Use what-if analysis to meet data center capacity needs

The authors describe the challenges of capacity requirements and planning for dynamic IT architectures and why what-if analysis is a critical tool for capacity management in modern data centers.

Today, technologies such as big data, the Internet of Things (IoT), and artificial intelligence (AI) are booming, generating large amounts of data and enabling organizations to drive business innovation. However, they also consume a lot of IT resources. These emerging technologies require important basic computing and transaction support from data centers, and data center managers are struggling to keep up with demand.

How can modern data center managers respond with agility and flexibility to meet growing demands? The answer lies in dynamic, scalable provisioning and the ability to efficiently manage computing resources at the speed of IT.

[[205973]]

Common barriers to planning effective capacity

The data center is a massive system of interconnected devices, in some cases exceeding a million pieces of equipment from hundreds of vendors. Colocation facilities, data center migrations, and cloud computing/hybrid environments have further complicated an industry that was previously dominated by enterprise data centers. Therefore, configuration efforts must take into account the complexity of today's data centers.

However, many data center managers are struggling to understand the real-time capacity of power, space, cooling, network connections, power connections, and power loads at the data center level through UPS, control panels, iPDUs, and ports. So how do you know what equipment is there and where each device is located? How is it configured and connected? Most importantly, does the organization have enough space, power, cooling, and network capacity to provide new equipment for the business?

Understanding the correlations between these physical elements is critical to understanding potential shortages in data center resources. However, managing the complexity of data center capacity planning is difficult, inefficient, and unreliable with traditional spreadsheets and CAD drawing programs. That’s why so many organizations are adopting data center business intelligence software and dashboards to analyze, simplify, and accelerate capacity demand planning.

What-If Analysis: Beyond Generic Metrics

Understanding metrics such as floor space remaining, cabinets with the most open data and power ports, and budgeted power are solid starting points for capacity management. Capacity forecast graphs with historical trends can help create a clearer picture. However, to fully utilize existing capacity and purchase more if necessary, organizations need more advanced analytics.

That’s where what-if scenarios come in. When used as part of a data center infrastructure management (DCIM) software solution, what-if analysis allows organizations to understand the potential impact of changes in the data center, especially when adding and removing capacity, so they can quickly and accurately predict the future state and determine whether adding additional resources can be postponed or if more resources need to be purchased.

The simplest and most useful approach is to look at scenarios related to capacity on a per-project basis. For example, as a data center manager, you may have multiple concurrent projects in the pipeline. You may have several consolidation projects, projects to retire assets, and new projects for which you need to reserve resources. What-if analysis allows managers to see the net effects and tradeoffs for individual or groups of projects so that they can:

• Determine exactly if and for which projects more resources are needed. While an organization may want to avoid expensive capital expenditures whenever possible, the results of a what-if analysis may indicate that the organization will fall short and will actually need to invest more resources. Knowing which projects resources are needed can also help the organization save money accordingly.
• Most capacity planning metrics focus only on addition, but doing so creates an incomplete landscape that does not fully represent the organization’s actual capacity. Since the organization’s goal is to hypothesize the impact that analytics projects have on the organization’s capacity, it is critical to include both allocated and released resources.
• Communicate reasoning to management easily and effectively. As a data center worker, you need to analyze data at an abstract level where insights are easy to grasp. When explaining your chosen course of action to management, you need to provide a summary that clearly states the capacity that needs to be purchased or deferred. What-if analysis charts can provide intuitive analysis to present this information at a glance. This cannot be done if you waste time manually combing through spreadsheets to find significant data.

What-if scenarios can take your capacity planning to the next level, enabling provisioning at the speed of IT while also providing justification for spending to your management team.

How DCIM helps organizations meet their IT needs

What-if analysis is often just one component of a comprehensive DCIM solution that can help organizations quickly provision and plan capacity. Organizations also need to deploy power meters for iPDU and UPS metering, distribution circuits, feeder circuits, and various sensors for temperature, humidity, airflow, pressure, etc. deployed throughout the data center environment.

Once the necessary instruments and meters are deployed, the organization's DCIM software can collect and analyze data to easily view the capacity level of each component of the data center and determine where resources are over-provisioned, wasted or under-provisioned resources, potential risks, etc. Over time, predictive analysis and what-if analysis help organizations plan infrastructure deployment in a timely manner to support the launch of dynamic IT architectures, applications and services.

Putting it all together

While the metrics most commonly used to plan data center capacity are helpful, they are not enough to provide a complete picture of future capacity needs. What-if scenarios can bridge this gap by processing as well as supplementing so that organizations can accurately determine whether capital expenditures for new resources are needed. In addition, what-if analysis charts enable data center staff and their management to view the same data and understand at a glance the rationale for data center capacity plans.

With what-if analysis, organizations can quickly gain the insights they need to quickly configure and manage the needs of dynamic IT architectures. When what-if analysis is used with DCIM software that collects and analyzes instrument data through smart devices and sensors in the data center, it allows organizations to more accurately plan capacity and reduce waste, thereby maximizing productivity and improving customer satisfaction and the ability of management teams.