About 450 students got real-world experience analyzing and developing visualizations of 3 million rows and 80 columns of big data – the kind that energy managers face in their work.

The data was provided by data-driven energy management systems (EMS) company GridPoint, which partnered with colleges throughout the United States for the fourth annual DataFest, with a total of 15 schools participating in five separate events.

The data points were generated from three years of hourly readings at 110 buildings and ranged from energy-consuming assets such as lighting panels and kitchen appliances to weather data points such as precipitation and temperature. The dataset real world analysis projects included the difficulties presented by missing data, inverted readings and other realities professional face daily.

The student teams analyzed the large, complex dataset to understand total building energy use, spot assets that consumed a large amount of energy, and analyze patterns and trends in energy consumption. Teams presented their findings to a panel of judges who selected top performers based on the level of sophistication, information and usefulness in their reports.

Some teams chose to focus on savings analysis while others showcased the market potential of an energy management system. Still others looked at the way weather events like Hurricane Irene impacted buildings and disrupted businesses.

The events were hosted by Duke University on March 21 and ended at the University of California, Los Angeles on May 4. Princeton University, Emory University; Five Colleges, a consortium composed of Amherst College, Hampshire College, Mount Holyoke College, Smith College and the University of Massachusetts Amherst also hosted events.