatus 100 is improved. Further, after the file is divided, the file in the continuous cluster area 303 is automatically moved to another free area, so that the next multiple stream data is recorded in the area provided on the outer circumference side of the disk and having high recording performance. Can be used.

  The embodiment of the present invention can be realized by, for example, a computer executing a program. Also, means for supplying a program to a computer, for example, a computer-readable recording medium such as a CD-ROM recording such a program, or a transmission medium such as the Internet for transmitting such a program is also applied as an embodiment of the present invention. Can do. The above program can also be applied as an embodiment of the present invention. The above program, recording medium, transmission medium, and program product are included in the scope of the present invention.

It is the figure which showed the structure of the information recording device 100 in embodiment of this invention. It is the figure which showed the structure of the stream data from the cameras 111-113 in embodiment of this invention. It is the figure which showed the recording system of the stream data in the hard disk 101 in embodiment of this invention. It is the figure which showed the recording system of the data area 302 in embodiment of this invention. It is the figure which showed the recording system of the data area 302 in embodiment of this invention. It is a figure explaining the file division | segmentation system in embodiment of this invention. It is a flowchart of the stream recording operation of the information recording apparatus 100 in the embodiment of the present invention. It is the figure which showed the mode after the inside of the continuous cluster area | region 303 was defragmented. FIG. 5 is a diagram showing a state of moving a file in a continuous cluster area 303. It is a figure for demonstrating the recording system of the stream data in the conventional information recording device.

Explanation of symbols

100: Information recording device 101: Hard disk device 102: CPU
103: FLASH ROM
104: SDRAM
105: Host bus 106: PCI bridge 107: PCI bus 108: Network controller 109: Disk controller 110: Network 111: Camera 112: Camera 113: Camera

Claims (8)

File recording means for recording a plurality of stream data coming simultaneously as one file in a continuous area on the hard disk;
File dividing means for dividing the file into a plurality of files for each stream data;
Along with the file division by the file division means, a file relocation means for canceling the fragmented state of the divided files,
An information recording apparatus comprising:   The information recording apparatus according to claim 1, wherein the file rearrangement unit includes a unit that moves a file in the continuous area to the outside of the continuous area. File recording means for recording stream packet sequences from a plurality of video sources as a single file in a continuous area on the hard disk;
File dividing means for dividing the file into a plurality of files for each stream data;
Have
The file recording means records one stream packet for one unit area in a state including identification information for specifying a video source of the stream packet,
The information recording apparatus, wherein the file dividing unit operates the unit area based on the recorded identification information. A recording step of recording a plurality of stream data coming simultaneously as one file in a continuous area on the hard disk;
A file dividing step of dividing the file into a plurality of files for each stream data;
Along with the file division by the file division step, a rearrangement step for eliminating the fragmented state of the divided files,
An information recording method characterized by comprising:   5. The information recording method according to claim 4, wherein the rearrangement step includes a step of moving a file in the continuous area to the outside of the continuous area. A file recording step of recording stream packet sequences from a plurality of video sources as a single file in a continuous area on the hard disk;
A file dividing step of dividing the file into a plurality of files for each stream data;
Have
The file recording step records one stream packet for one unit area in a state including identification information for specifying a video source of the stream packet,
In the information recording method, the file dividing step performs an operation on the unit area based on the recorded identification information. On the computer,
A file recording procedure for recording stream packet sequences from a plurality of video sources as a single file in a continuous area on the hard disk;
A file dividing procedure for dividing the file into a plurality of files for each stream data;
A program for executing
The file recording procedure records one stream packet for one unit area in a state including identification information for specifying a video source of the stream packet,
The file division procedure performs the operation of the unit area based on the recorded identification information. On the computer,
A file recording procedure for recording stream packet sequences from a plurality of video sources as a single file in a continuous area on the hard disk;
A file dividing procedure for dividing the file into a plurality of files for each stream data;
A computer-readable storage medium storing a program for executing
The file recording procedure records one stream packet for one unit area in a state including identification information for specifying a video source of the stream packet,
The computer-readable storage medium characterized in that the file division procedure operates the unit area based on the recorded identification information.

Images