Maximizing the Benefits of Capacity Planning
IT is the epicenter of your company's nervous system. Capacity planning can keep it healthy. We examine planning methods and highlight their pitfalls.
- By Scott Adams
- February 7, 2006
Think of IT as the epicenter of your company’s nervous system – an enormous network of nerves that coordinates the actions and reactions of customers and employees. IT must ensure information is processed without delay throughout the organization.
Chaos, lost revenues and a damaged brand can result from just a minor delay. Today’s IT leaders understand the need for having and implementing a capacity plan. They know the cost savings/benefits, functional components, and potential pitfalls of a well-oiled, smooth-running data center.
The “capacity planning = buy more hardware” approach of the dot-com era increased costs and made management difficult. Throwing money at an availability problem (i.e., buying more hardware) only adds to the inefficiency and complexity issues in your data center. This could lead to waste and costly downtime. However, some companies continue to operate in this mode.
The best capacity planning/performance management software can help you identify as much as 50 percent of the IT resources sitting idle in your data center. After discovering your underutilized resources, you can deploy new services without spending more capital.
Capacity planning software can predict response times from components in a multi-tier environment so IT can be prepared to process forecasted increases in business data while meeting service level agreements without wasting resources. You can even mitigate the risk of stacking multiple applications on a single server by understanding the impact on service before consolidation.
Performance tools can also report on service levels, analyze end-to-end performance, and investigate bottlenecks in a variety of ways (e.g., drilling down to the users and processes active at the time of a problem), explore cause-effect relationships, and uncover cycles and patterns in system behavior.
Techniques for Success
To realize these benefits, several capacity planning techniques are available—each useful in different situations.
- Trending uses simple extrapolation of resource utilization over time. The advantage of this technique is that it does not require sophisticated tools. A spreadsheet will do, but it can be difficult to take into account non-linear behavior buried deep within a multi-tiered system.
- Linear trend analysis looks at historical data and projects a linear trend line, applying upper and lower confidence intervals as well as a threshold at which resources will become inadequate. This capacity planning method is a quick sanity check and identifies over-utilized resources, but it does not allow for experimentation with different configurations.
- Simulation modeling, a much more sophisticated capacity planning technique than trending, simulates the queuing events that occur during execution. The downside of simulation modeling is that it can be time-consuming to build and run the models.
- Analytic modeling calculates how a queuing network will perform. A model is built based on a description of the system. Hypothetical changes can be made to system configuration or business workloads, and the model will predict how the changes will affect performance. Done properly, analytic modeling is a fast and accurate capacity-planning technique.
Analytic modeling and load testing can be used together to determine the optimal configuration for meeting required service levels when rolling out new or modified applications. Time and cost considerations often make it prohibitive to conduct such tests on the actual hardware using production-level workloads. Instead, smaller representative loads can be applied to a scaled-down set of test servers while performance analysis software takes a baseline reading of performance. Analytic modeling can then be used to rapidly predict how various configurations will perform under a production-level workload, all without the need to purchase and test with the actual configurations under consideration.
One critical objective of capacity planning is to ensure current and future needs will be met while mitigating risk and minimizing costs. When you can confidently plan for business forecasts and consistently meet service levels, you will provide more business value and demonstrate to the CFO that IT contributes to the bottom line.
Unfortunately, management complexity, budgets, transaction volumes, lack of business planning, and the overall business environment all contribute to the difficulty of planning for future resource needs. Without understanding the complete business environment, most IT leaders will fail in their attempts to accurately predict needs.
Gartner notes that the lack of capacity in a company’s infrastructure is often blamed on the IT organization, when in reality it is the business units that are unable to judge or provide the needed information. Many times the business units see it as an IT problem, not as a business problem.
Another issue many IT managers face is the continuing expectation that IT should do more with less. According to Gartner, through 2008, the costs of data center facilities, maintenance, and support will continue to rise by between 3 and 8 percent per year. As costs rise, IT will continue to be held accountable for measurable results. The good news is that capacity planning can help an IT organization get the most from IT, meeting service levels with a minimum investment.
Capacity planning is an ongoing, strategic function that helps IT leaders address hot topics such as cost reduction, risk mitigation, alignment with business priorities, and disaster recovery.
By creating an optimization plan with your vendor, IT will be more proactive and will help the company meet or exceed customer expectations.
Scott Adams is a senior software engineer for TeamQuest Corporation.