Balancing Dollars & Demand

• By Charlene O’Hanlon
• 09/01/07

How do IT administrators prod their IT budgets to meet institutional goals? Four intrepid campus tech leaders share their concerns, woes, and triumphs. Listen in.

When it comes to building and maintaining a cutting-edge infrastructure on a limited budget, nobody knows more about the pain points, challenges, and workarounds than the IT professionals tasked with overseeing campus IT. In a recent roundtable discussion with Campus Technology contributing writer Charlene O'Hanlon, four IT administrators—James Stoner, director of scientific computing at the University of Denver (CO); Steve Engorn, CIO of Villa Julie College (MD); Ryan Laus, network manager at Central Michigan University; and Stan Gatewood, CISO of The University of Georgia—all shared their experiences and advice for peers.

The schools they represent are diverse: The University of Denver has around 5,000 students, undergrad and graduate, and nearly 1,000 staff and faculty. It has a fairly large IT network infrastructure, and a campus laptop initiative that is pivotal to maintaining a conservative IT budget (students must bring their own laptops to the university). Villa Julie College, with two campuses just outside of Baltimore, has a full-time enrollment of 2,500, with another 400 or 500 part-time, evening, and online students. The college's current focus is academic, not research. Central Michigan University lies in the heart of Michigan, in the small city of Mount Pleasant. The university boasts 20,000 students on campus at the main site, another 8,000 students at offcampus sites, and around 50 sites worldwide. The University of Georgia, with a 600-plus-acre main campus, is located in Athens, and is an undergraduate, graduate, and doctoral R1 institution, with heavy emphasis on research. There are approximately 35,000 students and 10,000 faculty and staff, and the institution has a presence in all 160-plus counties in the state. There are also satellite campuses in South America, China, Italy, and various other sites in Europe.

Campus Technology: Please tell us about your institutions' core IT focuses.

James Stoner: Aside from our laptop initiative, we have other unique aspects with respect to our infrastructure. Our research infrastructure and scientific research are more important than any other aspect. We do a lot of genomics and life sciences, so there's always that instrumentation aspect, for instance. Most of our really high-end IT infrastructure has been accomplished through collaboration and building relationships with external companies. That's been our key in maintaining our budget constraints, at least in the sciences. The value-add there is exponential, where we can use our external technical support to help maintain our internal IT structure.

Steve Engorn: For the most part, our setting in both campuses is very rural, and one is in a valley. They're about six miles apart. The newer campus in Owings Mills houses all of our residents; our older campus in Stevenson has a number of academic buildings. Our challenge is to have both campuses act as one, even though they're six miles apart.

We typically have two phone PBXs so, based upon how our network has been structured, it seems you're on one campus no matter where you are. We're running voice over IP on both of our campuses to alleviate wiring costs. And we've done a lot of construction in a very short period of time, and we're still doing more.

Ryan Laus: We have a central IT department, but also have a decentralized part of IT—different departmental techs for the colleges. Central IT handles things like the e-mail and learning systems such as Blackboard, networking, and cabling. The colleges, our distributed folks, handle some of the specialized faculty and staff needs. We are very research-attentive. We do undergraduate, graduate, and doctoral, and we're trying to make more of a name for ourselves in the research area.

Stan Gatewood: We're strong in research in agriculture, veterinary medicine, and biology. We have a fairly decentralized IT infrastructure, in that we do have a central organization that handles our backbone, our telephone systems, and our wiring to the buildings, but the schools—14 schools within our university— handle their own IT to the desktop.

CT: With the exception of Villa Julie, it sounds like many of you are dealing with heavy-duty research needs at your campuses. How does that affect your IT budget? How do you go about weighing what is needed for research against what is needed for operations for the college in general, and what your students need?

Gatewood: We have several research alliances that we're a member of, so we do get money from those alliances, plus some of our large grants, et cetera. But research is headed up by the Office of the Vice President for Research. He does have a budget as well as being a member of the Georgia Research Alliance and a few other associations. So his department gets a piece of the overall University of Georgia pie as far as budget is concerned, and it carves out the necessary funding or monies to work with its research clusters and its overall IT.

CT: How is that budget faring?

Gatewood: The overall IT budget health probably needs a booster shot of B12, just like everyone else's budget. Monies are getting slimmer and slimmer, and we have to spread it thinner and thinner. There's not a whole bunch of research money out there, but there's plenty of research to be done. Most of it is state or federally funded, but we must also measure in or count in such things as the overall cost of IT—which also has skyrocketed— and the maintenance, upgrading, and rescaling of our systems. We're definitely doing more with less than we ever have before.

