|
For an enterprise, the security of network data is extremely important, and once important data is destroyed or lost, it will have a significant impact on the enterprise, and even irreparable losses. People are familiar with the term data storage backup, but they don't fully understand what backup really means. In the minds of ordinary people, backup and copy are often equated, and backup is simply regarded as a completely stylized and monotonous operation process such as replacing tapes and numbering tapes. In fact, this is not the case, because in addition to copying, it also includes more important content, namely management. Backup management includes the planability of backups, the automated operation of tape drives, the preservation of history, and logging. In fact, backup management is a comprehensive concept, which not only includes the formulation of systems and the management of tapes, but also can decide the introduction of backup technology, such as the selection of backup technology, the selection of backup equipment, the selection of media and even the selection of software technology. Some people often compare the content of hardware backups such as dual-server hot backups, disk array backups, and disk mirror backups with data storage backups. In fact, all hardware backups cannot replace data storage backups, hardware backups are just sacrificing one system, one device, etc. in exchange for the security of another system or device for a short period of time. If human error, natural disasters, power failures, virus attacks, etc., the consequences will be unimaginable, such as causing all systems to be paralyzed, all equipment to be unable to operate, and the resulting data loss cannot be recovered. It turns out that only data storage backups can provide people with foolproof data security protection. With the expansion of network applications and the deepening of computer applications, the explosive growth of enterprise data storage has led to the increasing complexity of enterprise users' storage systems, and the cost of storage management is also rising significantly. Storage concepts such as data backup, disaster recovery, and storage resource management are being accepted by users, and network storage is gradually becoming more widely used. At present, the storage products on the market mainly include disk arrays, tape drives and tape libraries, optical disc libraries, etc., among which tape equipment occupies an important position in the storage market with its mature technology, low price, complete product line, easy to use and other advantages. Three ways to store enterprise data In the early stage of enterprise establishment, the scale of users' data is not large, and the storage requirements are very simple, just store relevant data in a certain place. The ultimate purpose of storing data is not only to be able to store it safely, but also to ensure that the data can be called at any time. We first adopted a storage method called "DAS (Direct Attached Storage)". The server structure of this storage solution is like the PC architecture, and the external data storage devices (such as disk arrays, optical disk drives, tape drives, etc.) are directly connected to the server's internal bus, and the data storage equipment is part of the entire server structure, and the server is also responsible for the data storage of the entire network. The direct connection method of DAS can solve the storage space expansion and high-performance transmission needs of a single server, and the capacity of a single external storage system has developed from less than 1TB to 2TB. In addition, DAS can also form a dual-machine high-availability system based on disk arrays to meet the requirements of high availability for data storage. From a trend perspective, DAS will continue to be used as a storage mode. NAS (Network Attached Storage) is a comprehensive improvement on the previously inefficient DAS storage method, which is a file server developed separately for network data storage independent of PC servers. The NAS server is centrally connected to all network data storage devices (such as various disk arrays, tapes, optical disk drives, etc.), and the storage capacity can be well expanded. It provides a simple, cost-effective, highly available, highly scalable, and low total cost of ownership (TCO) network storage solution. SAN (Storage Area Network) is completely different from NAS, it does not centralize all storage devices in a special NAS server, but connects these storage devices individually through fiber optic switches to form a Fibre Channel network, and then this network is connected to the existing LAN of the enterprise Channel (FC, Fibre Channel) protocol, which is a standard integration established by ANSI for network and channel I/O interfaces, supporting a variety of advanced protocols such as HIPPI, IPI, SCSI, IP, ATM, etc. In a SAN, data is stored in a centralized manner, enhancing data manageability while adapting to data sharing in the same storage pool under multiple operating systems, reducing total cost of ownership. Choose a reasonable data storage scheme The architecture of enterprise storage applications mainly includes three modes: DAS, NAS and SAN, and the three modes have obvious differences from the logic of the architecture. An enterprise storage has the following requirements: performance, security, scalability, ease of use, total cost of ownership, services, and more. Since the construction of storage systems for enterprise users is not an overnight thing, and will go through the