In every industry, demand for IT capacity continues to grow. Digitalization, the Internet of Things and cloud computing are just a few of the factors that require vastly increased compute, data and networking resources.
The solution is to go hyperscale: in other words, to reimagine the data center in order to scale beyond the cost and capacity limitations of today's architecture. Hyperscale data centers are smarter, more flexible and better able to achieve increased resource utilization levels while reducing total cost of ownership.
Transforming data centers in this way goes beyond hardware to impact the software layer, including hypervisors, operating systems and cloud platforms. Composition and resource orchestration, workload execution and – ultimately – data and applications are just some of the areas transformed as a result. In fact, this architecture allows for entirely new approaches to building, managing and running services and applications.
However, challenges must be overcome if hyperscale is to be fully realized. As well as revealing the scope of this vision, this paper sets out the key considerations and questions for data center operators at every layer – from hardware to applications.