Stoner: We have very much the same system as Stan has over at Georgia. Our vice provost for graduate research has a budget and, in our universe, it's based on the amount of funding the university has, plus a small chunk from the university added to that. So his budget directly reflects the amount of research the university is doing, which is interesting when you think about it. Somewhere between 5 and maybe 15 percent is the funding ratio for grant applications right now—substantially less than it's been across the years. What we've been doing is taking the small one- or two-lab IT infrastructures that were out there, and migrating them back to a more central location, so that the funds get used in a more standardized way, from the support to the hardware sides of things.

CT: So you're using a cluster computing environment? An IT ‘pool' from which different research arms or different departments can access their IT power? That seems to be an excellent way for the university to save some money and allocate resources more evenly.

Stoner: That's exactly what we've done. We've built a centralized IT function for things like business functions, registrar, et cetera. That's all handled by the university technology services. My world is very, very specific in the sciences, so we're one of the distributed sides of the overall picture.

Laus: We have similar issues. One of our biggest concerns has been to make sure that we have sufficient bandwidth out to the internet so that our professors and [students] can get the research done and move the data around when they need it. So, one of our biggest initiatives has been to team up with other institutions across the state, to make sure we have the sufficient fiber, bandwidth, and redundancy we need to keep the institution running.

CT: As a private college, I'm assuming that Villa Julie has a somewhat different budget process. Steve, how would you describe the health of your IT budget?

Engorn: Our college is continuing to grow. Seven years ago we had 1,200 students; now we're close to 2,500. Our goal is 3,000, so our challenge is going to be to maintain and enhance our infrastructure. The college is very willing to support that, because of the increased enrollment. I have not seen any decreases in my budget, and I have a strong relationship with the president, the cabinet, and the comptroller. We sit down as a team and decide what we need to fund for the coming year, based on the goals and objectives, and it seems to work. I'm not given carte blanche, so we have to do a lot of digging to make sure we have the right solution, and make sure it's the right price. But we also see some savings because of the statewide consortium that was actually started by the university systems of Maryland, where we now are able to buy at a very, very reduced rate. Even as a private institution, I can buy off of the state contract, which gives me some leverage. That doesn't mean that [the state] always gets the best price, but it gives me a starting point when I'm negotiating with vendors.

Gatewood: We have a similar situation. Basically, we negotiate statewide contracts, so any state agency under the governor of the state of Georgia can purchase out of these contracts. We have a tremendous amount of buying power; we just need to make sure that everybody knows about it, because we get even deeper discounts when everyone comes together at the same time to make major PC purchases, for instance.

CT: Are any of you doing what the University of Denver has done with its laptop initiative? Are you finding such programs help keep your costs down?

Engorn: We're looking at that now, and plan to pilot a laptop initiative in the fall. We're actually looking at the University of Wisconsin-Stout model, where they give every incoming freshman a laptop. They have the Apple component, a standard laptop, and a higher-end laptop, based on what each department is requiring. We believe we're going to see some significant savings overall in general maintenance of labs and so forth.

Stoner: Our students are responsible for purchasing their own laptops, but we have a list of specifications, and preferred vendors we work with that will do imaging and so forth.

CT: If you are not providing the laptops, then how are you handling help desk requests, for instance? Are you offering that as a service to the students since you are requiring them to bring a laptop?

Stoner: We do have a help desk function that's very well-versed at handling these things. We will provide the technology support if students have a disaster, and we also have deep-discounted student pricing for a lot of the software packages we require, so they're actually affordable.

Laus:We take the same approach both in terms of requiring students to bring their own machines, as well as having our help desk support them. They can call into the help desk for pretty much any computer needs they have. We also provide them with an antivirus package for free if they don't have one, or if they want to switch to ours, since they won't have to pay for the updates on it. We find that works really well to help keep the cost of support down. We also have a PC repair division that is a for-charge service.

CT: Has going with a wired versus a wireless network—or a hybrid of the two—had any impact on IT expenditures?

Stoner: We use wireless primarily in places where it's become either untenable or too expensive to install wired outside of the common areas. We use it as an adjunct to wired or wireless, if we need to. But the goal is always to come back to a copper port. Since we have some old buildings on campus, wireless is not tenable in some cases.

