Safe city storage needs
The safe city video surveillance system generally has many monitoring points, a large number of cameras, and a long storage time (more than one month of monitoring data storage time). Therefore, video surveillance system storage must support large capacity, and the capacity has high scalability to meet the needs of long-term large-capacity video image storage. The storage system assembles platforms at video and image centers or video image information front ends such as municipal bureaus, county bureaus, police stations, and video image information storage devices to video-tape their converged video image information resources. Equipped with video image information storage device, video recording of its own video image information resources.
Safe city's image information storage technology requirements:
· The video playback resolution of each channel of video image information should not be lower than the D1 SD and 720P HD formats;
· Use standard MPEG-4, H.264 format for encoding, and can be decoded by VLC;
· Ensure that video can be recorded at least for more than 30 days at D1/720P video resolution;
· Fast search for video, with complete search conditions. Supports multiple users to play back the same video at the same time, and has a certain ability to extend the intelligent search based on objects;
· Has a user-friendly, interactive interface and can play clips. Has the ability to download video backups according to files or time periods;
Storage time synchronization, digital image storage devices can support NTP;
Set the time error should be within 10 seconds;
· Automatically clean up disk policies based on conditions such as time, capacity, etc.
Provides secondary development software SDK packages and API interface functions;
• When the primary monitoring center is “downtime,†the secondary monitoring center can have seven days of monitoring video storage. When the primary monitoring center storage is restored to normal, the storage is automatically transferred to the primary monitoring center for storage, and the local video data is copied to the upper level monitoring center for storage in the background to ensure the integrity of the video.
· The centralized storage of the primary monitoring center is used to store important event video databases, which need to be stored for a long period of time;
· Based on the performance of IP networks, and based on * business requirements to achieve effective video management applications, to build a stable, reliable, simple architecture, easy to maintain and upgrade the digital video storage system.
Video data storage technology
According to the safe city monitoring network construction framework, the following methods can be implemented in centralized storage of system video data.
DAS
DAS is directly connected to the storage, using SCSI and FC technology, external storage devices connected through a network cable or fiber optic link directly to a workstation, data storage is part of the overall server structure.
NAS
NAS is a network attached storage, which is a professional network file storage server, or network directly connected storage device, using NFS or CIFS protocol for file access via TCP/IP.
SAN
The SAN is a storage area network and a dedicated storage network centered on data storage. The network structure is scalable and can achieve block-level I/O data access between storage devices and application servers. According to the protocol and media used, the SAN is divided into three types: FC-SAN, IP-SAN, and IB-SAN.
IP-SAN/NAS
The iSCSI protocol used by the IP-SAN/NAS in the SAN structure serves as a new Internet SCSI standard for accessing data blocks at the IP network, and establishes a bridge between the SAN and the NAS. iSCSI was originally initiated by both Cisco and IBM and then submitted to the Internet Engineering Task Force (IETF) for standardization. It is a technology that combines NAS and SAN. It is a storage server that can send and receive Block-level data over a TCP/IP network.
Choice of storage considerations
In the Safe City project, when choosing a storage solution, consider:
· Adopt H.264 or MPEG-4 video format, but various encoders have their own playback library;
Digital video quantity affects storage capacity and storage performance requirements;
• Video bandwidth/recording time: Affects storage capacity and performance requirements. The general video frequency range is from 1Mbps to 8Mbps, but it is usually in the range of 2Mbps-4Mpbs to meet SD and HD signals.
Video recording should be able to maintain 30 days of storage;
• Recording methods of video streams: Interfaces that affect storage space. Currently knowing the information, all DVS/NVS/encoders use the file system to record video (video) by producing video files at regular intervals;
· Stable reliability requirements: Some video surveillance systems have high requirements for continuity of monitoring services, and require that equipment failures do not affect the current monitoring and recording;
* With the development of * business, video storage should be able to expand.
Using IP-SAN as the Best Network Storage Application Technology in Pingan City
IP-SAN is a network area storage architecture based on IP Ethernet. It uses the iSCSI protocol instead of the Fibre Channel (FC) protocol to transmit data. The IP-SAN architecture eliminates the need to use expensive fiber optic networks, FC-HBA cards, and Fibre Channel. Instead of storage devices, use IP Ethernet, Ethernet, and iSCSI storage devices. Therefore, the cost of the IP-SAN storage architecture is much lower than that of the FC-SAN. Similar to FC-SAN, IP-SAN can also divide storage devices into one or more file volumes and export them to front-end application clients. Client computers can create new file system operations on these imported volumes. The client computer accesses these volumes as device-level block access. The block-level access characteristics determine the high I/O performance and low latency of iSCSI data access.
