SNMP is 28 years old. Its foundations were established in the early 80’s and SNMPv1 was standardized in 1988. It was pretty much the only type of monitoring that existed during the takeoff of the connected world. It has served us well, and it will remain relevant for the foreseeable future. But it has become more important to measure the actual user experience than the devices that connect the user to their services and applications.
Moving to the Cloud
The cloud has virtualized routers, switches, and servers. Even the most resistant industries, like banking and healthcare, are moving into the cloud, dropping any private or hybrid cloud initiatives along the way. As a consequence, the footprint of private networking infrastructure is shrinking with business operations just needing connection from the user to the cloud provider.
We are moving towards an environment that consists of just edge devices at the location where customers or employees need connectivity. On top of that, these devices might be managed through a cloud-based console or dashboard. Of course, you need to know if a router toasts and if CPU utilization spikes, but the fewer the number of devices under management, the less important SNMP becomes.
Finally, networking hardware is very mature and reliable. To the point where we have started seeing white label switches that run third-party OS’s. For example, Cumulus Linux runs on Dell, HP and other switches. Of course, hardware has a finite lifetime, but checking if a switch is toasted is the last thing on your mind when you start troubleshooting. Also, when that happens, the consequences are so obvious that you will figure it out pretty quickly.
At the end of the day, what we care about is if employees and customers can access applications and services over wired and wireless networks and about the quality of their experience. And this is something SNMP doesn’t capture.
Real End User Experience Monitoring
Real End User (REU) experience monitoring gives us exactly that: it measures how applications and services are accessed at each network site. These end-to-end measurements are vendor-agnostic and take into account the on-site infrastructure, the connection to the cloud (ISP, dark-fiber, MPLS), and the server response time. In addition, when you break down the performance by DNS resolution, network latency, and application response time you are able to determine the source of an application slowness of service interruption. On top of that, REU monitoring enables proactive knowledge of these issues even before they affect the actual employees or customers.
More and more network professionals are realizing that SNMP is not enough these days. All our customers already have some kind of SNMP monitoring, such as Solarwinds, IBM Tivoli, or Nagios. They added NetBeez to their arsenal because they realized the big need and room for improvement to the quality of service they provide to their users. In addition, for them it’s a time-saver to know within seconds when a remote site is impacted by performance issues and have the data and visibility to troubleshoot.