Software-defined data centers have endeavored to bring software-enabled optimization that exists at all layers of the stack today (storage layer, CPU layer, network layer, etc.) into the physical data center, giving adopters the ability to manage those assets in a way that correlates to the workloads utilizing them.

While this concept is still an emerging term and practice (coined in 2012), there is no denying its current prowess and expected future success. Recent statistics show the software-defined storage market was worth about $360 million in 2013 and is expected to explode to a worth of about $3.7 billion by 2016. With this kind of growth on the horizon, we wanted to take some time to define five attributes that make up a successful software-defined data center.

Security/Compliance

.

• As communications are happening through many different avenues (cloud, personal devices, virtual portals, etc.), business has become enhanced, placing a massive amount of strain on IT. Older data-centers are equipped with physical security measures, but are ill-equipped for cyber-attacks that may occur power and cooling infrastructure. As a result, it is vital to ensure the software-defined data center has taken a closer look and modernized the security profile of the data center itself. (Reference)

.

Reliability

 .

• Data center downtime remains a costly line item for organizations and costs have continued to rise in the past three years, according to a Poneman/Emerson Network Power survey. While costs continue to rise as a result of downtime, the number of causes of failures has shrunk thanks to the virtualization of most physical IT infrastructure. The survey results show 85 percent of respondents reported that they had lost primary utility power to their facility at least once during the past two years.
• To solve these issues, software-defined data centers should look to create layers of abstraction that isolates applications from local power dependencies. Power should move the application load not only at times of a disaster, but fluently and dynamically to ensure uptime is maintained. (Reference)

 .

Efficiency/Sustainability

 .

• Energy efficiency remains an item at the top most enterprises cost concerns list. Power stands as the largest operational cost within massive data centers, creating a need for a solution that maximizes energy efficiency without sacrificing performance. (Reference)
• Today’s software-defined data centers should look into solutions like:
  • Allowing power and cooling capacity to be provisioned based on individual application’s needs.
  • Ability to move workloads to other data centers where energy costs are lower.
  • Panoramic visibility into the entire data center footprint so informed efficiency decisions can be made.

.

High Performance Support

 .

• High performance computing (HPC) is both computationally and data intensive, a practice often used by both enterprise and governmental organizations. HPC equipment traditionally operates at a higher density per rack than less computationally intense equipment, straining power and cooling infrastructure. Today’s software-defined data centers need to have the ability to support these heavier loads through data modules or other solutions. (Reference)

.

Agility/Scalability

 .

• Growth is the end goal for any business. More users and visitors generally equate to more revenue, and goals become met. However, when business expands rapidly, servers begin to work harder, and the pain of having to expand the data center becomes real. A software-defined data center should allow data center users to meet increased demand without massive physical upgrades. While the purchase of data center modules would meet demands, it wouldn’t come without a significant price tag. Software-defined data centers allow modular data center additions at a smaller, “as needed” scale. (Reference)

.

As software-defined networks and software-defined data centers continue to come to the forefront of data center discussions and decisions among IT executives and business leaders, it’s important to recognize these five attributes as vital to success. When installed, ran, and used correctly, software-defined data centers can provide a reliable solution to bridge the gap between legacy enterprise applications and new cloud computing services.