Problem solve Get help with specific problems with your technologies, process and projects.

Five steps to calculate data storage capacity requirements

This tip offers five best practices for controlling storage costs with effective storage capacity planning.

Digital information is empowering businesses at an increasing rate. The information stored is now of critical value...

to both large and small companies. As a result, storage costs are quickly outpacing other areas of IT spending. The key to controlling the costs is good storage capacity planning. Getting started with capacity planning is like saving for retirement, you can't start too early. In this tip, I'll cover five key steps that will lay the foundation for good storage capacity planning.

  1. Collect the data. You may not be ready to use the storage capacity information to transform the storage service just yet, but when you are, you'll wish you had a baseline with history. The fastest way to start collecting data is to buy a tool designed to do so. The marketplace has quite a few, but first check with your storage manufacturer for reporting tools, or look at some hardware-independent products like those from MonoSphere Inc.

    If you don't have the budget for a shrink-wrapped tool, you can always create your own. At its simplest, many small-midsized business (SMB) shops use spreadsheets to collect and store the data. However, this method can be labor-intensive when you grow beyond a couple of storage arrays and 10 or so hosts. If you are so inclined, you could also write your own collection scripts and store the data in a database. This should scale pretty well and will allow daily data points without much more effort.

    Whether you buy a product or not, be sure to collect the following key elements: On the storage infrastructure side, collect the raw, usable and allocated capacity for each tier. For storage networking ports, count the total number of ports and the number in use. For each server that uses non-local storage, grab the total capacity presented, the file system usable total, the file system utilized total and the database utilized by tier. If you create your own spreadsheet, for each data element, use Microsoft Excel's SLOPE function or a linear regression algorithm to determine the growth rate for each data element over time.

  2. Define standardized metrics. As the data starts to build, it can be overwhelming. Without a plan, the new information actually creates more questions than answers. For this reason, it's important to have a few key metrics that will be used to measure the storage service as a whole, and each internal customer's usage data.

    The storage service should be measured to make sure it's performing as desired. Key metrics include the total raw, usable and allocated capacity by tier over time. Using the growth rate of each tier and the total capacity available, determine the date when the current storage will be full. This helps plan storage purchases so you can take action when the growth doesn't match the budget. Also, calculate the cost of the storage service in cost per terabyte by dividing the total storage budget into the number of usable terabytes on the floor. An interesting way to combine cost and utilization into one simple number is a metric called the "effective cost of storage." This is the total storage budget divided by the utilized storage. Storage managers can extract more value to the business by lowering the storage budget or increasing the utilization -- one simple number, two good results.

    In the next step, we'll be meeting with application owners, so it would be a good idea to develop application by application utilization, growth rate and cost metrics similar to the overall infrastructure number, but specific to an application.

  3. Meet with the stakeholders. Take the standardized metrics collected above and meet with the internal application owners at least quarterly to review the metrics. Help them to see the utilization, growth rates and costs of their application. Use this as an opportunity to capture the business drivers that may affect storage growth, manually overriding the calculated growth rates when applicable. Working with the app owners on storage capacity will help them understand the impact of their actions and the IT group's intention to improve the service.

    Meet with IT management on a monthly basis to review the storage service numbers. Discuss the trends, costs and storage budget. When senior management feels informed about the business drivers effecting storage and less surprised about immanent purchases, life will be easier for everyone.

  4. Set some goals. After meeting with the stakeholders, it's time to set some improvement goals. Decide how many days of storage to keep on hand. Will you buy storage once or four times per year? Set a goal for reducing the effective cost of storage, 15% per year should be attainable. Getting to that new number will drive several questions: Are the applications tiered correctly? Should you increase utilization or allocation rates? How can we lower the purchase price of storage?
  5. Make some changes and measure again. In order to meet the goals we set in step 4, some changes will be required. Reclaim the underutilized storage from applications, or maybe it's time to brush up the storage request process to ensure too much storage isn't being allocated to applications. Lowering the effective cost of storage may also require new purchasing practices like RFPs or aligning the buys with a manufacturer's end of quarter to get better discounts.

Storage capacity planning is critical to controlling cost and stability. Storage budgets aren't growing at the same rate as data. Managing resources better isn't an option. Taking these five practical steps will bring you to mature capacity planning.

About the author: Brian Peterson is an independent IT infrastructure analyst. He has a deep background in enterprise storage and open-systems computing platforms. He has consulted with hundreds of enterprise customers who struggled with the challenges of disaster recovery, scalability, technology refreshes and controlling costs. 

Dig Deeper on Storage management tools