Network monitoring and AIOps

Network monitoring tools answer two questions: what is on the network, and is it working right now. The older generation polled devices over SNMP and sent an alert when a counter crossed a threshold. The tools listed here still do that, and they add automatic discovery, live topology maps, flow analysis and AI features that group related alerts so one failed switch does not page the on-call engineer dozens of times.

They split roughly by scale. Auvik, Domotz and PRTG suit offices, schools and the clients of a managed service provider, where the job is to see every device on a few sites and back up the configs. Kentik, ThousandEyes and LogicMonitor are built for larger networks, service providers and companies that care how traffic crosses the internet to SaaS apps. Zabbix is the open source option for teams that want to run everything themselves.

"AIOps" means different things depending on the vendor. In most of these products it covers anomaly detection on metrics, correlation of alerts into incidents, and plain-language summaries of what changed. Ask for a demo on your own data, because a model that learns normal behavior needs a few weeks of history before its alerts mean much.

Featured AI IT operations tools: network monitoring and aiops

SponsoredNetwork monitoring and AIOps This position is open Put your tool at the top of the network monitoring and aiops page. Paid spots are always labeled Sponsored. From $99 a month. Get featured

What a featured listing costs ยท from $99 a month

9 AI IT operations tools in network monitoring and aiops

Auvik

Cloud network monitoring that discovers devices, draws a live network map and backs up switch and firewall configs.

  • Built for MSPs (multi-client)
  • Developer API

Domotz

Network monitoring and device discovery aimed at MSPs and integrators, with remote access to devices on client networks.

  • Built for MSPs (multi-client)

PRTG Network Monitor

by Paessler

Sensor-based monitoring of networks, servers and applications.

  • Free plan or free tier
  • Self-hosted option
  • Developer API

LogicMonitor

Hosted monitoring for networks, servers and cloud resources, with AI-assisted alert grouping and anomaly detection.

  • Developer API

ThousandEyes

by Cisco

Monitors internet paths, SaaS apps and WAN links from many vantage points, showing where along the route a problem starts.

Kentik

Network observability built on flow data, used to analyze traffic, capacity and DDoS events across large networks.

  • Developer API

Zabbix

Open source monitoring for networks, servers and services, run on your own infrastructure with templates for common devices.

  • Free plan or free tier
  • Open source
  • Self-hosted option

NetBrain

Network automation and dynamic mapping for troubleshooting, running diagnostic checks across devices along a problem path.

SolarWinds Observability

by SolarWinds

Network and infrastructure monitoring from SolarWinds, available self-hosted or as a hosted service.

  • Self-hosted option

Before you choose, count your devices and sites, list the vendors you run, and decide where the monitoring server or collector will live. Check how alerts reach your on-call people, whether config backup supports your switch and firewall models, and how long metrics are kept. If you are an MSP, check that each client can be kept separate in maps, alerts and reports.

Collectors and agents deserve a close look. Cloud tools such as Auvik and Domotz put a small collector at each site that reaches out to the vendor, which is easy to deploy but means scan data leaves the building. Self-hosted tools keep the data on your servers and need someone to keep the server patched and sized as device counts grow.

Plan alerting before you switch anything on. A new monitoring tool will find problems nobody knew about, and if every one of them pages a person, the team will mute the alerts within a week. Start with device down, interface down on uplinks and high utilization on WAN links, route the rest to a daily report, and add alerts as you learn what matters on your network.

Questions about network monitoring and aiops

Do I need SNMP enabled for these tools to work?

For most device health data, yes. SNMP, and on newer gear streaming telemetry or vendor APIs, is how monitoring tools read interface counters and device status. Enable it with read-only credentials on a management network, and use SNMPv3 where devices support it.

What does flow monitoring add over basic monitoring?

Basic monitoring tells you a link is busy. Flow data such as NetFlow or sFlow tells you who and what is using it: which hosts, applications and destinations. That is what you need to answer "why is the internet slow" or to spot unusual traffic.

Which of these suit a managed service provider?

Auvik and Domotz are built with MSPs in mind, with separate client sites, maps and alerting. Check the MSP marker on each listing, and ask how billing and reports work when you add or remove client sites.

Can I run network monitoring on my own servers?

Yes. Zabbix is open source and self-hosted, and PRTG and SolarWinds offer editions you install yourself. Self-hosting keeps data in house but means you patch, back up and size the monitoring servers yourself.

Related categories