SolarWinds
Network Performance Monitor (NPM)
Monitor network device health and availability via SNMP, the classic SolarWinds use case.
SolarWinds built its reputation on network monitoring, and Network Performance Monitor (NPM) — historically delivered on the Orion Platform — is the flagship. NPM polls routers, switches, firewalls, and servers primarily over SNMP (with ICMP, WMI, and APIs) to track availability, interface bandwidth, errors, latency, and packet loss. It auto-discovers devices and topology, then visualizes health on maps and dashboards aimed at network and IT operations teams.
A building manager who walks the corridors on a fixed round, checking every door, light, and pipe, and immediately flags any that are broken or overloaded — for the network instead of a building.
Key Concepts
1
Its strength is breadth of device support and an operations-focused workflow: fault detection, the NetPath hop-by-hop path analysis for hybrid/cloud routes, and rapid time-to-value for traditional infrastructure. This is a different lineage from cloud-native APM tools — NPM targets the network and infrastructure layer that enterprises still run, including on-premises gear that cloud-first tools largely ignore. It is typically self-hosted, which suits regulated or on-prem-heavy environments.
2
What the interviewer is probing.
What the interviewer is probing.
3
1. "What does NPM monitor and how?" *Probing:* the basics. *Stalls:* "Networks." *Moves up:*
devices and interfaces through SNMP, ICMP and flow data, giving availability, utilisation and errors
per interface.
1. "What does NPM monitor and how?"
4
2. "What is NetFlow used for?" *Probing:* the capability. *Stalls:* "Traffic." *Moves up:*
identifying who and what is consuming bandwidth, which utilisation alone cannot tell you.
2. "What is NetFlow used for?"
5
3. "What is the usual false alarm source?" *Probing:* experience. *Stalls:* "None." *Moves up:*
polling timeouts on a busy or rate-limited device read as down, which is why alert conditions need
consecutive failures.
3. "What is the usual false alarm source?"
6
4. "Where does it fit against application observability?" *Probing:* the positioning. *Stalls:*
"The same." *Moves up:* it answers whether the network is the problem — complementary to APM, which
cannot see a saturated uplink.
4. "Where does it fit against application observability?"