Why enterprise datacenters are motivated to transform their next generation datacenter to on-premise CSP like infrastructures

by | Feb 13, 2017

This Document has limited distribution rights. It may not be distributed to the general public or shared beyond the employees of the licensee company. Upon the termination of an employee, whether voluntary or involuntary, access to this document must immediately cease. If the licensee company wishes to utilize this information in outward facing communications, please contact [email protected].

Introduction

Customer motivation is a powerful force for change. The world’s enterprise data centers are not only taking advantage of what off-premise services CSPs are offering, but they are also looking inward to see how they can capture shadow IT revenue for themselves.

Part of this investigation involves exploring new technologies developed by vendors that make these infrastructures more cost, space, and energy efficient. This involves a change in how an infrastructure is designed from a monolithic infrastructure to a distributed infrastructure that resembles more like the way nature consumes energy and communicates.

In addition, many vendors offer new technologies that support this shift in thinking and design. There are 4 main areas of motivation that seem to be popular for enterprise IT:

  • Economics of Scale;
  • Flexibility;
  • Efficiency; and
  • Competitiveness.

Economics of scale

Cost is one of the biggest motivators for IT customers today. There is a lot of marketing and advertising about new technologies supporting distributed scale-out architectures. Based on the needs and requirements of a particular datacenter, some of these technologies may not be appropriate or may not perform as advertised.

There are many factors that go into the cost of an infrastructure and some of the cost is still not well understood by IT customers to get a real total cost of ownership (TCO). Some of these factors will be based on:

  • Application needs and service level agreements (SLAs);
  • Budget initiatives and goals for the year;
  • Data migration services (local and geographic locations on premise and in the cloud);
  • Subscriptions to support a hybrid model when appropriate;
  • Scope, scale, and longevity (or refresh cycle); and
  • Data center and facilities infrastructure (power and cooling).

Flexibility

The ability to have more flexibility of where applications and data services live in a distributed scale-out ecosystem is very attractive to CSPs, and now for enterprise IT datacenters. Pay-as-you-grow or shrink-as-you-reduce has become more attractive from an incremental cost standpoint. Both hyperconverged and new scale-out solid state arrays, have made it easier to incrementally grow compute, network, and storage without needing large refresh budgets. In addition, using a distributed architecture in a virtualized environment increases flexibility of how compute, network, and storage are utilized in the infrastructure.

Efficiency

The ability to share resources and have a distributed ecosystem act as one system has been a dream of engineers for decades. Now, new vendors with new technologies allow us to do this in a more cost effective way, allowing us to come closer to realizing the dream that “the network is the computer” (Scott McNeally). This evolution to distributed architectures is a great contribution to overall infrastructure efficiency.

In a monolithic environment, there are wasted resources with the average computer used for native applications utilizing less than 10% of resources, and even with virtualization, it is between 30-50% on average.

In a distributed environment, compute and storage nodes can be much smaller, and do not have to carry every application supported by each node. Instead, simply deploy the necessary applications and services that are needed to support the infrastructure on each node as microservices, and they can float to where they are needed. This allows not only for less expensive equipment connected by high speed interconnects, but this allows for the next generation of low-power low-cost processors coming to market a chance to be part of the ecosystem to keep cost and efficiency.

Competitiveness

What is the motivation for enterprise IT to operate a datacenter like Google or Amazon? They are losing internal customers. That is why there are so many articles about shadow IT. It takes a long time to really understand what resources a department needs internally and IT to respond to the request, departments have turned to getting their resources in minutes through the public cloud and budgeting for these resources.

That is lost budget to the internal IT organization. But enterprise IT organizations realize they need to change their business model to their internal customers, and change their underlying infrastructure in order to support the new model. That means the methodology around how services are delivered needs to change to a rapid self-serving model delivering the resources and applications their internal customers need.

If enterprise IT doesn’t now show value to its customers those budget dollars will continue to go to external sources (such as SaaS providers) and make internal IT irrelevant. Some have embraced utilizing the public cloud and integrate them into their transformation by creating a hybrid cloud solution. If done right this can be a good cost effective solution and work in harmony in embracing shadow IT into their budget.

The second competitive motivation is data protection and security. If data is lost, corrupted or hacked, the IT organization has no way of protecting sensitive company data from these adverse actions if it is stored outside the company with no control from the company’s IT department. If IT is to be responsible for protecting the company’s sensitive data and secrets, then they have to be in control of where the data is stored and how it is protected.

Summary

There are many motivations that an IT department may have to transform their infrastructure to a cloud like environment. This transformation does require a major facelift to all areas of the enterprise. Moving from a monolithic architecture to distributed scale-out environments to support enterprise and modern applications, micro-services for flexibility, and a much more shared environment beyond just virtualization like containers architectures not only takes time but budget.

Deploying this new way of moving, storing, and protecting data will impact business models, services, and processes. This means the current state of the enterprise will be exposed and this exercise will help identify opportunities for investment for the new distributed architecture. Transform the architecture will transform how an enterprise does business.