February 3, 2016
In late January, the Mindsight blogging team sat down with Chris Rechtsteiner, the VP of Marketing for Server Central, to talk about the staggering scale and intricacies of the modern day data center. Chris has been working for Server Central for over two years, and served as a chief strategist at BlueLoop Concepts Inc. helping clients make data-informed decisions for their software and digital media.
Mindsight Blogging Team: Tell me a little bit about the Server Central data center.
Chris Rechtsteiner: Our flagship data center’s technical name is DFCH1, which stands for “DuPont Fabros Chicago 1.” It’s located in Elk Grove Village Illinois. The facility was built by DuPont Fabros, a real estate investment trust. They build massive, basically bullet-proof, data centers all over the world. One of the most sophisticated builders of data centers on the entire planet, and the center itself is managed by Server Central. This particular location is just under one half million square feet. Within that footprint there’s 275,000 square feet of raised floor—that’s essentially the location of where computing equipment network equipment can be located. The remainder of that space is storage and facilities that keep the site up and running.
MBT: And what would you say differentiates DFCH1 from other data centers?
CR: What makes this facility unique is that when Dupont builds a data center, they overbuild everything. There’s no clean or simple way to put it. They overbuild everything. People talk about N+1 or N+2 redundancies, right? Everything in [DFCH1]—whether it’s power, cooling, connectivity—everything is a minimum of N+1 ,and in most instances its N+2. What this means is that every single step of the way, there are so many failover capabilities that the likelihood of losing any one service component is really small.
Now, Server Central has been in business since 2000 and operating data centers since that point in time. You can architect any [application] to stay on line, but at the end of the day, the data center itself has to be there. It has to work. No matter how well architected the application may be, if you’re dependent on data centers that aren’t designed to always be online, it doesn’t matter.
We say this jokingly, but there’s really a lot of truth to it. If DFCH1 lost connectivity, lost power, and that facility was taken off line, the least of anybody’s concern is going to be the status of their application. It is going to take something so colossal for that to happen, and realistically in that scenario, people aren’t going to care [about data].
DFCH1 represents a culmination of years and years of data center development. This isn’t the first time anybody has ever done this. Every time one of these is built, the technology improves, the knowledge improves, and the foundation improves. Eventually, you reach a point—and DFCH1 is really that point—where at a redundancy level, you have everything covered. And looking forward, as the power density increases and the power requirements of the servers increase, the existing infrastructure to support it is already in place at DFCH1. The idea that you can support an increase in power density is really one of the hidden gems of this facility. It is so future-proof in that regard. It is really a fantastic foundation to build on.
MBT: And you mentioned that the DFCH1 is located in Elk Grove. What made that town an attractive location for this kind of facility over any other suburban community in the area?
CR: What makes Elk Grove so attractive is three things. One is its proximity to Chicago. It’s far enough away but its close enough. That’s really important. Server Central is a Chicago-based company, and DFCH1 is at a distance that can be managed easily from Chicago. Second, Elk Grove has an infrastructure—power, water, roads, all of that stuff— completely separate from the City of Chicago. For all intents and purposes, you have a completely different legal entity than the city.
Finally, Elk Grove is easily accessible from multiple freeways and from both airports. We’re 8 miles from O’Hare and 27 miles from Midway. DuPont Fabros is able to maintain the constant connection of those roads by providing snow plows and road maintenance support for the village of Elk Grove, when necessary, to make sure that anyone coming off of those freeways can make it to the data center regardless of weather. It’s an unbelievable physical location with O’Hare right nearby. You have tremendous power diversity. You have tremendous fuel access because of the airport. You have all the critical infrastructure that goes into building a major global transportation hub right at your disposal.
MBT: Still, Chicagoland has its fair share of flooding and tornadoes. Should there be some kind of disaster or natural event, what protections are in place to safeguard Server Central client data?
CR: Inside the data center footprint, there are four completely diverse power feeds. So, the base power coming into the facility is coming in from four different locations and from four different sources. Plus, DFCH1 is equipped with thirty-three 2.25 KW diesel powered generators. That is massive. That is enough to provide unlimited power for the data center should anything ever happen. There’s also an on-site fuel reserve and fuel contracts that will allow those reserves to be immediately refueled.
Realistically from a data center perspective, the disasters you’re going to have to deal with are very rarely going to be natural disasters. Sure you’re going to have hurricanes and things are going to flood like what happened in New York, but that is so rare. Typically, the disasters you’re going to have to deal with are man-made or natural events, like cutting fiber when people are putting in new streets. That kind of stuff happens all the time.
MBT: What are some of the more common natural events that DFCH1 faces, and what measures are taken to mitigate their threats?
CR: One of the key natural events that people don’t talk about is power consistency or power quality. Now, in any major city in the United States, the minute the temperatures get hot in the summer the power grid becomes absolutely overworked, and you essentially go into what are the equivalent of brownout periods. Some municipalities will talk about this very openly, and they will say “we are going to do rolling brownouts or we’re going to do rolling blackouts as ways to control power.” Then they roll those through. Others don’t talk about it, but do it.
