Using What-If Analysis to Address Data Center Capacity Demand

Aug. 31, 2017
A data center is an interconnected system with thousands or, in some cases, over a million pieces of equipment from hundreds of vendors. With what-if analysis done right, you can quickly derive the insights you need to quickly provision and manage resources for the needs of dynamic IT architectures. 

In this week’s Voices of the Industry, Herman Chan, President of Sunbird Software, discusses the challenges of capacity demand and planning for dynamic IT architectures and why what-if analysis is a critical tool for modern data center capacity management.

Herman Chan, President, Sunbird Software

Technology trends like big data, Internet of Things (IoT) and artificial intelligence (AI), are generating massive amounts of data that enable organizations to drive business innovation. Yet the benefit to the business often comes as a heavy drain on IT resources. These emerging technologies require significant underlying compute and transactional support from data centers, and today’s data center managers are struggling to keep pace with the demand.

How can modern data center managers address this increasing demand with responsiveness and agility? The answer lies in dynamic, scalable provisioning and the ability to effectively manage compute resources at the speed of IT.

Common Hurdles to Effective Capacity Planning

A data center is an interconnected system with thousands or, in some cases, over a million pieces of equipment from hundreds of vendors. Colocation facilities, data center migrations, and cloud/hybrid environments are further complicating an industry previously dominated by enterprise data centers. Provisioning efforts must consider the complexities of today’s data centers.

However, many data center managers struggle just with understanding real-time capacities for power, space, cooling, physical network connectivity, power connectivity, and power loads—from the data center level down to individual UPSs, panels, iPDUs, and ports. How do you know what equipment you have and where each piece of equipment is located? How is it configured and connected? Most importantly, do you have the space, power, cooling, and network capacity to provision new equipment for the business?

Understanding the correlations among these physical elements is critical to knowing where you may have a shortfall of resources. Managing the intricacies of data center capacity planning, though, is unnecessarily difficult, inefficient, and unreliable with traditional spreadsheets and CAD drawing programs. That’s why so many organizations are turning to data center business intelligence, dashboards, and analytics to simplify and accelerate capacity demand planning.

What-If Analysis: Beyond Common Metrics to the Bottom Line

Knowing metrics such as floor space remaining, cabinets with the most open data and power ports, and budgeted power is a solid starting point for capacity management. Capacity forecast charts with historical trending can aid in creating a clearer picture as well. However, to fully utilize your existing capacity and make the case for purchasing more if necessary, you need more advanced analysis.

[clickToTweet tweet=”Herman Chan – What-if scenarios can take your capacity planning to the next level #datacenters” quote=”Herman Chan – What-if scenarios can take your capacity planning to next level #datacenters”]

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 you to understand the potential net impact of changes in your data center—specifically additions and decommissions—so you can quickly and accurately predict a future state and ascertain if you can postpone adding additional resources or if you need to purchase more.

The easiest and most useful way to look at what-if scenarios relative to capacity is on a per-project basis. For example, as a data center manager, you likely have multiple concurrent projects in the pipeline. You might have several consolidation projects on the way, projects to retire end-of-life assets, and new projects in the pipeline that you will need to reserve resources for. What-if analysis enables you to see the net effects and trade-offs of individual or groups of projects so you can:

  • Accurately determine if you need more resources and for which projects. Although you may want to put off costly capital expenditures for as long as you can, the results of the what-if analysis may show that you will fall short and actually will need to invest in additional resources. Knowing which projects you need the resources for also can help you time your expenditures accordingly.
  • Account for decommissions as well as additions in your capacity planning. Most capacity planning metrics focus only on additions, but doing so only creates an incomplete picture that doesn’t truly represent your actual capacity. Since your objective for what-if analysis is the net result of your projects’ impact on your capacity, it’s critical to include decommissions and the freeing up of resources as well.
  • Easily and effectively communicate your reasoning to management. As the data center manager, you need to conduct your data analysis at a level of abstraction where the insights are easy to grasp. When you explain your chosen course of action to management, you’ll need to give a bottom-line summary that clearly justifies your need to purchase or ability to defer. What-if charts can provide a visual analysis for you to present this information at a glance. You can’t do that if you’re wasting time manually combing through spreadsheets to find the salient data.

What-if scenarios can take your capacity planning to the next level so you can not only provision at the speed of IT but also justify your expenditures to your management team.

How DCIM Helps You Keep Pace with IT Demand

What-if analysis often is only one component of a comprehensive DCIM solution that can help you rapidly provision and plan capacity. You also need instrumentation—iPDUs with power and energy metering, power meters upstream at UPSs, branch circuits, and building feeds, and environmental sensors including temperature, humidity, airflow, and pressure—deployed throughout your data center.

Once this level of instrumentation is achieved, your DCIM software can collect and analyze the data so you can easily see the capacity levels of every component in the data center and identify where you have over-provisioned resources and introduced waste or under-provisioned resources and introduced risk. Trending this information over time with predictive analytics and what-if analysis helps you plan timely infrastructure deployments to support the rollout of dynamic IT architecture, applications, and services.

Bringing It All Together

While the metrics most commonly used to plan data center capacity can be helpful, they are inadequate to provide a complete picture of future capacity needs. What-if scenarios can bridge this gap by accounting for decommissions as well as additions so you can accurately determine if you need to make capital expenditures for new resources. Additionally, what-if analytic charts enable both you and your upper-level management to view the same data and understand at a glance the justification for data center capacity plans.

With what-if analysis done right, you can quickly derive the insights you need to quickly provision and manage resources for the needs of dynamic IT architectures. When used with DCIM software that collects and analyzes instrumentation data from smart devices and sensors in your data center, what-if analysis allows you to plan capacity more accurately and with less waste, ultimately boosting your ability to increase your team’s productivity and delight your customers and management team.

Herman Chan is President of Sunbird Software.  He can be reached at [email protected].

About the Author

Voices of the Industry

Our Voice of the Industry feature showcases guest articles on thought leadership from sponsors of Data Center Frontier. For more information, see our Voices of the Industry description and guidelines.

Sponsored Recommendations

Guide to Environmental Sustainability Metrics for Data Centers

Unlock the power of Environmental, Social, and Governance (ESG) reporting in the data center industry with our comprehensive guide, proposing 28 key metrics across five categories...

The AI Disruption: Challenges and Guidance for Data Center Design

From large training clusters to small edge inference servers, AI is becoming a larger percentage of data center workloads. Learn more.

A better approach to boost data center capacity – Supply capacity agreements

Explore a transformative approach to data center capacity planning with insights on supply capacity agreements, addressing the impact of COVID-19, the AI race, and the evolving...

How Modernizing Aging Data Center Infrastructure Improves Sustainability

Explore the path to improved sustainability in data centers by modernizing aging infrastructure, uncovering challenges, three effective approaches, and specific examples outlined...

NicoElNino / Shutterstock

How Can Data Center Managers Handle Explosive Growth Driven By AI and ML?

Marc Caiola, Vice President of Global Data Solutions at nVent, explains how leveraging the right cooling and power technologies can help data centers manage the growth of AI and...

White Papers

Get the full report

Decarbonized Resilience

Nov. 14, 2021
A new white paper from Enchanted Rock explores four alternatives to diesel backups to see which offer both resiliency and an economical way to meet climate goals.