Virtualization and Cloud executives share their predictions for 2013.  Read them in this VMblog.com series exclusive.

Contributed article by Andrew Hillier, CTO and co-founder of CiRBA

Demand Management for Infrastructure a Key Focus in 2013

In 2012, most virtualized and cloud infrastructures were still teeming with excess capacity. While estimating high on infrastructure requirements enabled organizations to make the move to these new technologies with a lower risk of shortfalls, organizations are beginning to look at costs and density with increased scrutiny.

In 2013, organizations will turn their attention to the problem of over-provisioning, and will be exploring the levers that enable them to move to the next level of operational maturity and efficiency within virtualized infrastructures.  One of the key elements of this will be a push to measuring and modeling the true demand that will be placed on these environments. Operational teams will evolve from using simple capacity management processes and tools that trend utilization to estimating future capacity requirements based on models that are centered on active demand measurement and management.  

Most IT organizations currently lack the ability to accurately model future capacity requirements for virtualized infrastructure. They rely on old school methods founded in projecting requirements based on growth models, which creates one of two problems in new school environments: it either puts IT at risk of falling short of capacity and being unable to respond quickly to the business; or, more commonly, it forces IT to wildly over-provision capacity to reduce the risk of shortfalls in anticipation of what might be requested. Cloud stacks and self-service portals have been put in place to increase agility and responsiveness to the end users, but they actually make the management problem worse, as they increase volatility and make it even harder to forecast future requirements.

Organizations that succeed in 2013 will understand that the only way to effectively prepare for upcoming demands is to know what really is coming down the pipe. IT should be able to look forward in time to examine actual demand from lines of business and users and based on software release schedules, migration plans and organic growth factors, and use this to determine how much infrastructure will be needed. This visibility will enable them to determine if there is enough space to accommodate new workloads, and even reserve space for these new workloads on the planned placement date, enabling just in time hardware provisioning to control costs.

Essential to this strategy is enablement of new processes for modeling future utilization or demand, and combining these models with detailed analysis of current and future supply and demand, to get a precise forecast of capacity requirements. This forecasting strategy will ensure that applications have the capacity when they need it and prevent IT from having to over-provision resources to deal with uncertainty, thereby creating more efficient infrastructure. It will also enable IT teams to delegate the task of describing demand to the application owners and lines of business, giving them an active role in the process and allowing IT to get out of the thankless business of trying to figure out what users need, and instead focus on increasing efficiency and decreasing risk.

###

About the Author