What exists in this facility are Rotary UPS systems. Basically, this is a power conditioning unit. In the event of a blackout, there are these rotary UPS systems that can work in a very quick fashion should there be any power issues prior to a generator coming online. In essence, there’s no power downtime. The computer won’t see it. The server won’t see a power failure. It will all switch over so quickly that there’s no powering down and powering back up.
But, what these UPS systems also do is they condition the power. They’re constantly watching, and I mean every single second of every day. Every millisecond, they’re watching the power. If the power starts to get of poor quality, it can flip over to a generator, or it can use the power stored in the UPS to smooth it out. With the UPS systems, we can say, “You know what? We’re going to provide a clean level of power all the way through regardless of what’s going on outside.”
Remember, those sorts of natural events like brownouts put a huge amount of wear and tear on servers and infrastructure equipment from constantly getting power spikes. By smoothing that out, you get a much longer life of the devices, but you also have a much more consistent delivery of service. Everything is going to respond consistently. Applications are going to respond. Storage appliances are going to respond. Nothing is going to freak out, because it’s not clean power, and that happens a lot. Little things like these natural events happen all the time, and having those capabilities in place are important.
MBT: On the flip side of data protection, can you describe just some of the security features that would prevent someone from just walking in and tampering with data at Server Central?
CR: It’s a difficult place to get into. You can’t even get into the building without being badged, and all of those badges are done on background checks. It’s pretty significant.
But security really starts with a 24/7 security staff. That’s where it begins and ends. Our people are there all the time specifically for security. The entire facility, everything, is under closed circuit video surveillance 24/7/365, so everything is seen. All the access points are dual factor authentication with biometric and proximity scanners. This means that you might be able to have my badge, but you’re still not getting anywhere. I could hand it to you and say, “go,” and you’re still stuck. You’re stuck right where you are at that point in time.
That’s really important, because that dual factor is the foundation of our security. If you can get past step one, you can’t get past step 2. Each of these individual transition points from one area of DFCH1 to another, are set up with [security] traps. They work like a pseudo-DMZ. You can get literally caught between two places, and you can’t get in or out. All of these elements, when you start to stack each of them up, represent in totality a very complete set of physical security that’s going to prevent anyone from getting into the building. Now, should someone get in, and they got onto a data center floor, everything is still locked and under dual factor authentication. You can’t go around opening cabinets or cages. Even if you get in, what are you going to do?
At that point, it’s under our surveillance as well, because we have cameras on every single cabinet, every single cage in the data center pods. If we see people we don’t know, someone is going to check it out.
MBT: What would you say is the value of understanding the data center side of the technology industry?
CR: There’s an old adage, “If you really want to drive your car, learn how it works.” There’s a lot of truth in that. Once you know how it works, you drive it differently, you care for it differently. The data center is very similar. Once you know how it works, you have a new level of knowledge. You can build applications in a completely different way.
MBT: In a rapidly expanding tech industry, with new startups being founded every day, where do you see the value of a data center as thorough and iron-clad as DFCH1?
CR: The one thing I would say is that you spend a tremendous amount of time, money, and effort to architect applications that will run and meet your service levels, whatever they are. Then people turn around, take those applications, and say, “I don’t want to put it in that data center. I want to put it in this one because this one’s really cheap.” All too often, we see people who have over-invested in the application development and under-invested in the data center side, because they don’t know enough about it. It just seems like and unnecessary cost. Well, if it’s going to cost $1000 a month here and $500 a month there, put it where it’s $500 a month, because it’s half, right? That’s a huge issue. You have to understand where your application’s needs are.
No matter what the cost of investing up front to make sure the application is in a place designed to support it, the investment is going to be cheaper than moving from someplace that couldn’t maintain the service levels. Remember, at that point, it’s not a planned move. It’s a, “the sky is falling on me, get me out of here now” move. It happens all the time. The reality is that the cost associated with that is going to exceed anything you would invest up front in planning and taking care of [the application].
Learn More About Server Central and DFCH1
DFCH1 truly is an amazing data center. The massive scale of the facility alone is enough to instill awe, and companies throughout Chicago and beyond can take advantage of this incredible infrastructure. Contact us today to learn what Mindsight and Server Central can do to protect your data.
Mindsight, a Chicago IT consultancy and services provider, is an extension of your team. Our culture is built on transparency and trust, and our team is made up of extraordinary people – the kinds of people you would hire. We’ve always prided ourselves in delivering the full spectrum of IT services and solutions, from design and implementation to support and management. Our highly-certified engineers and process-oriented excellence have certainly been key to our success. But what really sets us apart is our straightforward and honest approach to every conversation, whether it is for a local business or global enterprise. Our customers rely on our thought leadership, responsiveness, and dedication to solving their toughest technology challenges.
For Further Reading
All over the country there are data center facilities out there trying to do everything possible to protect the essential business applications of their clients. Learn about just a few of the security measures standard across all data centers in our blog, “2 Key Ways to Protect Your Data: Colocation and Cloud Solutions.”
About the Interviewee: Chris Rechtsteiner
Chris Rechsteiner joined Server Central in 2013 and is responsible for the company’s corporate and product marketing operations. You can find him snowboarding, mountain biking, or coaching basketball.