With the advent of the era of big data, the storage demand of data volume is growing rapidly, and storage technology is attracting more and more attention from the industry. As a result, more and more enterprises take data storage as an important project to manage, thus bringing the rapid development of storage management technology. However, the diversity of storage devices makes it difficult to manage these devices efficiently. Virtualization is a promising solution for storage management. Storage virtualization technology solves the problems of storage device management efficiency, different types of storage resource integration, and the compatibility, scalability, reliability, fault tolerance, and disaster tolerance of heterogeneous storage systems.
The definition of storage virtualization by SNIA is that the abstract, hide or isolate the internal functions of storage (sub-) systems or storage services, so as to separate the management of storage or data management applications, servers, and network resources, so as to achieve the independent management of applications and networks.
Storage virtualization technology is an important structure of computer virtualization technology. Its idea is to separate the logical image of resources from physical storage and to provide a simplified and seamless virtual view of resources for the system and administrators. To users, virtualized storage resources are like a huge "storage pool," with no visible disk or care about which storage device their data is on. Storage virtualization technology has the ability to improve dynamic adaptability. It centralizes the storage resources into a large-capacity resource pool, changes the storage system and realizes data movement without interrupting the application, and realizes single point unified management of the storage system.
Storage virtualization can be realized at three levels: host-based virtualization, storage device-based virtualization, and network-based virtualization. It is implemented in two ways, namely, in-band virtualization and out-of-band virtualization. The results are block virtualization, disk virtualization, tape, tape drive, tape library virtualization, file system virtualization, file/record virtualization.
Host-based storage virtualization
Host-based storage virtualization is generally completed by the logical volume management software under the operating system, and different operating systems have different logical volume management software. This implementation enables the storage space of the server to span multiple heterogeneous disk arrays and is often used for data mirroring protection between different disk arrays.
Storage virtualization based on storage devices
Storage device-based storage virtualization, which adds virtualization capabilities to storage controllers, is commonly seen in mid - to high-end storage devices. Its purpose is to optimize user-oriented applications. It can fuse users' different storage systems into a single platform, solve data management problems, and realize information life cycle management through hierarchical storage, so as to further optimize the application environment. This technique is mainly used within the same storage device for data protection and data migration.
Storage virtualization based on network
Network-based storage virtualization is achieved by adding a virtualization engine to the storage area network (SAN). Mainly used for the integration of heterogeneous storage systems and unified data management.
Many people are concerned about where virtualization is deployed, whether it's host-based, network-based, or storage-based. Of course, all of these options have their own advantages and disadvantages. The most important is to allow storage virtualization to merge heterogeneous storage systems, merging them into a common storage pool for common management and protection. The goal of storage virtualization is to give enterprises and users the ability to select storage arrays from different vendors by providing them with enterprise-level capabilities such as dynamic configuration and live data migration.
When enterprise users apply storage virtualization technology, they need to do several tasks: First, plan the storage virtualization strategy of the enterprise; Secondly, the enterprise business data should be properly classified and planned. Third, understand the data life cycle; Fourth, planning the storage level and service level; Fifth, consider security and compatibility issues. The maximum utility of storage virtualization is realized as far as possible to better save resources and simplify management. Now, with Vinchin Backup & Recovery backup solution, the enterprises can as much as they want to use their virtualization storage for data backups. By the Data Deduplication and Compression of Vinchin, can help deduplicate and compress data from a backup repository to reduce storage space and costs.