Network Operations Centers can be somber places. I’ve seen quite a few during my last two decades in technology. To monitor and manage a vast network is equal parts nerve racking and mind numbing boredom as you stare at wall-to-wall screens, waiting for an alarm to go off.
Over the years, networks have grown to be more and more complex because there are so many interdependent factors that affect their behavior. These factors include traffic flows, network typologies, network protocols, hardware, software, and most importantly, the interactions among them.
More frightening, the steps required to do these very complex network-wide changes are in many cases still manual. In most enterprises this is done box-by-box one at a time – both time consuming and error prone. On top of that you’ve got to make sure that you’ve calculated for variance with lots of different flavors of swtiches and routers in the field. For example, to leverage a powerful feature such as Performance Routing (PfR), which can double your capacity, each WAN router must be properly configured and the overall WAN architecture adapted to the applications requirements. This can take many man-hours to implement, troubleshoot and optimize – which explains why most IT organizations spend 80-90% on operations, leaving little time for much needed innovation. Add in security, QoS, and mission critical applications and within seconds you can see this akin to kicking a sleeping beast. Once we move to the massive number of devices that are expected for the Internet of Things (IoT), then it simply becomes an unsustainable exercise in failure. Read More »