Data Centers: Cloud-centric Architecture

Over the past several decades, data centers have evolved from single-point, concentrated processing centers to dynamic, highly distributed, rapidly changing, virtualized centers that provide a wide variety of services to highly distributed global user populations.

A powerful new cloud computing paradigm has emerged wherein users request and receive information and services dynamically over network from an abstract set of resources. The resources are somewhere out there in the cloud. Users dont care where or how the resources are provided; they only care that their applications, data, and content are available when needed, and at the desired levels of performance and security.

To optimize results, CIOs are looking to see if the lessons of the cloud can be extended to their own IT departments. With cloud computing, the network has become paramount. Everything now runs over a network, within and across systems in the data center, between data centers, and between data centers and users. Globally, the network serves as the delivery and access vehicle. Without the network, there is no cloud.

As demands for cloud computing services grow and change at an exponential pace, it becomes extremely critical for data center planners to address both current and evolving needs and to choose a flexible, dynamic cloud data center design that can effectively meet the complex challenges of the global information age.

Getting Ready

In the face of exponentially increasing complexity in compute and networking systems, it becomes critical to design data centers that reduce complexity.

Thus, we need an approach and products that greatly simplify data center design, deployment, and operation.

Simplify: Networks built on fragmented and oversubscribed tree structures have problems with scaling and consistent performance. Simplifying the data center network means minimizing the number of network elements required to achieve a particular design, thus reducing both capital and operating costs. Simplifying also means streamlining data center network operations with consistently implemented software and controls, meaning fewer devices, a smaller operational footprint, reduced complexity, easier management operations, and improved application performance.

Share: The economic and agility imperatives of the cloud-ready data center require network resources to be allocated, expanded, and reallocated efficiently at scale. Sharing the data center network essentially means intelligently partitioning the infrastructure to support diverse applications and user groups, and interconnecting large pools of resources with maximum agility.

Secure: Securing the data center network extends protection to support the rich, distributed architectures that many applications currently use. This requires a robust, multi-dimensional approach that enhances and extends traditional perimeter defenses. Increasing the granularity and agility of security policies enables trusted sharing of incoming information and resident data within the data center, while complementing the functions embedded in operating systems and applications.

Automate: Automating means the ability to capture the key steps involved in performing management, operational, and application tasks, and embedding task execution in software that adds intelligence to the overall data center operation.

Conclusion

Everything now depends on network communications. And, with network complexity growing exponentially in tandem with the growth in demand for global 247365 services, distributed processing, logical systems, virtual machines, multi-tier architectures, and inter-data center communications, all placing extreme burdens on traditional networking infrastructures, a new approach to the network is essential for the introduction and success of cloud data centers.