January 13, 2021
1. Front-end storage
The so-called front-end storage is a built-in storage component in the front-end equipment of the network video surveillance system (such as a network video encoder or a network camera), and the front-end equipment directly completes the local recording and storage of surveillance images.
Front-end storage has several advantages: one is that it can reduce the capacity pressure caused by centralized storage through distributed storage deployment; the other is that it can effectively alleviate the network traffic pressure caused by centralized storage; the third is that it can avoid centralized storage in The image is lost when the network fails.
For front-end storage, a single front-end encoding device usually has a small number of monitoring points and a short storage time, so the storage capacity is not high. Network cameras generally use CF cards or SD cards, and video servers generally use built-in hard drives. Compared with the previous stand-alone storage, this is basically the same.
The essential difference from the previous stand-alone storage is that in order to ensure the flexibility and convenience of user access, all front-end storage in the network video surveillance system must not only provide point-to-point stand-alone access, but also be able to provide it through a unified interface. Centralized sharing of all content. To this end, the network video surveillance system uniformly manages and schedules all front-end storage through the central business platform, and realizes the networking of storage space and storage content. In this way, users can either directly log in to a single front-end device for on-demand playback of video materials, or log in to the central business platform for centralized retrieval and playback of all front-end video materials.
2. Central storage
In the network video surveillance system, more central storage is deployed. The front-end equipment collects the monitoring point image and encodes and compresses it into a digital monitoring code stream, and then transmits it to the central service platform through the network, and the central service platform distributes the code stream to the network video recording unit for centralized storage.
In many large-scale video surveillance networking applications, a multi-level distributed central storage method, that is, sub-central storage, can also be used. On the one hand, it can reduce the pressure on storage capacity and network traffic caused by centralized storage at a central point. Significantly improve the reliability of the system.
The use of central/sub-central storage has obvious advantages in the following aspects: first, it is more convenient for users to retrieve and call video resources; second, the integrity of the stored content is easier to guarantee, and it will not be due to a front-end device Theft or damage leads to the loss of important content; third, resource scheduling can be carried out reasonably, and storage space can be allocated for front-end equipment on demand, thereby saving resources; fourth, it is conducive to formulating diversified storage strategies to meet the individual needs of users ; Fifth, it is convenient to maintain, convenient for centralized detection and timely troubleshooting.
For applications where the number of monitoring points is relatively small and the storage time is not long, the central/sub-center storage can be deployed in a relatively simple way such as the server plugging in the hard disk or the external disk cabinet, called DAS (Direct Access Storage), and Stand-alone is similar. With the widespread recognition of the advantages of network video surveillance, more and more large and even super-large video surveillance systems are beginning to emerge, such as the social security surveillance system under the construction of "Safe City", China Telecom and China Netcom are fully promoting The two major operational-level video surveillance systems, "Global Eye" and "Wide Vision", are facing the large-scale access of front-end equipment and the demand for large-capacity centralized storage. The previous stand-alone storage methods cannot meet the application requirements of these systems in terms of flexible expansion of capacity. More advanced network storage equipment and storage technologies must be adopted. Typical of these are SAN, NAS and iSCSI.
Unlike SAN based on a specialized Fibre Channel protocol, NAS (Network Access Storage) is based on an IP network to realize the interconnection of servers and storage arrays, using TCP/IP protocol for communication, and file-level I/O for data transmission. In contrast, NAS equipment is easier to install, debug, use, and manage, and the deployment cost is relatively low.
iSCSI is a new standard protocol of IETF, which is a transmission standard that converts SCSI block data into network packets through IP network. It transmits data through IP network like NAS, but in terms of data access method , It uses the BlockProtocol protocol which is different from NAS but the same as SAN. Therefore, the value that iSCSI brings to users lies in: First, iSCSI makes it possible to transmit SCSI data packets over Ethernet, which makes SAN get rid of expensive optical fiber networks. The original functions can be realized through IP networks, which reduces Management complexity reduces costs. Second, due to the complexity of user application requirements, two storage networks, SAN and NAS, are often deployed at the same time, and iSCSI can integrate the two.
These features of iSCSI are in line with the current status and direction of the development of video surveillance, especially in the field of operation-level video surveillance. The storage scale is large and the investment is high. Based on the current mature IP network, the central/sub-center storage system is constructed. iSCSI Technology is undoubtedly a good reference.
3. Hybrid storage
For video surveillance networks that are more complex and require very high storage security and reliability, the deployment method of both centralized storage and front-end storage can be used, which has the advantages of both and avoids possible risks. A better choice. But it will bring management complexity and high construction cost, which needs to be determined according to the specific situation.