A network storage device is like a terminal with only a storage function, which works independently and has a fixed system inside, but can set some parameter functions by itself, and can be connected to a server or computer for settings. The network storage device is actually a simplified, The miniaturized server is also composed of motherboard, CPU, SoC chip level system, memory, network chip, etc. Although it is simplified from the basis of the server, it still has powerful functions. It can not only be used as HTTP server, FTP server and file server, but also has perfect data backup, network printer sharing, user/user group management functions. Can automatically download PT, BT.
1. Introduction to network storage devices
At present, the professional network storage technologies used by high-end servers are roughly divided into four types, including DAS, NAS, SAN, and iscsl. They can use RAID arrays to provide efficient and secure storage space.
2. Direct Attached Storage (DAS) for network storage devices
Direct-attached storage refers to the use of a storage device directly connected to a server through a SCSI interface. Some people also call it a disk array. [1] DAS has low acquisition cost and simple configuration. The use process is not much different from using a local hard disk. The only requirement for the server is an external SCSI port, so it is very attractive to small businesses. . But DAS also has many problems:
2.1 The server itself is easy to become a system bottleneck;
2.2 The server fails and the data is inaccessible;
2.3 For a system with multiple servers, the devices are scattered and management is inconvenient. When multiple servers use DAS at the same time, the storage space cannot be dynamically allocated among the servers, which may cause considerable waste of resources;
2.4 The data backup operation is complicated, only some brands of DAS servers are relatively simple to use, and DAS is widely used for file backup.
3. Network Attached Storage (NAS) for Network Storage Devices
A NAS is actually a storage device with a thin server. This thin server is actually a network file server. The NAS device is directly connected to the TCP/IP network, and the network server accesses the management data through the TCP/IP network.
As a thin server system, NAS is easy to install and deploy, and it is easy to manage and use. At the same time, because the client can directly access data in the NAS without going through the server, the system overhead can be reduced for the server.
NAS provides a solution for heterogeneous platforms using a unified storage system. Since NAS only needs to add a thin server system outside a basic disk array cabinet, the hardware requirements are very low, and the software cost is not high. Even free LINUX solutions can be used, and the cost is only slightly higher than that of directly attached storage. Branded NAS network storage devices has its own operating system, such as TerraMaster's TOS. The main problems with NAS are:
3.1 Since the stored data is transmitted over a common data network, it is susceptible to other traffic on the network. When there is other large data traffic on the network, it will seriously affect the system performance;
3.2 Since the stored data is transmitted through the ordinary data network, it is prone to security problems such as data leakage;
3.3 Storage can only be accessed by file, and cannot directly access physical data blocks like ordinary file systems, so it will seriously affect system efficiency in some cases, such as large databases cannot use NAS.
4. Storage Area Network (SAN) for network storage devices
A SAN is actually a dedicated network that is established for storage and is independent of the TCP/IP network. At present, the general SAN provides a transmission rate of 2Gb/S to 4Gb/S. At the same time, the SAN network exists independently of the data network, so the access speed is very fast. In addition, the SAN generally adopts a high-end RAID array, which makes the performance of the SAN in several professional It is the leader in network storage devices technology. Because the SAN is based on a dedicated network, it is highly scalable. It is very convenient to add a certain storage space to a SAN system or to add several servers that use storage space. Through the tape drive of the SAN interface, the SAN system can conveniently and efficiently realize centralized data backup.
As an emerging storage method, SAN is the future development direction of storage technology. However, it also has some disadvantages:
4.1 Expensive. Whether it is a SAN array cabinet or a Fibre Channel switch necessary for a SAN, the price is very expensive, and even the price of the Fibre Channel card used in the server is not easily accepted by small commercial enterprises.
4.2 It is necessary to establish an optical fiber network separately, and it is difficult to expand in different places.
5. iSCSI for network storage devices
The cost of using a special storage area network is very high, and using a common data network to transmit SCSI data to achieve similar functions to SAN can greatly reduce the cost and improve the flexibility of the system. iSCSI is such a technology, which utilizes the ordinary TCP/IP network to transmit the SCSI data blocks originally transmitted by the storage area network. The cost of iSCSI is much lower than that of SAN.
With the popularization of gigabit network, 10G network has gradually entered the mainstream, so that the speed of iSCSI does not have much disadvantage compared to SAN. The main problems with iSCSI currently are:
5.1 For emerging technologies, there are fewer manufacturers providing complete solutions, and the technical requirements for managers are high;
5.2 When accessing iSCSI data through a common network card, decoding into SCSI requires CPU operation, which increases the system performance overhead. If a special iSCSI network card is used, although the system performance overhead can be reduced, it will greatly increase the cost;
5.3 Use the data network for access, and the redundancy of the access speed is affected by the network operation.
6. The main difference between the four network storage technologies of network storage devices
Through the above comparative research, the four network storage technology solutions have their own advantages and disadvantages. For small and service-intensive commercial enterprises, a simple DAS solution can be used. For small and medium-sized commercial enterprises, the number of servers is relatively small, there are certain requirements for centralized data management, and there is no need for large-scale databases, the NAS solution can be used.
For medium and large commercial enterprises, SAN and iSCSI are better choices. If the servers that want to use storage are relatively centralized and have extremely high requirements on system performance, the SAN solution can be considered; for servers that want to use storage are relatively scattered and have low performance requirements, the iSCSI solution can be considered.