This kind of mass storage IP-SAN applied to the safe city has the following advantages of the IP network:
· Flexible storage network expansion, adaptive application changes;
The proven transmission equipment guarantees the reliability of operation;
Ethernet transitions from 1G to higher speeds, requiring only a simple upgrade to get a significant performance boost;
IP long-distance expansion capabilities, easy remote data replication and disaster recovery;
· The use of a large number of familiar network technology and management talent will reduce system training and labor costs;
The biggest contribution of IP-based storage is to integrate storage networks into mainstream data communications.
Massive video storage solution design
Due to the large number of video accesses in the Ping An City project, most of the storage systems are designed with a distributed storage and centralized storage network architecture. They are collected by video front-end platforms such as municipal bureaus, county bureaus, police stations, and video information front-end platforms. Equipped with video information storage devices, video recording of the video information resources collected. The architecture of these two storage methods is roughly as follows:
Distributed network storage: In the police station and county monitoring center access the front-end video surveillance image, deploy network storage devices, the front-end video information including high-definition network camera directly stored in the network storage device;
· Centralized network storage: IP network cameras (including network dome cameras and high-definition network cameras) are distributed in various police stations and video data are stored centrally in the network storage devices of the county bureau's monitoring center. The police station calls the video images within the jurisdiction through the network to realize the real-time monitoring and management of the monitoring system.
Storage selection
For urban video surveillance with a large number of accesses, in order to reduce the pressure of video transmission from the police station to the district/county branch and ensure the application of front-line combat units at the police station, the system adopts a distributed design to support the coexistence of multiple storage devices, and the county branch office and the superior city office The command center uses a centralized storage device to back up key monitoring images. The storage method can generally be planned as follows:
Send a command center to store video in real time for 30 days in D1 format and allocate IP-SAN storage;
· For police stations with fewer monitoring points, video images can be stored directly to higher level county* sub-bureaus;
· The command center of the municipal (sub) bureau stores access to important video images and social video resources in its jurisdiction.
Storage management server management
Deploy storage management servers at the police stations and county monitoring centers. The network storage management service software with embedded platform software is a dedicated storage management software module for large-scale network monitoring system environments. Real-time recording, alarm recording, and centralized centralized backup of all remote images. And afterwards retrieval search service, support video recording under the plan video. Support multiple storage methods such as DAS, NAS, and SAN. IP-SAN storage is recommended for video storage. The network storage management server can be deployed in a distributed manner according to requirements. Currently, it can be deployed in police stations, county branch offices, and city offices. The number of video channels that can be managed by a storage management server is not limited, but the number of concurrently processed storage video channels is limited by server performance. The current test result is that the next storage server in the Giga Gigabit network environment can concurrently transmit the corresponding 1000-channel CIF video stream.
The network storage server software obtains the information of all video sources in the system through the network and stores it in a large-capacity storage device. The system can support a distributed cluster storage structure consisting of n calculators. At the bottom of the communication of these computers is our high-performance communication platform technology. In normal operation, each storage computer is responsible for storing certain video information in the entire system. First we need to define the storage groups. Each storage group has n channels of images. We bind a storage group to a storage server so that the entire system's image will be stored on different storage servers. The model used is an integrated approach to front-end distributed video and back-end centralized video. The objects managed by the storage management server can be PC server storage devices, DVR devices, IP network cameras, disk arrays, or tape libraries.
Storage Fault Tolerant Technology Design
In the Safe City project, the storage system should be able to provide a multi-level fault-tolerance mechanism to ensure that video storage operations can be performed stably. First, fault-tolerance technology on the system architecture. Our system supports distributed clustering technology. In simple terms, each server normally performs the current storage task, but at the same time it acts as a backup cluster node for other servers.
When a node of the cluster system fails, our underlying communication protocol can accurately monitor the problem of this machine. This is called the member departure event. The member departure event will be delivered by the trusted protocol to each node in the cluster. , Its backup server will immediately assume the responsibility of the failed node, video recording. This is fault tolerance. When the failed node is repaired and restarted, the underlying communication protocol can deliver this rejoin time to each node. This is called a member join event. The backup server stops recording after receiving this time. This is the error recovery function.
