With the increasing improvement of social informatization construction, enterprises have gradually realized the informatization and digitization of office operations. A large amount of important data recording the normal operation of enterprise business is stored in storage terminals. These important data records With the company's system files, management information, technical files and current financial accounts, the loss or damage of these data will lead to immeasurable devastating consequences for the enterprise. But for most enterprises, when it comes to data disaster recovery and backup, the most direct reflection is to increase the budget, because they need to purchase more hosts, storage devices and corresponding software. So, does building a disaster recovery backup system require the construction of high-end network equipment, hardware machines, and software systems? How to consider the necessity of enterprises building data disaster recovery and backup projects in a more scientific and effective way? In fact, we can scientifically measure an enterprise's disaster recovery needs from the following two indicators: RTO, Recovery Time Objective, refers to the time after a disaster occurs, from the moment the system crashes and the business is suspended to the time the system recovers to the point where it can support various departments. When the business resumes operation, the time period between these two points is called RTO. Generally speaking, the shorter the RTO time, the shorter the time required to return to a usable state. Although from a management perspective, a shorter RTO time is better, however, it also means more cost investment, which may require the purchase of faster storage devices or high-availability software. For companies in different industries, their RTO goals are generally different. Even in the same industry, each company will have different RTO goals due to different business development scales. As mentioned above, the shorter the RTO target, the greater the cost investment. Each enterprise has its unit time profit index at that stage of development. Once the index is determined, the loss caused by business suspension over time can be calculated. Inspur storage engineers suggest that when an enterprise plans to build a disaster recovery system, it must first find an RTO target that is more suitable for the enterprise itself. That is, under the definition of this target, the investment in disaster backup should not be greater than the corresponding business loss. . RPO, Recover Point Object, recovery point indicator refers to the point in time before the disaster that the disaster recovery system can restore data to before the disaster occurs. It is an indicator that measures how much production data an enterprise will lose after a disaster occurs. . Ideally, we want RTO=0. RPO=0 means that the disaster will have no impact on the company's production and will neither cause production to stop nor cause production data to be lost. Judging from the current level of computer technology, we can build this type of disaster recovery system for users. The most famous examples are the settlement systems of VISA and Master, because these two bank settlement organizations occupy an important position in global bank settlement business. status, their settlement system does not allow any pauses and data loss, even in extreme and devastating disaster situations such as "911" and major earthquakes. However, the investment to implement such a disaster recovery system is huge. It requires a combination of storage data replication technology, server operating system mirroring technology, cluster technology, database high availability design, application system high availability design, synchronous disaster recovery technology, asynchronous disaster recovery technology, and intra-city Detailed rules and regulations such as disaster recovery plan, off-site disaster recovery plan, corresponding management process and accident response process, as well as staffing and administrative support means (communications, transportation, etc.), are integrated to complete a complete disaster recovery plan (It is actually a dual production center or multiple production center solution, and there is no simple disaster recovery center). However, the investment in this solution is too huge. At present, except for enterprises with special characteristics such as China UnionPay, there may not be many enterprises in China that will implement this system. Therefore, what is a reasonable investment scale in the construction of disaster recovery systems for Chinese enterprises? A senior engineer from Inspur Storage suggested that if the business department can confirm the RTO/RPO indicators, then the technical department can determine the scale of investment by choosing appropriate disaster recovery technology and supporting management processes, and then use the most optimized construction method to achieve data recovery. Disaster recovery and protection purposes. For example, if the business department confirms that after a disaster, production can be resumed within 3 hours to meet user needs, and business system data cannot be lost, then RTO = 3 hours and RPO = 0, then data replication technology based on the storage platform must be selected. Synchronous disaster recovery plan; if the business department confirms that the business analysis system can be restored within 3 days after the disaster occurs, and the previous data loss is negligible, then RTO = 3 days and RPO is zero, then choose low-end ATA disk implementation Offsite backup can meet data protection requirements. It is worth mentioning that a huge amount of money is invested in building a disaster recovery center for a once-in-a-century disaster. The equipment in the disaster recovery center cannot bring benefits to the enterprise before the disaster occurs. This is a problem that is difficult for corporate decision-makers to accept. Therefore, Inspur Storage It is recommended that when it comes to data disaster recovery protection for small and medium-sized enterprises, investment resources can be reasonably allocated, and the disaster recovery center can be built into a second production center. Together with the production center, it can become the center for the enterprise to support the normal operation of the enterprise and achieve mutual tolerance. Disasters can not only reduce the total cost of ownership (TCO, Total Cost of Ownership), but also increase the return on investment (ROI, Return Of Investment).

In short, data disaster recovery and backup technology, with its increasingly mature technology development and civilian implementation form, can help enterprise users with different needs and levels to prevent the damage of sudden disasters and achieve the backup and protection of important data! Related articles analyze the three major disaster recovery technologies and the levels and technologies of data disaster recovery backup