How can I monitor resource usage on my Windows VPS?
What this is
Watching what your Windows VPS is doing with its CPU, memory, disk, and network, the built-in tools cover three depths (now / detailed now / history), and knowing which to open is most of the skill. (If you're here because the VPS is currently slow, that guide applies this toolkit to the hunt.)
Layer 0: the panel Graphs (the outside view)
Your VPS's Graphs tab plots CPU, memory, disk, and traffic over time, measured from the host, so it works even when the VPS itself is wedged, and it's the fastest answer to "when did this start" and "is my plan the right size". The tools below add the inside view: which process.
Layer 1: Task Manager, the glance
Ctrl+Shift+Esc → Details tab, sort by CPU or Memory, and the hungriest process is at the top; the Performance tab shows the machine-wide curves for the last 60 seconds. This answers most questions in ten seconds, and its limits are exactly two: no per-process disk/network breakdowns worth trusting, and no memory of anything before you opened it.
Layer 2: Resource Monitor, the detail
The link at the bottom of Task Manager's Performance tab, or run resmon. Four tabs, each per-process:
- CPU: like Task Manager, plus per-service breakdown.
- Memory: the honest per-process working sets, and the machine-wide In Use / Standby / Free map (standby is cache, not a problem).
- Disk: which process is hitting which file, and disk queue length, the tab that finds runaway logging and answers "what is grinding the disk".
- Network: per-process traffic with remote addresses, Windows' answer to nethogs, and the place a miner or exfiltration shows its destinations.
Filter by ticking a process at the top and every tab scopes to it, the fastest way to profile one suspect.
Layer 3: Performance Monitor, the history
Task Manager and resmon evaporate when closed; perfmon records. The two uses that matter on a VPS:
- Live counters: Monitoring Tools → Performance Monitor, add counters (green +), Processor, Memory → Available MBytes, PhysicalDisk → Avg. Disk Queue Length, LogicalDisk → % Free Space, and watch them plot.
- Data Collector Sets, the 3 AM answer: right-click User Defined → New → Data Collector Set, use the System Performance template (or your counter list), set a schedule, and it logs to disk continuously. The next time something happens overnight, the report (under Reports) has the curves for that exact window, the built-in equivalent of what Netdata does on Linux.
Keep collector logs on a leash (they grow; point them at a folder you watch, and set a stop condition), and pair the resource history with the Event Log for the why behind any spike, timestamps across the two are how cases close.
Which tool, when
| Question | Tool |
|---|---|
| "What's eating CPU right now?" | Task Manager, Details |
| "What's hammering the disk / talking to the network?" | Resource Monitor |
| "What happened at 3 AM?" | Performance Monitor collector set (+ Event Log) |
| "When did this start / is the plan too small?" | Panel Graphs |
Still need help?
You can open a support ticket. So we can help on the first reply, it's worth mentioning:
- the VPS hostname or IP,
- which resource looks wrong and when (a screenshot of Task Manager or the Graphs helps),
- what was running at the time.
Related questions
- "How do I see what's using CPU, disk, or network on Windows Server?"
- "What's the Windows equivalent of htop or nethogs?"
- "How do I record resource usage history on my Windows VPS?"
- "How do I set up a Data Collector Set?"