The fault tolerance on the second level is that the entire cluster is disconnected from the image source and real-time video recording is not possible. At this time, we will use the plan and actual video implementation to make up. Please refer to the plans and supplements below.
The third level of fault tolerance is when video recording is completed. Each video file ends and a new video file starts. The system is redundant with some data to ensure that no image frames are lost.
Storage video plan and make up plan
Each video group in the safe city storage system can have its own video plan. The bottom layer of the video execution is our plan and task scheduling engine. In each video recording process, the system records the running time of the video recording, and maintains the server of the system. Based on the difference between the scheduled schedule and the actual implementation, we can calculate the time period for which the recording needs to be made. The patching system will automatically download the missing video information from the appropriate DVR.
Video disk reorganization strategy
When the storage system discovers that the storage capacity exceeds the maximum warning line specified by the system, the liquidation task will start. This task is managed by the maintenance server. The maintenance server will periodically check the disk status, perform disk cleanup, and automatically export the disk. The system provides a centralized disk reorganization strategy that includes FIFO, recent access, maximum access, and retention priority. In the first-in first-out case, all the first recorded image data will be deleted. In the recent access strategy, the more recently accessed video data will be retained, the more access policies will consider the frequency of access to data by the entire system user, and the more people visit, the longer the video data will be retained. In all of these strategies, the retention priority can be defined. If the priority of a channel is higher, its data will be deleted later, and the system can define the length of the reservation.
Conclusion
The development of high-definition and mass storage to the network stage is no longer a matter of simple storage in the Safe City project, but how to achieve fast and accurate video access and access among the massive information is of concern to everyone. The storage management function is particularly important. Therefore, how to provide users with high-performance, stable, easy-to-expand, and management solutions is an urgent need for a safe city storage solution.
The safe city video surveillance system generally has many monitoring points, a large number of cameras, and a long storage time (more than one month of monitoring data storage time). Therefore, video surveillance system storage must support large capacity, and the capacity has high scalability to meet the needs of long-term large-capacity video image storage. The storage system assembles platforms at video and image centers or video image information front ends such as municipal bureaus, county bureaus, police stations, and video image information storage devices to video-tape their converged video image information resources. Equipped with video image information storage device, video recording of its own video image information resources.
Safe city's image information storage technology requirements:
· The video playback resolution of each channel of video image information should not be lower than the D1 SD and 720P HD formats;
· Use standard MPEG-4, H.264 format for encoding, and can be decoded by VLC;
· Ensure that video can be recorded at least for more than 30 days at D1/720P video resolution;
· Fast search for video, with complete search conditions. Supports multiple users to play back the same video at the same time, and has a certain ability to extend the intelligent search based on objects;
· Has a user-friendly, interactive interface and can play clips. Has the ability to download video backups according to files or time periods;
Storage time synchronization, digital image storage devices can support NTP;
Set the time error should be within 10 seconds;
· Automatically clean up disk policies based on conditions such as time, capacity, etc.
Provides secondary development software SDK packages and API interface functions;
• When the primary monitoring center is “downtime,†the secondary monitoring center can have seven days of monitoring video storage. When the primary monitoring center storage is restored to normal, the storage is automatically transferred to the primary monitoring center for storage, and the local video data is copied to the upper level monitoring center for storage in the background to ensure the integrity of the video.
· The centralized storage of the primary monitoring center is used to store important event video databases, which need to be stored for a long period of time;
· Based on the performance of IP networks, and based on * business requirements to achieve effective video management applications, to build a stable, reliable, simple architecture, easy to maintain and upgrade the digital video storage system.
Video data storage technology
According to the safe city monitoring network construction framework, the following methods can be implemented in centralized storage of system video data.
DAS
DAS is directly connected to the storage, using SCSI and FC technology, external storage devices connected through a network cable or fiber optic link directly to a workstation, data storage is part of the overall server structure.
NAS
NAS is a network attached storage, which is a professional network file storage server, or network directly connected storage device, using NFS or CIFS protocol for file access via TCP/IP.
SAN
The SAN is a storage area network and a dedicated storage network centered on data storage. The network structure is scalable and can achieve block-level I/O data access between storage devices and application servers. According to the protocol and media used, the SAN is divided into three types: FC-SAN, IP-SAN, and IB-SAN.
