From WAN to Data Centers, the Landscape is Changing – Here’s What You Need to Know

Have you ever had a heated discussion with a home internet technician about the broadband speeds that you’re paying for −, i.e. paying for a 1000 Mbps fiber connection but actually receiving 250 Mbps? It’s frustrating to understand why the speeds coming out of the fiber optic cable don’t necessarily translate when it comes to real-life use, such as upload and download speeds, streaming, and more. That’s because there is a multitude of factors that play into the speeds users receive, such as the speeds recognized by your laptop, time of day, and a number of people on the network, whether you are relying on a Wi-Fi modem compared to using an Ethernet cable, to name a few. Understanding the speed and performance of a Wide Area Networks (WAN) is much like the home internet scenario. Sometimes it’s hard to determine what factors have the most significant impact on the performance of the network and corresponding applications that run on that network. As reliance on public WAN has increased dramatically over the years in nearly every industry, from healthcare to retail, hospitality, and branch office workers, it’s become increasingly important to understand the relationships between WAN performance and the technologies that support them. Let’s take a closer look at critical considerations for understanding how key applications use the network and how latency and other issues can affect their performance.
  • Evaluating a traditional WAN setup and optimization options – Setting up a traditional WAN involves provisioning connectivity from carriers plus tracking and managing bandwidth, setting up security and compliance policies, as well as maintaining connections for branch offices. When considering costs compared to speeds, private circuit links like MPLS are more expensive but deliver faster speeds than broadband connections, or 4GLTE. Done through a data center or technology provider services may also include WAN optimization technologies that mostly try to improve the data flow across the public internet to improve speeds and application performance. For those organizations looking to improve the performance of their WAN, tools like traffic shaping, traffic prioritization, deduplication for disaster recovery, and business continuity applications, can also help in this area. By increasing the capacity of hosting data locally, WAN optimization can also improve because frequently used data is readily available.
  • Ask about your ‘committed information rate’ – A Committed Information Rate (CIR) is generally an agreement from a data provider regarding your WAN contract. The rate essentially guarantees that even though you share a public bandwidth pool with other users, your enterprise is promised a specified rate of data and speed, regardless of how busy the link gets. Network administrators can implement quality-of-service (QoS) configurations to regulate traffic and create policies to drop ‘excess’ traffic before it enters the WAN. For instance, parameters may need to be set specifically for traditional peak usage time, which is generally in the evenings. It’s also essential for admins to consider the trajectory and route of their data paths. Meaning if data travels across time zones, typical congestion times may stretch as the peak times shift from one time zone to the next.
  • SD-WAN gets smart by being application aware – Software-defined WAN is a technology that aggregates multiple Internet connections into a single virtual connection that’s more reliable than those separate connections. SD-WAN is also transported agnostic meaning it can rely on everything from MPLS links to broadband, 4GLTE, and satellite. Working from pre-set policies and with traffic-analysis capabilities, the software selects the most appropriate connection at any moment to shape traffic in real-time. Because less traffic is full MPLS-dependant, operational costs shrink and application performance improves. By breaking out traffic into categories based on application need- i.e., business-critical versus non-business critical- that results in lower latency and frees up bandwidth for business applications running over the connections.
  • Don’t forget latency issues – For companies evaluating their WAN performance, measuring the amount of time it takes traffic to traverse the WAN or WAN latency, is an excellent place to start. Network admins should set a WAN latency benchmark to get a handle on WAN performance and to help isolate performance issues that may occur later. Teams can do this by ‘pinging’ across the links of  WAN routers. Measurement data is collected in various ways, but basically, implementations send four or five echo requests and report back the individual and average round-trip WAN latency as well as any packet loss. While collecting measurement data is relatively straightforward, latency is another issue that involves complex factors. Particularly when the application comes from an external party, and you don’t know its internal operation, like many Software-as-a-Service (SaaS) applications. Teams can talk to technology partners about validating an application in a high-latency WAN environment, before migrating from a LAN to WAN environment, for example.
It today’s interconnected IT environment, it’s essential organizations understand how mission-critical applications use the network and how various issues can affect WAN performance. Armed with this knowledge, network administrators can make smarter decisions about network deployment and WAN optimization.

Leave a Reply