Ever wonder where the Internet lives? (No Ted Stevens, it is not in a “series of tubes.”)
Actually, it’s more like a series of servers, or computers, many of which are located in data center facilities. Some of these are on-premises, which means they’re owned and operated by an organization that provides the Internet-based content, service or application. Others are in the cloud, which is just an airy word for a facility that is owned and operated by a paid vendor, for example, Amazon Web Services or Google Cloud. Some are in colocation facilities, which means the equipment, bandwidth and storage space are rented out to clients.
Regardless of the model used, chances are, most of the Web activity you engage in for work or play on any given day entails your computer connecting to a remote server somewhere. This could be a few hundred miles away, or it could be across an ocean. What’s important to understand is that the Internet is entirely dependent upon these facilities. And while it’s easy for data center managers to get wrapped up in their day-to-day processes and forget just how important security really is, data centers are essentially the backbone of the Internet: Trying to keep Web services up and running without these facilities would be like trying to walk without a spine.
What are the cyberthreats to data centers?
There’s a long list of data center hazards that includes floods, fires and power outages, which have been caused by squirrels chewing on wires, birds flying into transformers and the occasional summer blackout. Every single one of them, plus many other possible threats, can result in downtime that cripples Web services.
But the risk of physical intrusion is just as significant for data center management. According to Data Center Knowledge contributor Jason Verge, thieves actually managed to break into a data center facility in Denmark by cutting a hole in a wall. Verge reported that the perpetrators stole basic equipment, including network cards. While this damage was about as minimal as they could have hoped for, it does raise some questions about how this was allowed to happen.
“How did thieves cut through a wall? How did they get in and out undetected? Why wasn’t security staff aware? What was stolen and how did that hurt the customer?” asked The Data Center Journal contributor Josh Moody. “Security should be multilayered and require multiple points of two-factor authentication along with biometric scanning at every colocation-room door.”
“In many cases, cyberattackers and thieves are after much more than network cards.”
In many cases, cyberattackers and thieves are after much more than network cards. Take the NSA’s data center in Utah: The massive 20-building super complex is home to what the NSA refers to as a “100,000 sq-ft mission critical Tier III data center.”
According to The Atlantic reporter Walter Kirn, the facility was being used, among other things, to house intelligence collected by the NSA. Given the amount of potentially sensitive data, which may or may not be pertinent to national security, it’s hardly surprising that there are an estimated 300 million attempted cyberattacks on the facility every day, according to The Hacker News. Still, that’s an awfully large number of attempted hacks, and if nothing else, it highlights how just how important cybersecurity is in government-operated data center facilities.
Likewise, data centers that house protected health information are ideal targets for hackers. The Identity Theft Resource Center estimated that there were hundreds of health care-related data breaches in 2015. While many of the more infamous incidents entailed elaborate virtual schemes, a cyberattack caused by a physical intrusion of a data storage facility is not outside the realm of possibility.
It might be time to change the locks
Given the undeniable significance of data center facilities as well as possible implications of a physical intrusion – and we don’t just mean downtime here – data center managers need to make sure physical security is as strong as ever. For starters, multifactor authentication at all possible entry points is absolutely essential.
This could entail the use of a one time password that is sent to a predetermined mobile device each time an employee taps an eID on a card reader. It could also mean fingerprint and retina-scanning technology, as suggested by Moody. The same goes for internal entry ways to sectors of a facility that might have limited access to a select few employees, as well as doorways in colocation facilities.
“Mobile device management creates unique identities for mobile devices.”
Stronger data center security also entails securing mobiles devices that may be in use for daily operations such as communication between staff and alert notifications. To this end, mobile device management creates unique mobile identities for devices. This makes them more reliable as authentication tokens in data centers.
Whether you’re running a massive, top-secret facility that houses NSA intelligence, you’re a health care provider that relies on its data center for access to patients’ medical records or you’re a world-leading cloud provider, slacking on authentication is just about the dumbest thing you can do.
Keep your organization’s data – or your customers’ data in the case of a cloud vendor – safe with smart, strong authentication.