.png?width=200&height=59&name=IPTellogo-byline-2020-dark-v2-pnd-optimised-500%20(1).png "IPTellogo-byline-2020-dark-v2-pnd-optimised-500 (1)")
AIOps, or artificial intelligence for IT operations, uses machine learning, data analytics and automation to improve the performance, reliability and security of IT systems. At IPTel we're always looking for ways to engineer improvements into our service, and AIOps is one of them. In this blog, we'll look at how AI now works inside our managed services monitoring platform, Highlight, and what it means for the networks we support.
Highlight: Our Managed Services Monitoring Platform
Highlight is a cloud-based service observability platform built for managed service providers. It gives our support team one view of every customer network we manage, across LAN, WAN, Wi-Fi and SD-WAN, and it's the platform behind the monitoring in our managed services.
Highlight connects directly to the platforms our customers run. It integrates with the Meraki Dashboard API (security appliances, switches, access points and cellular gateways), Cisco Catalyst SD-WAN and the ThousandEyes API, and it covers other devices through SNMP and ICMP. That means a branch firewall, a ThousandEyes test and a WAN circuit can all be seen side by side.
It's also multi-tenant by design, so each customer's services are kept separate while our engineers still get the bird's-eye view they need.
Alert Noise: How Do We Reduce It and Prioritise Issues?
Highlight cuts alert noise with what it calls hyper-relevant alerting. Alerts are tuned to what matters for each customer, with one-minute polling and settings designed to reduce false positives, so our engineers see real problems quickly instead of wading through an alert storm.
As a support team we often see alert storms across a network, and correlating them to find the real cause can be slow work. Highlight's Insight views, including Heat Tiles and Grid View, show service health at a global, regional, site or connection level. A problem at one site stands out straight away, and so does a pattern across many sites.
Root Cause: How Do We Resolve Issues Faster?
We resolve issues faster by seeing the whole service in one place. Highlight's Grid View shows LAN and WAN elements together, so an engineer can tell in seconds whether a fault sits with the circuit, the SD-WAN edge, the switching or the Wi-Fi.
Whether it's a one-off alert or something we've seen before, that context gives our support team a head start. Knowing which part of the network to check and validate first lets us resolve issues sooner and gives customers confidence in our ability to meet our SLAs.
Highlight's reporting also pulls data from every technology into one set of reports, so a service review covers every technology in one place rather than one report per vendor.
Power of AI: Capacity Forecasting & Anomaly Detection
Highlight's AI features look forward as well as back. Capacity forecasting uses machine learning on historical traffic data to project future load on each circuit and flag when it's likely to run out of headroom. Anomaly detection spots sudden, unusual step changes in traffic and performance, as distinct from normal growth and day-to-day fluctuation.
Together they let our support team act before users notice. A circuit trending towards capacity can be upgraded on a planned timeline rather than in a rush, and an unexpected shift in traffic at a site gets looked at early, whether it turns out to be a fault, a change in how the site is used or something that needs a security review.
Pro-Tip: Forecasts are only as good as the history behind them. Give a new site a few weeks of clean data before you rely on its capacity trend, and tell your provider about known seasonal peaks so they're not mistaken for anomalies.
AI in Managed Services: FAQs
What monitoring platform does IPTel use for managed services?
We use Highlight, a cloud-based service observability platform for managed service providers. It monitors our customers' Meraki, Cisco Catalyst SD-WAN, ThousandEyes and other network services from one multi-tenant platform.
How does AI help in network monitoring?
AI helps by forecasting when circuits will reach capacity and by detecting unusual changes in traffic and performance. That lets the support team fix or plan for issues before they affect users.
Does Highlight work with Cisco Meraki?
Yes. Highlight integrates with the Meraki Dashboard API and supports Meraki security appliances, switches, Wi-Fi access points and cellular gateways.
Can I see the monitoring data for my own network?
Talk to us about the reporting and visibility included in your managed service and we'll show you what's available for your sites.
AI in Managed Services: Summary
AI is now part of how we run managed services every day. With Highlight we get one view of every customer network, alerting tuned to real service health, and machine learning that forecasts capacity and flags anomalies early. The result is fewer surprises, faster fixes and a service our customers can rely on.
Need Help With Managed Services?
If you'd like proactive, AI-assisted monitoring for your network, find out more about our managed services, start with a network assessment, or get in touch with our team at sales@iptel.com.au.