Optimizing your enterprise network performance

Optimizing your enterprise network performance
(Image credit: Shutterstock / carlos castilla)

Optimizing the routes data travels over the internet has never been more important. With more people working from remote locations, coupled with the ever-increasing rate of cloud computing and cloud storage adoption, the demands on enterprise networks have never been greater. Typically, when internet buyers consider the route their data travels, the ‘last mile’ has often been the focal point. 

About the author

Michael Davies works at GTT Communications.

The key connectivity performance criteria applied in the evaluation of last mile connectivity options typically relate to bandwidth, resilience, price and security. In the process, the ‘middle mile’ will have been overlooked. However, it is just as crucial a part of the route data takes and needs to be equally considered for its performance characteristics.

The middle mile is often responsible for carrying traffic the bulk of the distance across the internet. In essence, it is the route made up of the high-capacity IP backbone connections that your data takes at any given time as it moves from point to point around the globe. The middle mile hands off or picks up traffic where it connects with the last mile, ideally at the nearest point-of-presence (PoP) facility to where the data transits.

Understanding how the middle mile works is a business-critical task. Increasingly, the experience users have when connecting to cloud-based applications depends very much on the middle mile aspects of an enterprise’s cloud connectivity strategy. Knowing what, why and how the middle mile is managed is critical to the enterprise network’s performance and reliability.

The middle mile explained

The internet comprises multiple interconnected Internet Service Providers (ISPs) that operate networks, which are individually owned and managed. There are thousands of such networks. The middle mile can be described as a combination of multiple ISP networks that traffic may have to traverse on its way from one end of the internet to the another. These ISP networks are informally categorized as Tier 1, 2 or 3 networks — with Tier 1 generally being the largest and most global and Tier 3 the most local.

However, if traffic has to travel across a number of differing ISPs of various tiers (1, 2, or 3), then it will be subject to challenges. These challenges stem from how those individual networks route the data/traffic, where those networks interconnect and what points of congestion exist within and between those networks. One key issue that may arise is latency. The longer the route the data must traverse, the bigger the risk of latency. An enterprise connecting to a Tier 3 network may, for example, see its cloud traffic route to multiple Tier 2s, then a Tier 1, and then finally to the cloud provider. These extra hops can result in unreliable, end-to-end network performance that negatively impacts user experience.

For enterprises moving from traditional private MPLS-based networks to internet-based WANs, the quality of the internet is particularly relevant. Such moves are often made to reap the benefits in cost savings, agility and cloud connectivity that the internet can provide. To avoid challenges and make the MPLS-to-Internet WAN transition simpler, businesses can choose to use a single Tier 1 provider that is able to provide a more direct route from one end of the internet to the other.

This choice ultimately allows for better route management, capacity and reliability and allows the enterprise to maintain a level of control from end to end. It is fundamental to have the right services in place so that the way data/traffic is routed via multiple PoPs within the middle mile ensures a smoother route.

Getting your middle mile into shape

Here are four tips to make sure your business’ middle mile is in the best possible shape:

  • Work with a top-ranked, Tier 1 ISP – Having a provider with direct control over the middle mile helps businesses avoid unexpected challenges, as they can minimize latency by sticking to one ISP network from one end to the other.
  • Look for a provider with a strong PoP roster – choosing a provider with globally distributed PoPs can ensure more stability, less latency and a smoother route to the destination point.
  • Have last mile flexibility as a priority – When it comes to last mile connectivity, choosing a provider willing to offer a broad range of local access partners can help ensure you have the flexibility and agility you need to balance your cost-performance-risk needs and tolerances on a per-site basis.
  • Take an intelligent approach – Adding SD-WAN technology to help manage your traffic will help you intelligently and dynamically adjust to network changes and traffic flows. SD-WAN ensures that traffic takes the quickest and most direct route to its destination. 

Why the middle mile is critical to your cloud strategy

Connectivity is key to any cloud strategy. Therefore, ensuring you have a Tier 1 network provider at the center of your cloud connectivity and IT infrastructure enables your traffic to transit across one backbone network – rather than via multiple ISPs – to the desired PoPs, before travelling across the last mile to the service destination. This ensures low-latency connectivity and secure, private connections to cloud service providers. Your chosen provider should also offer extensive local access partnerships, providing you with the freedom to select any last mile service.

When shaping your cloud connectivity strategy, paying attention to the middle mile should be high on your agenda. A reliable middle mile that avoids congestion and routes traffic effectively around the internet’s core will help deliver a better performance for your enterprise network and your users.

Michael Davies works at GTT.