Vigilent is a pioneer in the Industrial IoT market. The company uses machine learning, wireless sensors, and prescriptive analytics to dynamically manage cooling in mission-critical environments. Vigilent has been in the business for over 10 years and has over 500 deployments across five continents worldwide. The company’s mission is to increase the profitability of its customers, and also to create a more sustainable planet by dramatically reducing CO2 emissions.
At present, Vigilent counts among its customers some of the world’s largest telecommunications, data center colocation, enterprise, government, and Internet organizations, including Avnet, Digital Realty, Hutchison, NTT, State of California, Spark, TELUS, and Verizon. The company has achieved global reach via direct sales, strategic OEM, and distribution partnerships.
About the Trailblazer
Dr. Clifford Federspiel, Founder and President/CTO of Vigilent had an early interest in controls and machine learning. Following his SMME and Ph.D. from the Massachusetts Institute of Technology, he joined the University of California, Berkeley, where he was a researcher at the Center for Environmental Design Research and the Electronics Research Laboratory. Dr. Federspiel launched Vigilent in 2004, building on his research combining AI and wireless networking to address cooling energy consumption. He has been a frequent lecturer, speaker at industry conferences, and contributor of peerreviewed papers and conference proceedings that address these subjects. More recently as a pioneer of real-world IoT systems, Dr. Federspiel contributes his perspective and next step thinking through his blog and published articles.
Transforming Business Needs
Data centers are digital factories at the center of the information economy, and critical to revenue generation and operational performance in most organizations. These digital factories require certain environmental conditions to be in place for the IT equipment to operate. Problems that lead to IT outages can have a multimillion-dollar impact within minutes. Vigilent helps ensure that these digital factories avoid environmental problems that cause outages.
“Our systems also significantly reduce the amount of energy consumption and carbon emissions, which translates to the bottom line. For example, NTT in Japan said Vigilent is saving them over $15M per year in energy costs. Vigilent also eliminates millions in CapEx that our customers would otherwise spend on cooling equipment that is not needed. When the Vigilent system is deployed, customers often find out they have plenty of cooling capacity and don’t need to buy any more equipment,” asserts Dr. Federspiel.
Controlled and Optimized Cooling
The Vigilent Dynamic Cooling Management System is a turnkey IoT system that automatically optimizes cooling and energy use in missioncritical facilities, a.k.a. ‘digital factories’ such as data centers and telecom exchanges. If an organization wants to use IoT to improve the performance and efficiency of their digital factories, Vigilent is there to help.
People running mission-critical facilities often face the challenge of constantly managing temperature and airflow. They typically avoid this challenge by running more cooling than they need to avoid overheating and potential outages. The problem is difficult because IT applications vary in intensity, and airflow is inherently complex since it is affected by floor layout and other factors. Facility managers try to manage this problem by over-cooling 24/7, but this can be risky and always results in wastage of money. The Vigilent System uses machine learning software to continuously match cooling with IT load at every spot across a data hall. Wireless temperature sensors are deployed throughout the facility, and wireless control modules are installed on each air conditioner. Through these sensors, Vigilent collects granular temperature data across the facility, understands the relationships between each air conditioner and how it influences cooling across the room and uses this knowledge to dynamically control the air conditioners and optimize cooling. The system is deployable in any existing data center in about two weeks.
Cut Cooling Costs
Many data center operators use the controls provided by air conditioning manufacturers to manage cooling. These controls are rudimentary and not designed to really optimize cooling. If they truly optimized cooling, the manufacturers would sell fewer air conditioners, which is not something they want to do.
Vigilent puts the customer in charge of their cooling vendors. They can pick whichever air conditioning vendor they want and compare how they perform. They don’t get locked into a control scheme that requires them to purchase cooling equipment from a particular vendor.
“It is also worth noting that there is one other company that uses machine learning to optimize cooling: Google”, states Dr. Federspiel.
He continues saying, “A few years ago, Google started to use Deep Mind to analyze and optimize cooling in its own data centers. It found that it could save around 40% of cooling costs by using machine learning to identify unexpected interactions, and taking steps to optimize its environment.”
Vigilent has done the same thing for more than a decade with similar, if not better, results. The main difference is that the company didn’t build it just for in-house use. Vigilent technology is available to everyone in the data center and telecom industries.
Robotic System is Indispensable
The team at Vigilent reveals that companies initially did not trust the technology and were unwilling to relinquish control of critical cooling equipment to a robotic system, despite pilot programs and undeniably consistent results. As the complexity of facilities has increased, the ability of humans to manage all elements of cooling equipment efficiently has decreased and stirred interest in automation. The spread of machine learning as an enabling technology for IoT has helped diminish concerns and pushed acceptance of technologies like the Vigilent System.
Vigilent expects to witness continued growth in the data center market. However, the company also focuses on other industrial applications that can potentially benefit from the use of its machine learning algorithms to optimize environmental controls in their operations, such as the retail, pharma, and indoor agriculture industries. Most cooling controls in these facilities are basic and suboptimal, similar to those in data centers. The company plans to explore these other vertical markets with strategic partners.