Gatewood: We're hybrid: We have wired and wireless, but wireless is pretty much ubiquitous. One of the things that we did almost immediately was to [set up] our own wireless standards. If schools and institutions and their administrative offices or units are left to put up their own wireless, invariably it's going to be screwed up, down, and unsecured, and maybe not in that order. So with our wireless standard, we have pretty much slid over to the driver's seat, and the central IT organization is handling all of the wired and wireless. The support staff can now focus and train on a given set of hardware, software, and application issues, not trying to be everything to everyone.

As far as budget is concerned, we found that the wireless access points and the wireless equipment have dropped so much in price and cost, it's really not affecting our overall budgets at all. You can buy a wireless access point for under $50, and there is no approval process needed for something so inexpensive. Still, when it comes to total cost of ownership, wireless is not all it's cracked up to be. When you secure wireless versus wired, the cost jumps exponentially.

Engorn: We don't quite have the wireless mesh in place, but we will look for any rogue wireless access points and we will disable those to keep the networks secure, decrease the number of security risks, and ensure adequate bandwidth for those who are supposed to have access. We have the software capable of doing that. At the same time, when we build any new facility, we're going to make sure that it's wired and wireless. The issue for us has primarily been—in a graphics design classroom, for instance—if I'm connecting wireless, there just isn't enough bandwidth to transmit large projects to and from the server. Now we're designing classrooms that have the capability of both. So with this laptop initiative, we probably will have wireless in the classrooms, but at the same time, have wired jacks just in case. The only thing we're going to require for laptop users is an extra battery. But I'm not going to be spending money building rooms that have many, many outlets, and so forth.

CT: Are schools that moved to wirelessonly facilities now finding that they have to go back and install ‘wired' too, after the fact?

Laus: In our College of Health Professions building we actually had to go through and install additional wired jacks in all of the classrooms because the wireless did not provide a stable enough connection. I think there are quite a few other schools that have experienced this as well, especially in the residence halls; students often expect the wireless network to perform just as fast as the wired network, but because of the nature of wireless—all traffic broadcasted— this is just not possible.

CT: Moving along, I'm wondering about corporate sponsorships and partnerships, and how those affect your IT budgets, be they research dollars or just overall IT budgets. How much does the corporate arena play into your IT budget considerations?

Gatewood: I can't speak for the institution here, but I think this is what they're thinking: We'd rather not factor in the corporate partnerships or sponsorships because you can't depend on them. What is really sexy and cool for corporations today is to partner with big research institutions, but what will it be tomorrow? A CEO may change overnight and decide that partnering with research institutions was a good idea but now is passé, and so will let it go. Then we're left with a humongous infrastructure we've built that depends upon their money and their resources, and the company leaves us high and dry. Therefore, we depend on our own funding and then the state funding, and other grants and things like that. We think it's a great idea to partner and we do have some very large corporate partners who donate hardware and software. But as an overall business decision, you really can't depend upon that. When and where we can, we handle our own budgetary needs. But, again, we're open to, of course, any collaborative effort, any partnerships that are out there.

Engorn: We'll build our budget on what the college can afford. Then if we end up getting a corporate sponsorship for a specific project, I may be able to buy some additional equipment, services, or software. For example, we had two distance learning rooms donated to us by Verizon. Otherwise, we may not have put them in. But it was all based upon us moving forward with an online distance learning nursing program, so there was a specific need; the vendor stepped up to the plate and put the rooms in. We're doing the same thing with a few other colleges in the area, and businesses that have put up some funds for specific projects. We will then have to incorporate the maintenance of the equipment or the software in the future, but the initial outlay was something that was just an added bonus, which we would gladly take. At the same time, we don't rely on that for our day-to-day, or even five-year plan.

Gatewood: In the past, I created a security operations center that had vendors' [donated] products in there. Yet, when the time came around to renew licenses and do any special professional service work, it was almost six to seven times what it would have cost us to purchase it ourselves. The bottom line is, we are very conscious of the fact that [a company] may have given us something, but somewhere down the road [we know that] somebody's got to pay for it.

