It’s safe to say that no one likes change. It makes us uncomfortable and puts us in positions that we are not familiar with. This is certainly true in our personal lives, but it’s also true in our jobs. It seems every time there is some kind of major technology shift everything changes, and IT needs to adapt, adopt new best practices, and develop new skills. Think back to the transition from mainframes to PC computing, TDM voice to VoIP, and physical servers to virtualization. Each of these seismic shifts required IT organizations to completely change the way they operate.
The enterprise WAN is in the midst of a major transformation, requiring network managers and administrators to do things differently and to develop new skills. The key to developing a plan for the new world is to understand the new rules of networking in the digital era. Below are five “rules of the road” for the WAN in the digital era:
- Simplification is the network engineer’s best friend. A decade ago networks were fairly simple. Well-defined parameters, corporate-owned endpoints, and users who worked on company premises with applications hosted in the data center. Today’s environment is radically different, where BYOD, WiFi, and cloud-based applications rule the workplace. The complexity of the WAN edge has gone through the roof making it significantly more difficult to manage and to align with business requirements. Simplifying the WAN edge must be a top initiative for network managers and administrators.
- The need for speed. Digital organizations move with speed and agility. Business leaders expect their technology infrastructure and applications to accelerate growth and profitability and to maintain competitive edge. With legacy WANs, the average time it takes to implement a change across the network is 4-5 months, making it unacceptable for an agile business. Network operations must find a way to run the network and applications faster to ensure that it does not become a bottleneck or hold the company back.
- Manual processes are the mortal enemy of network operations. The concept of network automation has been around for some time but has given many network engineers the “heebie jeebies” for several reasons. First, there’s the element of trust, since most high-level router jockeys think they can make changes faster and more accurately. As it turns out, this is nothing more than a fallacy – the #1 cause of network downtime remains human error from misconfiguration. Second, automation tools implement changes much faster than any person or team of people can. The time has come for network engineers to shed their legacy thinking that automation is bad and embrace it as the way forward.
- Broadband is business grade. Broadband has been widely used by consumers for decades now but many businesses have shied away from commodity connectivity, preferring higher-cost MPLS circuits with established Service Level Agreements (SLAs). On paper this might seem logical — pay more for the certainty that the network will perform as required by the business — but the fact is that most SLAs are nothing more than false hope. Even if the business can prove an SLA violation occurred, service providers typically only reimburse customers for the time they were in violation. An hour outage for a business could result in millions of dollars being lost and the telco reimburses them a few bucks? And somehow, that seems fair? A single broadband connection won’t solve these woes, but multiple broadband links with WAN optimization and aggregation tools can. Broadband is now viable for business connectivity.
- Visibility is mandatory. Historically I would put the lack of application visibility into the inconvenient category for network managers and administrators. A user would call saying a certain application was performing poorly and the long, arduous task of sleuthing the root cause would commence. However, in the past workers weren’t as reliant on the network to get stuff done as they are today. As more services move to the cloud and more things get connected, reliance on the network will only become more critical. End to end visibility and control of application traffic and performance will enable network operations to diagnose and address problems before they impact network or application performance. Without visibility and control, the best you can hope for is an educated guess. A finger in the air is no way to manage the network.