IP-SAN/NAS
The iSCSI protocol used by the IP-SAN/NAS in the SAN structure serves as a new Internet SCSI standard for accessing data blocks at the IP network, and establishes a bridge between the SAN and the NAS. iSCSI was originally initiated by both Cisco and IBM and then submitted to the Internet Engineering Task Force (IETF) for standardization. It is a technology that combines NAS and SAN. It is a storage server that can send and receive Block-level data over a TCP/IP network.
Choice of storage considerations
In the Safe City project, when choosing a storage solution, consider:
· Adopt H.264 or MPEG-4 video format, but various encoders have their own playback library;
Digital video quantity affects storage capacity and storage performance requirements;
• Video bandwidth/recording time: Affects storage capacity and performance requirements. The general video frequency range is from 1Mbps to 8Mbps, but it is usually in the range of 2Mbps-4Mpbs to meet SD and HD signals.
Video recording should be able to maintain 30 days of storage;
• Recording methods of video streams: Interfaces that affect storage space. Currently knowing the information, all DVS/NVS/encoders use the file system to record video (video) by producing video files at regular intervals;
· Stable reliability requirements: Some video surveillance systems have high requirements for continuity of monitoring services, and require that equipment failures do not affect the current monitoring and recording;
* With the development of * business, video storage should be able to expand.
Using IP-SAN as the Best Network Storage Application Technology in Pingan City
IP-SAN is a network area storage architecture based on IP Ethernet. It uses the iSCSI protocol instead of the Fibre Channel (FC) protocol to transmit data. The IP-SAN architecture eliminates the need to use expensive fiber optic networks, FC-HBA cards, and Fibre Channel. Instead of storage devices, use IP Ethernet, Ethernet, and iSCSI storage devices. Therefore, the cost of the IP-SAN storage architecture is much lower than that of the FC-SAN. Similar to FC-SAN, IP-SAN can also divide storage devices into one or more file volumes and export them to front-end application clients. Client computers can create new file system operations on these imported volumes. The client computer accesses these volumes as device-level block access. The block-level access characteristics determine the high I/O performance and low latency of iSCSI data access.
This kind of mass storage IP-SAN applied to the safe city has the following advantages of the IP network:
· Flexible storage network expansion, adaptive application changes;
The proven transmission equipment guarantees the reliability of operation;
Ethernet transitions from 1G to higher speeds, requiring only a simple upgrade to get a significant performance boost;
IP long-distance expansion capabilities, easy remote data replication and disaster recovery;
· The use of a large number of familiar network technology and management talent will reduce system training and labor costs;
The biggest contribution of IP-based storage is to integrate storage networks into mainstream data communications.
Massive video storage solution design
Due to the large number of video accesses in the Ping An City project, most of the storage systems are designed with a distributed storage and centralized storage network architecture. They are collected by video front-end platforms such as municipal bureaus, county bureaus, police stations, and video information front-end platforms. Equipped with video information storage devices, video recording of the video information resources collected. The architecture of these two storage methods is roughly as follows:
Distributed network storage: In the police station and county monitoring center access the front-end video surveillance image, deploy network storage devices, the front-end video information including high-definition network camera directly stored in the network storage device;
· Centralized network storage: IP network cameras (including network dome cameras and high-definition network cameras) are distributed in various police stations and video data are stored centrally in the network storage devices of the county bureau's monitoring center. The police station calls the video images within the jurisdiction through the network to realize the real-time monitoring and management of the monitoring system.
Storage selection
For urban video surveillance with a large number of accesses, in order to reduce the pressure of video transmission from the police station to the district/county branch and ensure the application of front-line combat units at the police station, the system adopts a distributed design to support the coexistence of multiple storage devices, and the county branch office and the superior city office The command center uses a centralized storage device to back up key monitoring images. The storage method can generally be planned as follows:
Send a command center to store video in real time for 30 days in D1 format and allocate IP-SAN storage;
· For police stations with fewer monitoring points, video images can be stored directly to higher level county* sub-bureaus;
· The command center of the municipal (sub) bureau stores access to important video images and social video resources in its jurisdiction.