Stoner: We have the same mechanism. We'll be more than happy to involve large corporations and such in the beginning of our planning processes. Mostly, though, we try to stay away from bringing in corporations to sponsor software licenses, mostly because of what Stan spoke about. If it's hardware, we're a little bit more willing to speak with them. We'll never say no, but it's much easier from both a budgeting and planning perspective if we're dealing with hardware as opposed to software licensing. Having said that, one of our biggest donors is a major software facility we've entered into a long-term, 10-year contract with. They've been a big player [in their market space], and they'll always remain a big player. But I cross my fingers when I say that; there's a lot of luck and serendipity involved when you deal with large corporations.

CT: Let's talk about the cost of maintaining the IT infrastructure. How do you factor in the human element and budget for it? Is it becoming more difficult to manage each year, and hit goals?

Engorn: My concern—and I think many institutions would agree—is that with a relatively small staff, you usually get a lot done. So, when upper management is looking at your accomplishments and project completions for the year, they say, "Well, you did X amount of work this year and you got it all done." But I'm asking for more staff because I'm actually working my great staff way too many hours to get the job done, and they're not going to be able to continue at this pace. I'm trying to communicate to the administration that I really need people in certain areas just in order to maintain. And as the institution is growing, individuals lose sight of the fact that you just can't grow an institution, at least at our pace, and not increase the IT staff.

Gatewood: We're running into another issue, too: We have to be very conscious of our compensation packages because we recruit people, send them to school, and allow them to dabble with all this high-tech stuff, and the next thing we know, corporate America comes along and sweeps an entire network division, or systems administrators, or network administrators. So we're in the process of taking a very serious look at the job descriptions and the role and scope statements and things like that, because recruiting [good people] is one thing, but retaining them is very important as well. Yes, we do factor in that we're going to have to give bonuses and merit raises, attract them to begin with, and then bring them in at a salary comparable to [that of the private sector in] Atlanta, which historically pays more than the state, or the state system or state schools.

Laus: At Central Michigan, we're severely understaffed. We even had a survey conducted a couple of years back and found that we're running at about a third of the staff we should be carrying for an institution of our size. The networking group, for instance, is a staff of about four, when we should be up somewhere around 11. So staffing has always been an issue for us, but it's the same problem: We do it fairly well, and even though our network continues to grow, we're just expected to keep up with it and continue doing the good job that we're doing. Some of that is finding pieces of technology that can help you manage or make your jobs easier, and free up some time to do more of the core networking things that we need to do.

CT: So what advice can you offer a peer who is looking to create a firstclass IT network with a limited budget, and limited manpower?

Stoner: My best advice is: Do your homework. A lot of us don't do enough homework, and there's always more homework you should do. Also, seek out best practices; that's one of the things that we've done. At the end of the day, good planning produces great results. If that means putting off the project for six to 12 months, you basically need to do that.

Engorn: My advice would be: Collaborate. Talk to other people at other institutions who may have done the same thing, or may be doing the same thing. We're not doing anything that is all that much of a secret. I meet monthly with all the local CIOs of both public and private institutions and we talk about everything. It's like a therapy session in many ways: If one of us has a question about our direction, there's nothing to hide. If someone's done something, we get the benefit of what they did right and what they did wrong. So to me, the collaboration with my peers is crucial.

Laus: Preparation or prior planning prevents you-know-what. Look at your options; look at what you really need to do. If it doesn't fit with the technology that you have, hold off on that project until you have all your ducks in a row. Another thing we try to do is standardize on things like our network equipment and some of our network setup. Because we have such a small staff, we don't have the time or the ability to learn a bunch of switch operating systems or anything else that we need to do with different vendors and different software, so we try to standardize on equipment and network setups and just try to replicate that across campus so that we can keep our scope very limited and be good at what we're trying to do.

Gatewood: My advice is twofold: First, strategic planning is essential, because if you don't plan, you will end up somewhere, and it may not be where you had hoped. And second, you need to network. Get out and talk to people of like-size institutions, maybe even some that are larger or smaller. Read reports and work with people, because none of our networks are islands: If we all connect to the internet, we're all in it together. I'll bet you we all have the same issues with networks, systems administration, and politics, not to mention the budgets we're talking about here today. Just get together with people and share. You need to learn from them.

::WEBEXTRAS ::
On-Demand Webinar: Funding Your Campus Communications Systems
Start Small: Wondering why you just can't attract those grant dollars for your tech initiatives? Try thinking ‘mini' before ‘maxi.'
Data-Driven Decision-Making: Dare to Share! Smaller institutions of higher ed are discovering that sharing software, data management, and expertise is key to DDD success.