Virtualization Technology: Disaster recovery in the cloud computing era

Traditionally, a business needs to develop a disaster recovery plan based on their respective recovery-time objectives (RTO), enabling them to recover operations within a certain amount of time. Typically, this may involve rebuilding infrastructure and applications over a period of more than 72 hours, and handling the purchase and deployment of tangible assets. In reality, this process may take a longer time. Downtime for critical applications is by no means an option, and people always recommend a fully redundant, http://www.aliyun.com/zixun/aggregation/6939.html "> Proactive solution, Allow two of these production systems to run in parallel in two geographically diverse facilities. In the past, this would cost six to eight times times more than a redundant solution.

Configuration speed

This situation has changed with the flexibility of cloud computing and virtualization technology. Now, the enterprise can deploy computing resources very quickly, dozens of or hundreds of servers can be done in an hour. This means it can replicate its production infrastructure without the need to buy additional servers and handle complex logistics systems. The server is only the first step in this process because it also needs to be restored and installed by the operating system, database and application.

Before that, the process of loading programs to the server may take days or even weeks. Another great benefit of cloud computing and virtualization is the ability to replicate data and configurations for computing environment snapshots at a particular time. This makes it easier to recover data or back up the data center in another location. Once the redundant hardware is in place, the snapshot begins to recover, reducing the time it takes to load the kernel software, namely, the operating system, the database, and the application tier.

Although these virtualization tools and technologies accelerate this process, and allow the computing environment to quickly supply relevant details. However, each system is different in application and infrastructure, for example, some systems have multiple application servers with load balancing and different databases on separate physical servers or virtual servers.

Blending method is preferred

It is for this reason that the hybrid cloud computing approach is an ideal choice because it combines physical, proprietary hardware, and technology, both within the enterprise or with an external data center. The enterprise still needs physical hardware to make the most of existing investments, improve data security in a physical device isolation, or improve performance as a dedicated power source.

Another scenario that utilizes the physical hardware for a disaster recovery strategy is particularly helpful when restoring a database. At the same time, there is an ongoing loss of orphan data from the last backup, causing the infrastructure to malfunction or catastrophic events. Databases also require enterprise security concerns because they can contain sensitive data, typically hosted in a cloud service, even in the event of a disaster. For many organizations, it makes sense that databases and other sensitive applications reside on dedicated hardware and can be replicated to other physical hardware in other locations.

Using a hybrid approach, cloud computing can be well applied to applications and Web servers, restoring production environments with snapshots, and then connecting to databases in disaster situations. In addition, applications and Web servers can be run in the cloud and connected to a database, but only a small computational task and memory are allocated to them. When a disaster occurs, more memory and computing power can be allocated almost seamlessly. This approach not only reduces recovery time, but also means a fully redundant system that can reach a production system of less than twice times the cost, rather than six to eight times times the previous cost.

Disaster recovery Strategy

It is advisable for an enterprise to place some datasets and applications within the enterprise and to place other data in the cloud services or Third-party data centers, even if this policy is mandatory. After all, if a company is hit by a disaster, it can minimize losses. But it also depends on the enterprise's ability to perform disaster recovery strategies. Use third party services to eliminate any potential dependencies and immediately increase resilience and strain capability. Other advantages of outsourced infrastructure include the ability of companies to leverage their investments in cloud computing, virtual technology, and skills.

There is absolutely no excuse for a business not to develop a disaster recovery plan for its core applications and infrastructure. Actively adopt technologies such as cloud computing and virtualization to replicate advanced storage hardware platforms and, when needed, adopt third-party services to quickly and cost-effectively build a disaster recovery platform.

(Responsible editor: Schpeppen)