Introduction to Network Performance Monitor

Discover Network Performance Monitor (NPM), designed to give you continuous visibility into the health and performance of your network devices across on-premises and hybrid environments.

Download the PDF: NPM Product Guide (formerly known as Administrator Guide)

Manage interfaces

Once NPM discovers a device, it automatically begins monitoring the health and status of its interfaces. From there, administrators can review current status, investigate problems like duplex mismatches, and adjust how individual interfaces are configured and polled — all without switching between separate monitoring and configuration tools.

Interface management covers both day-to-day monitoring and one-off configuration changes. To check current status, health, and downtime history, see View interface status, interface health, and details about downtime. To change polling intervals, thresholds, or other properties for an interface, see Edit interface properties. To pause monitoring, remove, or remotely shut down and enable interfaces, see Set interface management states and Remotely manage monitored interfaces.

Monitor and forecast capacity usage

NPM tracks historical usage on nodes, interfaces, and volumes so you can identify capacity problems before they affect performance. Rather than waiting for a resource to hit its limit, you can see how usage is trending and estimate when a warning or critical threshold will be reached.

Forecasts are calculated using either peak or average usage patterns, and the default method applies environment-wide unless you need to fine-tune it for specific objects. To view forecasted usage trends for a node, interface, or volume, see Forecast capacity for nodes, interfaces, or volumes. To change the calculation method or thresholds for your whole environment, see Change capacity forecasting settings globally. To apply different settings to a specific critical device, see Customize capacity forecasting settings for single nodes, interfaces, or volumes.

Network Routing Insights

Routing Insights correlates routing data across your network — routes, peers, interfaces, and VRFs — so you can understand how routing components relate to each other instead of pulling a routing table from each device individually. This makes it easier to assess the stability of a routing path and see how a single change might affect the rest of the network.

Because routing issues span protocols and devices, Routing Insights groups related data together instead of treating each device as a separate silo. To review routing tables for monitored devices, see Network Routing Insights: Routing table. To investigate route flaps or unstable peers, see Network Routing Insights: Routing neighbors. To understand how routes and peers map to a specific VRF, see Network Routing Insights: VRF.

Discover your network paths with NetPath™

NetPath shows you the actual path traffic takes between your users and the applications they depend on, whether that path stays entirely inside your network or crosses out to the internet. Instead of guessing where a slowdown is happening, you can see every hop along the way and tell whether the problem is internal, with your ISP, or somewhere else entirely.

To use NetPath, you deploy lightweight probes that continuously test a path to a defined service, and NetPath enriches that data with details from any other Orion Platform products you have installed. To confirm your environment meets the prerequisites, see NetPath requirements. To start monitoring a path, see Create a service and Create a probe. Once paths are running, use View a network path to review results, and Troubleshoot my network with NetPath data to diagnose a reported issue hop-by-hop.

Monitor firewalls

Firewalls sit at the edge of your network, so their health and configuration directly affect whether the services behind them stay reachable. NPM monitors firewall status, VPN tunnels, and platform-specific health so you can confirm your firewalls are up and passing traffic as expected, without switching to each vendor's own management console.

Available detail depends on your firewall vendor, since each exposes different platform-health and VPN information. To monitor Cisco ASA firewalls, including platform health, VPN tunnels, and access lists, see Cisco Secure Firewall ASA (Cisco ASA) firewalls. For Palo Alto Networks firewalls, see Palo Alto Networks Next-Generation Firewall (NGFW). For Cisco Firepower or Fortinet FortiGate firewalls, see Cisco Secure Firewall Firepower firewalls or Fortinet FortiGate firewalls.

Monitor load balancers

A load-balanced service depends on many components working together — virtual servers, pools, and individual pool members — and a failure in any one of them can be hard to trace from the load balancer's own interface alone. NPM gives you a single view of that entire delivery chain, so you can see which component is unhealthy and why a service might be degraded or unavailable.

NPM currently supports F5 BIG-IP load balancers, with drill-down data covering device state, high availability, and health monitors. To set up monitoring, see Set up Network Insight for F5 BIG-IP load balancers. To review the health of a load-balanced service and its members, see Monitor services delivered by F5 BIG-IP load balancers. To temporarily take a pool member out of service, see Remove F5 member from rotation.

