New Data Center Innovation Stack Could Redefine IT

By Chris Preimesberger

New business pressures, such as the need for real-time information and collaborative applications, are putting the effective storage and accessibility of business data at the center of the new IT. This, naturally, is requiring more agility from IT systems; the newer ones generally include this, but the older ones usually need upgrades.

The drive toward cloud-computing models has resulted in a strong emergence of converged infrastructure or “integrated stack” solutions. An integrated stack is a prepackaged and rapidly deployable system typically containing a mix of network, storage and server infrastructure bundled with virtualization, automation and orchestration software. The purpose of these stacks is to simplify purchasing requirements and accelerate the move to a virtualization or cloud-computing model.

There are two basic types of integrated stack solutions: 1) single vendor, in which all stack components are developed, manufactured and bundled by a single vendor; and 2) multivendor, in which products from two or more vendor partners are specifically engineered and bundled together to create the stack.

Highly mobile and social-media user expectations and shifting application requirements are imposing new demands on data center and cloud infrastructures. Data center managers need their IT infrastructures to be flexible, programmable, efficient and scalable; it is virtually impossible for a single vendor to innovate and deliver on every layer of the stack.

The key layers of the innovation stack are storage media, storage, network, processing, business logic and presentation. In the new stack, these layers are all implemented as services that manage their respective dynamic resource pools (compute, memory, storage, networking) rather than statically defined virtual machines.

A software-defined data center enables the innovation stack to define and allocate virtual resource pools for compute, networking and storage to all applications on demand, via common APIs. It also virtualizes the network for reliability and maximum resource utilization, in addition to transforming physical storage media to align it with logical application demands. In this setup, separate component management tools give way to full-stack automation and orchestration by policy.

A leading innovation stack example today is Amazon Web Services. AWS built an elaborate layer of services for developers—including multiple database options—that support multiple programming languages with management tools that are delivered via intuitive interfaces. Government and commercial organizations now are seeking stack providers that deliver the type of powerful, layered services that Amazon has built.

Several key verticals already have started to embrace the operational efficiencies and business benefits of the innovation stack. These include online retailers, advertising and marketing networks, and transportation companies that need to optimize trains or coordinate traffic lights to expedite the route for an ambulance going to a hospital, for example. Health care-related companies have adopted this infrastructure for automating electronic medical records, and X-ray and MRI data. Biotech firms are using it for sequencing DNA.

In 2014, IT organizations are being judged on speed and agility to keep pace with on-demand business challenges. This requires a fundamental shift in the way organizations plan and execute their data center strategies and demands that a strategic focus be placed on the effectiveness of their current IT.

The innovation stack provides an answer for the demands of increasing data stores and user expectations for real-time information. IT organizations need to infuse infrastructures with new flexibility and scalability, going beyond the static restrictions of conventional siloed systems. The new innovation stack stands to gain significant traction in 2014 because it delivers this business agility.