Storage management server management
Deploy storage management servers at the police stations and county monitoring centers. The network storage management service software with embedded platform software is a dedicated storage management software module for large-scale network monitoring system environments. Real-time recording, alarm recording, and centralized centralized backup of all remote images. And afterwards retrieval search service, support video recording under the plan video. Support multiple storage methods such as DAS, NAS, and SAN. IP-SAN storage is recommended for video storage. The network storage management server can be deployed in a distributed manner according to requirements. Currently, it can be deployed in police stations, county branch offices, and city offices. The number of video channels that can be managed by a storage management server is not limited, but the number of concurrently processed storage video channels is limited by server performance. The current test result is that the next storage server in the Giga Gigabit network environment can concurrently transmit the corresponding 1000-channel CIF video stream.
The network storage server software obtains the information of all video sources in the system through the network and stores it in a large-capacity storage device. The system can support a distributed cluster storage structure consisting of n calculators. At the bottom of the communication of these computers is our high-performance communication platform technology. In normal operation, each storage computer is responsible for storing certain video information in the entire system. First we need to define the storage groups. Each storage group has n channels of images. We bind a storage group to a storage server so that the entire system's image will be stored on different storage servers. The model used is an integrated approach to front-end distributed video and back-end centralized video. The objects managed by the storage management server can be PC server storage devices, DVR devices, IP network cameras, disk arrays, or tape libraries.
Storage Fault Tolerant Technology Design
In the Safe City project, the storage system should be able to provide a multi-level fault-tolerance mechanism to ensure that video storage operations can be performed stably. First, fault-tolerance technology on the system architecture. Our system supports distributed clustering technology. In simple terms, each server normally performs the current storage task, but at the same time it acts as a backup cluster node for other servers.
When a node of the cluster system fails, our underlying communication protocol can accurately monitor the problem of this machine. This is called the member departure event. The member departure event will be delivered by the trusted protocol to each node in the cluster. , Its backup server will immediately assume the responsibility of the failed node, video recording. This is fault tolerance. When the failed node is repaired and restarted, the underlying communication protocol can deliver this rejoin time to each node. This is called a member join event. The backup server stops recording after receiving this time. This is the error recovery function.
The fault tolerance on the second level is that the entire cluster is disconnected from the image source and real-time video recording is not possible. At this time, we will use the plan and actual video implementation to make up. Please refer to the plans and supplements below.
The third level of fault tolerance is when video recording is completed. Each video file ends and a new video file starts. The system is redundant with some data to ensure that no image frames are lost.
Storage video plan and make up plan
Each video group in the safe city storage system can have its own video plan. The bottom layer of the video execution is our plan and task scheduling engine. In each video recording process, the system records the running time of the video recording, and maintains the server of the system. Based on the difference between the scheduled schedule and the actual implementation, we can calculate the time period for which the recording needs to be made. The patching system will automatically download the missing video information from the appropriate DVR.
Video disk reorganization strategy
When the storage system discovers that the storage capacity exceeds the maximum warning line specified by the system, the liquidation task will start. This task is managed by the maintenance server. The maintenance server will periodically check the disk status, perform disk cleanup, and automatically export the disk. The system provides a centralized disk reorganization strategy that includes FIFO, recent access, maximum access, and retention priority. In the first-in first-out case, all the first recorded image data will be deleted. In the recent access strategy, the more recently accessed video data will be retained, the more access policies will consider the frequency of access to data by the entire system user, and the more people visit, the longer the video data will be retained. In all of these strategies, the retention priority can be defined. If the priority of a channel is higher, its data will be deleted later, and the system can define the length of the reservation.
Conclusion
The development of high-definition and mass storage to the network stage is no longer a matter of simple storage in the Safe City project, but how to achieve fast and accurate video access and access among the massive information is of concern to everyone. The storage management function is particularly important. Therefore, how to provide users with high-performance, stable, easy-to-expand, and management solutions is an urgent need for a safe city storage solution.
WUXI HOLLY rubber sound insulation board is an acoustic material developed with unique and advanced technology.Compared with traditional materials, the closed cell design ensures the same amount of sound insulation while reducing the density of the material and the weight is lighter.
Sound insulation sponge foam products are widely used in studio, recording room, residence, hotel, night club, schools, ballroom, hospital,
office building, industrial workshop, electrical and mechanical equipment, aerospace, metallurgy, machinery, medical equipment,
environmental protection, automotive, marine, cooling tower, automobile and motorcycle engine, architectural decoration, household appliance, etc.
Sound Insulation Board,Soundproofing Panels For Walls,Sound Reduction Board,Outdoor Soundproofing Material
Wuxi Holly International Trading Co., Ltd , https://www.hollynbrs.com