The Green Campus

IT Meets BAS

• By Rama Ramaswami
• 05/01/08

Universities have buildings and plants with systems which, in a green world, need to be monitored cohesively via building automation systems, or BAS. Now it's critical that those systems are integrated with IT.

MOST GREEN PROJECTS are relatively easy to carry out in a university setting, but IT systems present unique challenges. In this first of our new series on The Green Campus, CT speaks with Denis Du Bois, a specialist on green energy and technology, editor of Energy Priorities magazine, blog contributor for the Seattle Post-Intelligencer, host of the Energy Minute Podcast series, and member of the board of the MIT Enterprise Forum of the Northwest.

Campus Technology: What does green information technology mean, in the context of higher education institutions? What should university CIOs, for example, be thinking about when they plan the greening of campus data centers?
DuBois: University CIOs can have an important impact on green information technology. The question is whether they want to lead or follow trends. They should think of computing in general as part of a supply chain: Components need to be recycled, there are energy inputs, the equipment takes up space that has to be cooled, and workers need access to the equipment. All of these have environmental influences. Within the data center, CIOs need to think of these broader influences, although they certainly should think about tactical things like new processes and new servers. Pretty soon everybody's favorite server will be available in green.

Building a green data center or retrofitting an existing facility is a huge undertaking. What are the major factors to consider?
I like to think of data centers as three concentric circles of influence: At the center is the data center itself, where you can set up more energy-efficient servers and virtualize them to save space. In virtualization- the term comes from the mainframe world- one machine can run different platforms. Thus, the new system "mimics" the older one. You don't need multiple pieces of hardware for different environments. The second circle of the data center consists of users such as students and researchers. Here you can have thin-client workstations that use less energy, and with network power management systems, a single person can manage every PC in the system and shut it down when it is unused. In the third circle are the building systems, which are on separate protocols. IT generally doesn't pay a lot of attention to them. But facilities have fairly sophisticated operating systems, and there comes a point where building operations can really leverage the IT network to ensure maximum energy efficiency. Web services can let systems talk to each other. That's the direction in which IT is going.

How does the integration of IT and building systems apply to universities?
There are many opportunities for it, particularly on large campuses. Universities have lots of buildings, cogeneration plants, and so on, and managing that portfolio is difficult. The systems are diverse and operations is separate from administration. So, the data in each system need to be normalized so that someone can look at the entire portfolio of systems at the same time. Following that reasoning, building automation systems-or BAS- should be integrated into the IT network. This involves taking all these separate networks-for example, lighting, or ventilation in the labs-and integrating them. It's a different way of thinking about IT's role.

It's also a different way of thinking about the role of building automation.
Absolutely. Take demand response technology, for example, where customers lower their electricity needs in response to dynamic energy prices, or when a utility requests curtailment. The newest systems are automated demand response systems, in which commercial facilities can automatically shed load as needed. The next realm of thinking is when the utilities do this on their own, without the customer's involvement. But this requires a lot of technology, a lot of security, and a lot of thinking on the part of IT and building management, to prevent hacking or abuse. IT has enjoyed the safety of isolation for a long time; prior to the advent of the internet, you needed to have a key to physically enter the building before you could damage anything. That is no longer the case. So, BAS vendors need to make assurances to IT about security, and to ensure that IT believes them.

What are some other challenges to integrating IT and BAS?
Seventy percent of all data centers are constrained for power, cooling, or both. Many schools haven't moved to organize their data centers. They typically are shoehorned into buildings or scattered around campus, and are hard to maintain. If they've got the air conditioning going, the classrooms become too cold, and it's a strain on the electricity and cooling systems. By contrast, building automation is a mature technology-it's not broken, and it has a specific function. But IT is not used to handling building automation functions. Universities are not sure how much of a setback they'll have if they integrate the two, and I can understand why: What if they can't get things to work? In the labs, for example, climatic conditions are important. You don't want heating or cooling systems not working for a long time. But until the data center can take on the responsibility for managing a fire alarm, IT doesn't know the meaning of "mission-critical." That certainly raises the bar for IT.

Universities have a lot of old buildings. Will their systems pose a problem for integration with IT?
Some of the legacy systems can't be integrated, but many building automation systems have open standards. Only the oldest are locked out, and even those sometimes can be converted. They have open standards protocols for the most part, and the technology is here to tie them to the IT network. It doesn't require new inventions or technology, and there are a number of vendors of this type of software.

How much does integration cost? How can universities get started?
The cost will vary a lot, but the return is energy savings, as well as the opportunity to participate in demand response. I believe universities are great places to implement some of these new ideas. They have large data centers, large computer networks, large campuses, complex occupancy patterns, cogeneration plants in some cases- all of these lend themselves to energy-saving experiments. The campus itself can be a lab. Students can be involved in these projects and that, in turn, will lead to the adoption of the practices by the private sector. Building automation is a difficult technology, so you find out what you need, you prioritize, you allocate resources. That's an ideal class project! The IT department also can be involved-for example, by modeling buildings to LEED standards before they're built. They're all software functions: You get the specs on a building and what components you need, you fine-tune it before you build it, you put systems in it to monitor building performance constantly, and you identify inefficiencies before you get the energy bill. There's even newer technology out there to help you determine which parts of the data center are coolest and which computers to move there. It's all software. You see? It's what IT lives and breathes.

::WEBEXTRAS ::
The Changing Face of Auxiliary Services As new needs crop up on campus, Auxiliary Services finds solutions for mobile transactions, mountains of junk mail, even green construction.

Where Green and IT Meet