The data center is a crucial part of today’s society. Without it, the world would come to a grinding halt. Data center managers are cautiously implementing private cloud computing by incrementally leveraging virtualization technology to automate the management and deployment of applications and workloads – a pragmatic way to transform utility computing with automation.

What gets data center executives excited about a private cloud initiative is the promise of instantly scalable resources at a lower cost. But what gets data center operations managers really excited is when they hear security breach or system outage. There are some harsh realities that exist for the data center operations managers that create a gap between promise and reality. 

The first reality is the regulatory environment within the data center. Regulatory auditors are accustomed to doing things in a very safe manner. This creates the server-hugging data center manager, who believes he or she can trust the cloud, but must show the controls in place when managing customer and financial data. The result is a less agile implementation.

The second reality is limitation of virtual machine scaling to cover workload demands. Simply implementing virtualization technology in the network does not guarantee efficiency or redundancy. Recovering from a failure is hard, so operators tend to add and take away resources incrementally. This defeats the notion of instantly scalable. 

The final reality is that cost. Data center managers are inherently conservative. The tendency is to provision the processing power and storage needs to peak workloads plus some design margin. The database administrator may also look to add additional margin to ensure uptime. This margin stacking drives up the cost and can wipe out any expected savings by moving to the cloud.

So what’s the final score? Implementing a private cloud strategy can be scalable and cheaper, but unique automation tools are the critical link. These tools enable data center operations managers to provision when they need it and only what they need. Only then can they meet the performance and cost goals set by their data center execs.