Close

HPE Aruba Networking Blogs

Aruba Central Cloud Gets AI Insights for SD-Branch WAN Gateways

By Jeff Olson, Director of SD-WAN Product and Technical Marketing, HPE Aruba Networking

IT teams have a difficult job when it comes to resolving application issues today as many apps are cloud-hosted and rely on WAN connections. If a performance problem is not tied to the on-prem network, a lot of time is spent working to identify root causes. Without comprehensive end-to-end visibility and automated insights, finding and resolving a simple problem can take hours or days.

As an SD-WAN and SD-Branch vendor, we’re often asked, What is the next big thing when it comes to improving IT efficiency related to WAN issues? Our point of view – and the direction of many of our innovations – is that AI and machine learning will play a big role. Monitoring application and network performance across your branches and providing insights when we see unfavorable behavior is how we save you time and grief.

It’s also important to consider an AI implementation that addresses all aspects of managing branch networking inside and out, including your wireless and wired networks that extend to the WAN. Luckily, we’ve built a complete, full-service AIOps portfolio with more than 40 AI Insights into Aruba Central that’s easy to use. IT admins can now deliver SLAs that achieve the highest level of end user satisfaction, all while optimizing visibility and efficiency.

But because WANs are dynamic by nature, it’s often difficult to determine how well every site, link, and application is performing overall.  For this reason, we’ve enhanced our AIOps platform (existing gateway insights) to provide site-specific WAN health insights and recommendations to help you optimize network performance and identify the following:

  • How do I know if my WAN uplinks at each location are performing at the optimal level?
  • Do I have underutilized links at any of my locations?
  • Can I easily see the performance I’m getting and make the necessary adjustments?
  • Are there preferred configurations or best practices that I can benefit from?

Automating the Solution

Aruba addresses these questions with new AI-powered WAN gateway insights that provide the required visibility to enable IT to tune the network for optimal performance across hybrid WAN uplinks, based on gateway location and compared to your peers.  Let’s take a closer look at each Aruba Insight:

Insight #1: Gateways are using underperforming WAN links

Aruba AIOps identifies individual WAN uplinks that have available bandwidth but are under-performing or are underutilized when compared across all available WAN uplinks on the branch gateway. This information is provided across a hybrid WAN with up to four wired uplinks comprised of any combination of private (MPLS), broadband, Metro-Ethernet, and VSAT. For example, the gateway may be transmitting a large percentage of traffic over a VSAT uplink when the MPLS service is available and underutilized. With this Insight, Dynamic Path Steering policies are adjusted to correct this behavior and send traffic over better performing circuits.

Benefit

Dynamic Path Steering policies for the WAN gateway may become out-of-date over time as new links are added and policy changes are made, resulting in suboptimal performance at a particular location. This AI Insight allows IT teams to adjust the WAN policies to optimize the overall throughput performance achieved on each WAN gateway.

 Insight #2: Gateway WAN uplinks have higher latency than comparable peers

Aruba AIOps provides visibility into how WAN uplinks are performing at a global level and for each branch gateway when compared to similar peer networks with WAN gateways located in the same geographic area. When evaluating uplink performance (loss, latency, and jitter) it is important to compare your gateways to other gateways located in the same location, city, and region. Achievable uplink performance can vary greatly between rural and urban locations and across circuit providers.

Benefit

IT teams managing WAN networks lack visibility into how well their existing WAN uplinks are performing relative to industry best practices and similar deployments. This AI Insight provides visibility into how well WAN gateway uplinks are performing when compared to other peer gateways that reside in the same location (region, county, or city).  This becomes very relevant when selecting WAN service providers, and also when fine-tuning WAN policies; an SLA that may work well for an urban area might be too restrictive in some remote locations.

These SD-WAN insights result in more efficient usage of hybrid WAN uplinks to deliver the following business outcomes:

  • An enhanced user experience which can lead to higher productivity
  • Time savings and operational efficiencies for IT admins managing the network
  • Reduced downtime with a more finely tuned hybrid WAN link set
  • Lower overall WAN circuit costs

Is your network performing at the top of the spectrum, or just at an average level at some locations?

With Aruba AIOps you’ll gain better network visibility, allowing you to tune your network policies to deliver optimal WAN performance at the gateway location and across your entire global enterprise—all from a single UI dashboard: Aruba Central.

 To learn more