Storage revolution in data center

Over the years, the server and storage environment have their own independent parallel development path. However, with the explosive growth of servers and storage arrays, IT managers are facing unprecedented challenges. In the past, they had to view these platforms as disparate entities, requiring different networks, management strategies, and even different employees to maintain the effective functioning of these platforms.

But as more and more companies launch data center architectures, the two separate islands need to be integrated. To further simplify and improve the performance of the data center, the enterprise needs to be fully merged, and by using a new set of technologies, the chances of success of this consolidation work are very high. These technologies include data center management and automation tools, blade servers, public and grid computing, storage area networks, grid storage, information lifecycle management, policy-based management tools, and the most important virtualization technologies.

In a server or storage marketplace, virtualization is not a new concept. Many enterprises have created separate server and storage pool of virtual technology, so that users and applications do not need to understand the physical components behind the server and storage resources, and these companies have gained great benefits from it. But integrated server and storage virtualization is the key to implementing a truly managed consolidation.

The origins of Virtualization

In the server market, virtualization was initially used in mainframes. In this environment, virtualization tools can help achieve workload management and improve mainframe utilization.

In the late 90, virtualization Tools for UNIX and Windows servers also emerged. These tools enable multiple virtual operating systems to run on the same physical machine, but logically and completely independent, with a complete hardware configuration. These tools are sometimes referred to as server resource management, including partition managers, virtual machines, virtual partitions, and logical partitions.

In the storage area, the earliest virtualization occurred in the early 90, the first RAID subsystem. By the end of the 90, many new storage virtualization technologies had emerged, most of which were aimed at improving management and utilization. Since then, storage virtualization has evolved from an independent technology to an important feature of storage infrastructure management tools. That is, it resides not only on the host server and on the disk array, but also on the intelligent switches in the storage network more and more.

Storage virtualization can also achieve higher levels of management capabilities. With virtualization features, data management tools can better handle snapshots, replication, on-demand capacity, and policy-based decision-making. Volume management is also considered a form of virtualization and is now a necessary tool for most data centers to manage storage networks and large storage arrays. In the coming years, it will become a standard feature in the entry-level storage array and is widely used in IP storage and entry-level storage networks.

Toward a fully virtual data center

With the development of server and storage virtualization capabilities, IT personnel are beginning to rethink their traditional device-centric client/server data Center architecture. Virtualization can rearrange components within a data center as a new model and treat them as shared resources. The end result is that all storage, server, and network resources will be virtualized as a pool of resources.

This ideal architecture can make full use of the evolving computer and network technology, improve the performance of each layer and greatly improve the speed and reliability of applications. The primary responsibility of the IT staff is to study how to deploy and manage these technologies and make them a practical datacenter architecture that emphasizes service. At the same time, large system manufacturers and emerging enterprises are also constantly introducing new technologies to enable data centers to develop into a practical model.

In the case of merging server and storage virtualization, management tools must have the ability to maintain and consolidate servers and storage resources. This type of management tool will emerge in large numbers in the coming years. Most of these technologies will appear in the Emerging data center Automation market. As more and more users deploy blades, a new generation of storage arrays, storage management tools, and larger storage networks, the market will expect to reach $1 billion trillion in revenue in 2006.

In the next five years, labor-intensive manual jobs like today will turn into automated and highly intelligent jobs. However, the biggest problem is that most virtualization tools have a large degree of relevance to the hardware platform or operating system, so their ability to support multi-vendor hardware and software platforms and heterogeneous systems becomes critical.