Virtualized disaster recovery ensures data security

When the master server suddenly fails or stops working due to other damages, it starts to start the VM that is synchronized with the master server and backed up, send an email to the administrator or send a notification directly to the mobile terminal of the Administrator. After the master server recovers, the VM includes the operating system, database, and application.ProgramAnd other related data are seamlessly migrated back to the original master server. To complete these operations, you only need to click a few clicks, and the user does not feel any business interruption. "All this is the convenience brought about by the Disaster Recovery Solution Using virtualization technology ." Yang inghong, an engineer at Novell, said.

A large number of users' business systems rely on it for support. If there is a problem in the IT system environment, it will be difficult to ensure the normal operation of the business and bring about great losses. Various disaster recovery technologies and systems are popular in the market. However, due to the high installation, configuration, and complexity of traditional disaster recovery systems, they are expensive, 80% of users' investment only protects 20% of servers, and does not guarantee service operation.

Disaster recovery aims to restore the system immediately after a disaster occurs and put it into use as soon as possible. Therefore, disaster recovery uses various technologies, including data backup, data replication, and other technologies, will focus on the continuity of the business. The main indicators for measuring these disaster recovery technologies are RPO (recovery point object, recovery point objective) and RTO (recovery time object, recovery time objective ).

Test recovery time

In the Disaster Recovery Solution of virtualization technology, the target to be backed up is defined as a workload, which refers to some application runtime environments independent of the hardware platform, including operating systems, data, and applications.

The core idea of the virtualization technology disaster recovery solution is two-way replication. Data in other places generates an available copy in real time, which can be put into use without data recovery, restore the service after the interruption is restored. The biggest advantage of data replication is that the copy data is available immediately and there is no data recovery time. RTO is very good. Because it is real-time replication, RPO is also very good, and almost no data is lost.

Let's see how the disaster recovery system of virtualization technology works, as shown in solution 1. For devices that have installed the Novell automatic migration technology, such as a server or a disaster recovery device provided by the vendor, the first step is to use p2v (physical server to virtual machine, physical server migration to virtual machine) technology, migrate physical server workload to virtual server, or use V2V (Virtual Machine to virtual machine, virtual machine to virtual machine migration) technology, migrate the workload of protected virtual machines to the disaster recovery system through the network for centralized operation as virtual machines. This virtualization-based disaster recovery technology can provide users with a system that supports both local disaster recovery and Network Disaster Recovery.

You can set it according to the policy. For example, a physical server can be used to operate an external system and set an incremental replication period from minute to week. Each time a physical server (an external server) has any data changes, the network can be incrementally synchronized to the corresponding virtual machine in the disaster recovery system. This ensures that all the updated data of the server is synchronized to the disaster recovery system while the server is operating. Once a physical server or protected system fails, the disaster recovery system can quickly take over the service.

Because the virtual machine environment is the same as the protected server environment, it can provide external services after it is started. The user connects to this virtual server in the disaster recovery system to ensure that the protected server has the least downtime and the least data loss. After the protected physical server is repaired, run the V to P (Virtual Machine to physical server, virtual machine migration to physical server) operation in the disaster recovery system, that is, the virtual machines in the disaster recovery system can be migrated back to the physical server. This is the entire disaster recovery process.

Bidirectional migration

One of the most critical technologies in the virtualization disaster recovery solution is the two-way migration technology. 2, we can see that P (physical server, physical server), V (virtual machine, virtual machine), I (image archive, image) two-way, automated, and policy-based migration can be achieved between the three. P2P migrates the workload from one physical machine to another. p2v migrates the workload of one physical machine to a virtualization platform; v2V and v2p migrate virtual machines from one virtualization platform to another through the network, or migrate workload from one virtual machine to one physical machine.

After understanding the two-way, automated, and policy-based migration between P, V, and I, it is easy to understand the virtualization-based disaster recovery solution. Yang inghong said that more vendors, including Novell, are currently starting to provide complete and reliable migration products.

A major disadvantage of traditional remote disaster recovery is the high cost. Not only is the data replication system expensive, but it also needs to increase the hard disk storage space and host system. Considering the cost of network wiring and remote data center, the cost of remote data disaster recovery is greatly increased, the maintenance cost is much higher than that of data backup. Virtualization Technology is also highly competitive in terms of energy saving and consumption reduction because of its low cost.

The advantages of the virtualization disaster recovery technology include: in the case of system failures or disasters, the process of recovering workload is very simple; to reduce the cost of disaster recovery solutions, usually, one-to-one redundant hardware and software resources are required. It is better to implement disaster recovery and facilitate the completion of Continuous testing, and can protect the workload in the Hybrid Processing and virtual machines. during incremental replication, you can set network bandwidth to avoid network bandwidth depletion and affect other business operations.

If the physical server breaks down completely, you can replace a new server to restore the disaster recovery system to a new device. For example, when the master server is upgraded and the hardware configuration of the server changes, the VM can still be migrated or restored to the new server, or from the workstation to the blade server.

Save money and energy

With the increasing importance and closeness of IT and business, users also adopt some disaster recovery solutions, but a large number of servers are not well protected. Because the current cost of some disaster recovery servers is very expensive, especially for remote or cross-Wan disaster recovery, it is unacceptable for most users. Because the adoption of virtualization solutions can effectively reduce costs, most small and medium-sized enterprises, or some branches of the company, can adopt this technology for disaster recovery.

Second, some server users think it is not very important, such as the Mail Server File Server and web server. In fact, they are also very important and contain some important information. If these servers fail, they will also have some impact on users' business.

Applicable to temporary disaster recovery. For some large users or enterprises, they have their own R & D and testing systems. Because the testing process needs to last for a period of time, temporary disaster recovery is also required, without disaster recovery, testing is easily interrupted and the system recovery is difficult. Another type of customers is small and medium-sized customers who are willing to choose cost-effective systems.

In addition, Disaster Recovery Technology Based on host heterogeneous replication is one of the future development trends and has a broader market. Because most customers have heterogeneous host environments and support data replication technology in heterogeneous environments, they can use the existing environment to replicate each other among hosts. virtualization is the solution to this goal.

Source: http://server.ccw.com.cn/jssc/htm2009/20090403_612091_2.shtml