There’s more to campus Wi-Fi management than device numbers.
GUEST COLUMN | by Perry Correll
Campus IT administrators are no stranger to the challenges of delivering a quality Wi-Fi experience for students and administrators. They often see high device density as the thorniest problem, and rightfully so, but it’s important to not overlook other common roadblocks. It’s not just laptops connecting to the network anymore; these days we have smartphones, tablets, wearables and gaming devices all clamoring for a share of bandwidth – and who knows what popular new device will arrive on the scene next? To that end, administrators need to plan, monitor and manage their environments to address both
I often get asked how many devices an individual radio, access point (AP) or even an entire network can support. The problem is, you shouldn’t even be asking this question.
expected and unexpected growth. In particular, they must consider four key factors: client types, available spectrum, number of devices and application demands, as each of these variables has a strong impact on overall network capacity.
Here’s the bottom-line: Not all clients are created equal. For instance, students use laptops in a significantly different way than they use an Apple Watch. Likewise, each distinct Wi-Fi client type employs different Wi-Fi capabilities. Laptops represent the ideal for a wireless product, boasting big batteries and large antennas, which enable them to connect to a weaker signal. By contrast, smaller and less expensive gadgets demand a stronger RF signal to perform. For campus IT to support all users on all devices, including those devices which are emerging as part of the Internet of Things (IoT), administrators will have to design their networks in a way that provides even the weakest clients with robust Wi-Fi service.
And as you plan your Wi-Fi infrastructure, beware this common marketing ploy – just because a product says 802.11n or 802.11ac on the package doesn’t mean it actually supports the highest possible data rates. For example, .11ac Wave 1 data rates currently support 1.3 Gbps, but the majority of consumer and handheld devices such as tablets and smartphones can only provide data rates at a fraction of that speed.
Be Strategic With Your Spectrum
Wi-Fi operates in two radio frequency bands: 2.4 GHz and 5 GHz. Until 2009, 2.4 GHz was the primary band for the vast majority of devices. Today’s mobile devices now come in two versions – either they operate on 2.4 GHz only, or can use both 2.4 GHz and 5 GHz. As a result, 2.4 GHz has become the “lowest common denominator” in Wi-Fi coverage, and this spectrum may always need to be supported at some level. But with more and more new devices supporting the 5 GHz band, IT teams should make the 5GHz spectrum their primary focus when it comes time to architect their campus Wi-Fi networks. The transition to 5GHz is an essential step in the move to a mobile-dominant user landscape, as this band provides eight times more channels, resulting in far more bandwidth and capacity. But don’t give up on the 2.4 GHz bands altogether. The 2.4GHz spectrum still offers three useful channels, a kind of backup option that could perhaps be best utilized for something like a guest network.
Getting A Handle on Density
I often get asked how many devices an individual radio, access point (AP) or even an entire network can support. The problem is, you shouldn’t even be asking this question. Instead, the right question revolves around how many devices you can effectively support on the network and still ensure a quality user experience.
Every Wi-Fi vendor works off of a specification sheet, which typically claims support for 64, 128 or even 256 clients per radio, translating to 512 clients per AP at the upper end of the range. But is it actually possible to achieve this level of coverage? Sure, but let’s consider the resulting user experience.
Wi-Fi remains a shared media and each radio has a finite amount of bandwidth available for each client. The reality is that the number of users effectively supported in any environment will vary considerably by use case. For instance, a classroom could be designed to support only 25 simultaneous users, while a cafeteria might need to support up to 40. Meanwhile, a larger venue like a stadium could possibly provide coverage to more than 50 Wi-Fi users depending on the anticipated demand. It all depends on the applications, which are most likely to be used in each context, and the quality of user experience that is desired for the location. Wi-Fi users in a classroom will most likely need more bandwidth per device to support data-intensive tasks like streaming a video lecture, while fans at a sports game can probably make do with more minimal support for needs like texting and chat.
It’s also important to understand real-world limitations. We often hear that a new Wi-Fi technology like 11n or 11ac will support more clients per radio. For example, if 11n (300 Mbps) can effectively support 30 users per radio, then 11ac (1.3 Gbps) should be able to support four times as many clients, right? Not quite. This might work in a white paper, but in practice, it doesn’t represent a real-world setting where client speeds and capabilities tend to vary considerably. Even a few low speed clients on your 11ac network can substantially weaken overall performance for all users.
A good Wi-Fi engineer should be able to design a network that can handle almost any number of devices. But even the most talented planner will find it impossible to create a network that supports all possible bandwidth demands, especially when considering the new educational applications that are being introduced to classrooms worldwide. That’s where traffic management comes in.
In a high school or university campus, there’s no argument that some applications will carry far more value than others. For a teacher completing a remote lecture, a crystal-clear videoconference will be crucial. Meanwhile, the need for hundreds of students to sync files with their iCloud accounts is simply not as essential or time sensitive. Traffic management enables network administrators to apply policies that optimize overall performance by prioritizing, limiting and even blocking specific programs or classes of applications. When this strategy is implemented, a high density of clients can be supported through managing how much bandwidth is used by any one application.
There are a few fundamentals to effective network traffic control – first, find out which programs are eating up all of your bandwidth; then set policies based on the value of each application. The tools available to you include prioritization, imposing bandwidth limits, specifying time of day access or even blocking specific apps altogether.
The truth is that there is no magic bullet that will ensure a totally flawless campus network. There is no ‘perfect’ technology or ‘perfect’ design that will address all your challenges. IT teams must appreciate the fact that Wi-Fi is a shared technology with multiple device types, capabilities and use cases all competing for the same bandwidth. Continual monitoring of your network to detect actual client performance levels is a never-ending process, but it’s your best shot for providing a high quality, reliable user experience for everyone who needs it.
Perry Correll is principal technologist for Xirrus, a Wi-Fi technology company based in Thousand Oaks, Calif.