Monitor SD-WAN

SD-WAN deployments route traffic dynamically across multiple links, which makes it harder to tell whether a performance problem is coming from the underlying network or from the SD-WAN overlay itself. NPM monitors your SD-WAN orchestrators and edge devices together, so you can see tunnel health, jitter, latency, and packet loss alongside the rest of your network data instead of relying on a separate vendor dashboard.

Each vendor is added and monitored slightly differently, typically by connecting to the orchestrator first and then bringing in edge devices. To set up monitoring for your vendor, see Cisco Catalyst SD-WAN — cEdge devices, HPE Aruba Networking EdgeConnect SD-WAN, Fortinet Secure SD-WAN, Palo Alto Networks Prisma SD-WAN, or VeloCloud SD-WAN.

Monitor software-defined networks

Software-defined networking separates network control from the underlying hardware, and NPM brings that same visibility into the SolarWinds Platform Web Console so you can monitor SDN fabrics alongside your traditional infrastructure instead of relying on a separate management plane.

Right now, that coverage is Cisco Application Centric Infrastructure (ACI) — fabric health and tenant-level details included. To get started, see Monitor Cisco ACI. For built-in alerting and reporting once ACI monitoring is running, see ACI alerts and reports.

Monitor switches

Switches are often the densest part of your topology — a single switch can have dozens of interfaces connecting to other devices — so switch and port-level health is central to diagnosing most network issues. NPM monitors switches from all major vendors, giving you a consistent view of device and interface status regardless of the underlying hardware.

Setup and available detail vary slightly by vendor, particularly for Cisco Nexus devices, which have their own dedicated setup and data views. To monitor a specific vendor's switches, see Arista switches, Aruba switches, Cisco Catalyst switches, Fortinet switches, or Monitor Meraki switches. For Cisco Nexus devices specifically, see Monitor Cisco Nexus nodes.

Monitor switch stacks

Switch stacks let you manage a group of physically stacked switches as a single logical unit, and NPM extends that same single-pane view into monitoring — you can see stack membership, ring status, and the health of individual stack members without treating each switch as a separate device.

Because stacking works differently across vendors, monitoring details vary slightly by hardware. To set up monitoring for your stack, see Monitor Cisco switch stack, Monitor Aruba switch stacks, or Monitor Juniper Virtual Chassis (Juniper switch stack).

Monitor wireless infrastructure

Wireless problems are notoriously hard to diagnose from wired-network data alone — a dead zone, an overloaded access point, and a failing controller all look completely different from a switch port going down. NPM monitors wireless controllers, access points, and connected clients so you can see coverage, signal strength, and controller health in one place, and visualize physical coverage with heat maps instead of waiting for client complaints to find dead zones.

Because wireless hardware varies widely by vendor, NPM supports controllers and access points across most major brands, plus a dedicated heat-mapping workflow for supported Cisco access points. To review currently monitored access points, clients, and rogues, see View wireless data. To set up monitoring for a specific vendor, see topics such as Aruba Central wireless, Juniper Mist wireless, or Monitor Meraki devices. To visualize wireless coverage and identify dead zones, see Create wireless heat maps.

Configure polling

Not every device can be monitored the same way — some support standard SNMP, while SD-WAN orchestrators, wireless controllers, and certain Cisco devices need API-based or CLI-based polling instead. Choosing the right polling method up front affects both the level of detail you get and the load placed on your polling engines.

This matters most in orchestrator-managed environments, where you typically add the orchestrator first and then bring in edge devices using API-only or combined polling. To add edge devices for monitoring, see Add edge devices or Add edge devices (API only). For vendor-specific API polling considerations, such as wireless controllers or SD-WAN orchestrators, see API polling for wireless.

Troubleshoot network performance issues with Performance Analysis dashboards

When a performance issue spans multiple devices or metrics, checking each resource's individual view one at a time makes it hard to see how they relate over time. Performance Analysis (PerfStack) lets you drag metrics, events, and log data from across your environment onto a single, shared timeline, so you can visually correlate what was happening across your network when an issue occurred.

This is especially useful for intermittent or hard-to-reproduce issues where you need to compare data from multiple entities side by side. For an example of diagnosing dropped packets along a path, see PerfStack: Dropped packets scenario. For an example of investigating slow resources reported from a branch office, see PerfStack: Slow resources in a branch office scenario.