Storage For The Data Center Of The Future

The Data Center of the Future is envisioned as a computing utility that will use all of the server, storage and networking elements of the data center as a dynamic set of computing resources that can be flexibly allocated to customer applications. The ability to dynamically allocate some of these resource types, individually, in an uncoordinated fashion, already exists today. For example, the deployment of storage area networks (SANs) makes it possible to create pools of virtualized storage that can be assigned as needed to an application running on a server. Network services like load balancing, VLANs, and VPNs can also be virtualized to allow them to be dynamically assigned to applications. Server virtualization technology has recently become available from several server vendors as the last of the enabling technologies for the Data Center of the Future.

The IT industry is devoting a lot of engineering (and marketing) talent to this concept, which has also been referred to as the computing utility, grid computing, on-demand computing, the consolidated computing center, and data center virtualization. Whatever name you choose to call it, its defining feature is the automated dynamic assignment of server, network, and storage resources to business applications.

The key management component that will be needed to fully realize the Data Center of the Future is a Data Center Resource Manager (DCRM), which performs several important policy-based functions, utilizing the various system management utilities available in the data center as its eyes and hands. The DCRM maintains knowledge of what resources are available to allocate to applications, and schedules an application for execution based on priority and the availability of appropriate resources. It then allocates available resources to the application and configures those resources to host the application. It binds the application to its required data objects – logical units or file systems – and configures the data center security mechanisms to allow access to the application's storage from and only from those servers running the application. The DCRM then starts the application and monitors the execution of the application to detect aborts or performance problems due to the failure of data center hardware components, in which case it reallocates resources so as to keep the highest-priority applications running, even if that means aborting the execution of lowest-priority applications.