Cisco Subnet An independent Cisco community View more

The Hybrid WAN requires a new approach to network optimization

Now that the WAN is finally evolving, I think it’s time to take a look at the infrastructure that’s used to optimize it.

It’s been well documented that the current wide area network (WAN) and data center interconnect (DCI) architecture has been in place now for well over 20 years. That architecture – the legacy one we’ve all become familiar with – was designed for the days when no one expected much from the network. Slow, fast, whatever. It was the Internet and it was expected to be flaky. Not so anymore, though. We rely on the cloud and mobility to drive our lives. Businesses are network-centric, and that’s driving the evolution towards a new WAN model.

Now that the WAN is finally evolving, I think it’s time to take a look at the infrastructure that’s used to optimize it. The majority of equipment used to optimize the WAN is designed to solve a specific problem on a private, controlled network. The mismatch between legacy infrastructure and the demands of today’s WANs have led to the following problems:

  • Legacy WANs and DCIs are inefficient. Typical utilization of a WAN and DCI links are in the 30% to 40% range, with very few companies actually pushing utilization over 50%.
  • Traditional WANs handle bursts of traffic poorly. With current network architecture, everything can be running fine and then any unexpected surge in traffic can degrade the performance for everyone.
  • Legacy optimization techniques often collide with each other. Over the years, many different appliances have been developed to improve network performance. Solutions such as WAN optimization, application delivery controller, routers, and security devices are widely deployed today. However, each tool only solves part of the network performance challenge. Additionally, these often conflict with each other, creating new WAN and DCI problems.
  • Many management silos. Historically, different groups perform all of the management functions separately, such as visibility, performance, monitoring, optimization, and analytics. This can lead to long troubleshooting times and inefficient capacity planning.
  • Networks can be unstable. The corporate network, particularly the WAN, can be a pain to optimize. That’s why network managers are always fiddling, tweaking, and tuning the WAN and DCI links. Many of these changes, though, are made when IT is rushed and trying to fix a critical problem. The rapid changes combined with poor documentation and manual processes leads to a high amount of human error and unnecessary downtime.    

Obviously, there’s no single solution to all of a WAN’s woes today. What would be ideal is a superset of WAN acceleration, QoS, Firewalls, ADCs, analytics, and management software. This would allow network engineers to manage, optimize, and secure the user experience on a per-application basis. My idea of a utopian solution is that it would work over the traditional private network, but then optimize Internet traffic for cloud services.

While no vendor that solves this problem holistically yet, I am expecting to see more innovation in the optimization appliance market to meet the challenges of hybrid clouds. There’s a strong desire from companies today to re-architect the WAN. Now we just need a solution that makes this happen easily.

To comment on this article and other Network World content, visit our Facebook page or our Twitter stream.
Must read: Hidden Cause of Slow Internet and how to fix it
Notice to our Readers
We're now using social media to take your comments and feedback. Learn more about this here.