process of moving from stand-alone to networked storage, there are three storage solutions: DAS, NAS, and SAN for enterprise users to choose from. 1. Select the DAS scheme DAS direct storage relies on the server host operating system for data I/O read/write and storage maintenance management, data backup and recovery require occupying server host resources (including CPU, system I/O, etc.), data flow needs to be returned to the host and then to the tape drive (library) connected to the server, data backup usually occupies 20%~30% of the server host resources, so many enterprise users often perform daily data backup late at night or when the business system is not busy, so as not to affect the operation of the normal business system. The larger the amount of data stored in direct storage, the longer it takes to back up and restore, and the greater the dependency and impact on server hardware. This scheme was mainly used on early computers and servers, and because the demand for data storage was not large at that time, the storage capacity required by a single server could meet the daily data storage needs, so it was quite common in low-end network applications. 2. Select a NAS solution NAS devices generally integrate local backup software, which can back up important data in NAS devices locally without going through the server, and NAS devices provide hard disk RAID, redundant power supplies and fans, and redundant controllers, which can meet the stable applications of NAS. NAS devices are mainly used to realize file sharing applications under different operating system platforms, compared with traditional servers or DAS storage devices, NAS equipment is very simple to install, debug, use, and manage, and the use of NAS can save a certain amount of device management and maintenance costs. The NAS device provides an RJ-45 interface and a separate IP address, which can be directly connected to the switch of the backbone network or other LAN hubs, and the NAS device can be plug-and-play on the network through simple settings (such as setting the IP address of the machine, etc.), and there is no need to pause when expanding network data online, so as to ensure smooth data storage. NAS data storage solutions are designed based on local area networks, communicating according to the traditional TCP/IP protocol, oriented to messaging, and transmitting data in the form of file I/O. In a LAN environment, NAS can fully realize data-level sharing between heterogeneous platforms, such as Windows NT, Linux, Unix, and other platforms. For this reason, the use and maintenance cost of NAS storage solutions for enterprises is quite low, and the existing network administrators can take care of it. 3. Select a SAN plan Since SAN connects storage devices and servers through a separate SAN network, usually based on Fibre Channel, when there is a need for access to a large amount of data, the data can be transmitted at high speed between the relevant server and the storage device in the background through the SAN network, and the bandwidth occupation of the LAN is almost zero, and the server can access any storage device on the SAN, which improves the availability of data. In cases with high performance and reliability requirements, the use of advanced SAN data storage network can make data storage, backup and other activities independent of the original LAN network, thereby reducing the load on the LAN and ensuring the smooth progress of the original network application. At the same time, the SAN network adopts optical fiber transmission channels, which can obtain high-speed data transmission rates. The SAN solution simplifies management and centralized control, which is very practical for all storage devices to be centralized in the information center. SAN separates the storage and server platforms of the enterprise for 24 x 7 uninterrupted system availability and centralized management, on top of which a unified disaster recovery solution can be applied while cost-effectively scaling storage environments. Therefore, SAN is very suitable for non-linear editing, server clustering, remote disaster recovery, Internet data services, and other fields. The development trend of enterprise data storage The biggest drawback of DAS directly connected storage is that the storage resources of each part are isolated from each other, which can easily form islands. NAS has the characteristics of unattended, highly functional, stable performance, and dedicated functions, but NAS technology cannot meet the requirements of data storage systems with 99.999% reliability. The biggest advantage of SAN is that it can integrate storage devices from different manufacturers in heterogeneous environments to achieve resource sharing, so SAN will become the mainstream of enterprise storage applications. On the one hand, in order to ensure the smooth flow of data under different hardware platforms and different operating systems, the openness of storage architecture, the compatibility of storage hardware, and the adaptability of storage software come together, which is manifested as the integration of storage, and the integration of SAN and NAS is the general trend; On the other hand, virtual storage separates the storage seen by the host operating system from the actual physical storage, and virtual storage technology can improve the utilization rate of storage devices. At present, virtual technology has attracted the attention of almost all storage system manufacturers, and devices using virtual storage technology will become the new mainstream of the market.
|