A T1 Isn’t What It Used to Be: 3 Best Practices for WAN Performance
by Team AppNeta on

Our Systems Engineering team regularly helps customers deploy PathView Cloud to measure end user experience at remote offices. One of PathView’s unique features is the ability to measure actual achievable network capacity.  T1 class WAN services are a popular way to connect branch offices to corporate data centers and the Internet. I say ‘T1’ class because businesses are deploying WAN services that generally deliver on the promise of 1.54 Mbps capacity. These can be MPLS services, actual T1 or bonded T1, rate limited versions of higher speed links, and a variety or other configurations.

There’s a lot more to WAN services than just capacity, and our teams are seeing a variety of poorly performing T1 services that deliver poor quality capacity. Here are some examples of poorly performing T1 class WAN links and some best practices for managing your WAN performance.

Figure 1: North America Energy Company Branch – Data Center WAN

Measuring the WAN 1

Figure 2: Manufacturing Company WAN Link Connecting Plant to Corporate HQ

Measuring the WAN 2

Figure 3: Business Services Provider Oversubscribed WAN link

Measuring the WAN 3

Three WAN Performance Best Practices:

1) Know your WAN provider service level agreements (SLAs) and how the provider measures them.  What capacity, latency, packet loss, jitter, QoS policies are promised between locations? How do you know you are achieving these today? Do you get regular, accurate reporting from the provider?

2) Continuously monitor actual WAN performance vs. what is committed in your SLAs.

Have you validated WAN performance delivery vs. SLAs with your own tools and expertise?  When there are problems are you able to quickly segment them to provider WAN vs. your LAN?

Your End-User Experience monitoring solution should offer built-in performance SLA verification and track your SLA delivery every minute, 24×7, for a full year.  When there’s a problem, PathView Cloud’s expert alerting system automatically pinpoints the problematic network segment, regardless of ownership and management access.  You simply can ‘t get this with any other network performance tool.

3) Deploy a traffic utilization analysis solution to track existing capacity usage at your sites.

Do you need to buy more bandwidth, or can you eliminate non-business capacity usage by updating your terms of service, deploying quality of service (QoS), or other policies?  To find out what your organization needs to analyze network traffic (and performance problems!) in detail, try out a free trial of PathView Cloud with FlowView. You can deploy the cloud-service without requiring Netflow-capable devices and get unmatched Intelligent Application Visibility you need to track capacity usage.

Article Written By: Adam Edwards

Filed Under: Networking Technology, Performance Monitoring

Tags: jitter , latency , NPM , packets , PathView Cloud , SLA , T1 , WAN