How to do Capacity Planning for Your Data Center
Data center capacity planning for an enterprise is critical in wake of digital data explosion, and its ability to offer better insights into business processes. As storage costs are outpacing the other expenditures in IT, a good storage strategy can also help save costs. It can prove crucial in supporting the dynamic virtual environments allowing better server utilizations for uninterrupted business processes. In nut shell, the capacity planning is a proactive step towards knowing how much storage is available and how much more required, and how to leverage it for optimal performance of applications.
Learn further through the article to have a better approach for storage planning and proceedings for capacity planning.
Steps To Calculate Data Storage Capacity Requirements
- Collect the data: The immediate way to get started is to buy a tool designed to automate data collection. Before that, check with your storage service provider for reporting tools, or consider some hardware-independent products.
- Determine growth rate of data: Determining the growth rate can be done on a monthly or an annual basis and most economical method would be know how your storage capacity needs vary monthly.
- Consult with stakeholders: Consult with IT management to address complex levels of growth. Discuss the current organizational trends, cost and storage budget; it can assist in making an informed decision.
- Set realistic data storage goals: Get knowledge on a) How much storage space required per year? b) Are the applications tiered correctly? Find out if there are any new ways to improve efficiency of data storage. Also, set a goal to reduce the overall cost of storage.
Finding the Best Capacity Planning Tool
- Capacity forecast by utilization - If storage utilization is more than 85 percent, it’s imperative to add more storage soon to avoid applications glitch. A well-performing tool will monitor storage usage over the time and recommend storage purchase. These capacity planning tools typically yield more precise results as near-term and long-term trends can be chased.
- Translate growth rates into costs - If one can relate storage growth to the storage management costs, one can understand how compound annual storage growth rate affect your storage budget. "The magic number is 55 percent," says Phil Goodwin, president of Diogenes Analytical Laboratories Inc. "If your compound annual growth rate is 55 percent, then you have a neutral purchasing budget."
- Replace older tools with heterogeneous tools - No strategy can beat a way which reduces a number of storage management tools in an organization. Fewer tools will always simplify the learning curve for storage administrators. It is okay to use different management tools where needed, but the goal of simplification should be in the head.
- Storage management should include virtualization - Virtual servers are easy to create, move and delete. By opting server virtualization, one can host numerous logical (virtual) machines. The storage needs grow rapidly and unexpectedly after this, so it has to be on the top of storage capacity planning.
Virtual Machines (VMs) Change Everything
Capacity planning has become more unsustainable because, today, IT is more dynamic and complex—applications run in rapidly-changing, multi-layered, virtualized, and cloud-based environments. Hence, storage investments fail to sync with performance objectives and meet business goals. If investments are made on heterogeneous (physical and virtual) storage system, usage policies and reporting are centralized, and historical data trends observed, there will be a near possibility of forecasting future requirements.
Virtualization will complicate the picture, initially. A physical server has inherent physical limits; to consider buying more memory and CPU for increased data, or a whole new server. However, virtual servers surpass these limitations and are able to quickly scale up and down as per demands and changes made. But, this rapid scaling may add complexity to capacity planning, push the process of planning and growth prediction to the extreme.
In such scenarios, VM management system plays a vital role in diversifying the types of data and types of storage available satisfying certain criteria, such as high availability and data protection. To start with the planning process, try gathering data from VMs over a period and capture underlying trends. At last, quantifying and analyzing the trends will answer questions about capacity and performance of Virtual Machines, also help in determining what protocols, redundancy protection, and array features are to be implemented for the longer run.
Must Check: Storage Whitepaper