I frequently hear the question, “If everything is in the Cloud and we now have 10 or 20 times the bandwidth we used to have, do we still need the complexity of quality of service?” It’s a good question...
The answer can be determined by another question: Does the performance of any applications you use impact your business? If the answer is yes, then not having (or considering) QoS may put your business at risk.
Often though, the answer is genuinely no. For example, if an email takes a second or two longer or loses a few packets it may not impact your business. The same could be said for many other applications, particularly many Cloud-based internal applications.
However, if you do have applications with performance that can impact your business, then more bandwidth doesn’t make the problem go away - in fact, it can make it worse. Now we have Gigabit (and often 10 Gigabit) interfaces sending bursts of traffic into our networks. These interfaces make our new 100Mbps circuits seem small. Sometimes the impacts can be easily seen on basic performance graphs, e.g. Windows/Antivirus/Apple updates (or the Olympics live streamed) that can saturate your 100Mbps link for minutes at a time as multiple devices simultaneously stream updates. Other times a short burst of a fraction of a second can create enough congestion to drop your RDP session or create distortion in your voice or video conferencing call.
However, QoS doesn’t need to be as complicated as we think. Application-based classification takes some of the pain away, and many service providers will take care of this for you. And you may only need a basic ‘CIR’ or committed rate for the applications you really care about.
Basically, the problem still exists, regardless of the bandwidth we have, but it may not be as hard to solve as you think. Of course, if you’re unsure of where to start, you can talk to your service provider or an expert.
