System Interrupts sitting at 15%, 25%, even 40% CPU in Task Manager — with no way to end it, no file location, nothing to uninstall. It’s one of the most confusing entries in Windows because it isn’t a program at all. It’s the kernel’s accounting for time spent servicing hardware, and when it’s high, some device or driver is screaming for attention thousands of times per second.

The best fix isn’t a single tweak — it’s an ordered hunt: identify the interrupting device, fix or update its driver, and if the driver is fine, suspect the hardware. Here’s the order that resolves most cases with the least wasted effort. For the closely related audio-stutter variant of this problem, see our DPC latency guide .

Why this happens

  • Buggy device driver: A driver mishandling interrupts (network, storage, or USB drivers are classic) generates an interrupt storm.
  • Failing hardware: A dying drive, flaky USB device, or bad controller generates spurious interrupts as it errors and retries.
  • Power management conflicts: Aggressive CPU power states (C-states) interacting badly with a driver cause interrupt floods on wake.
  • Outdated BIOS/firmware: Interrupt routing is partly firmware’s job; old BIOS versions misroute on newer Windows builds.
  • Problematic USB devices: A malfunctioning USB peripheral (or hub) can generate thousands of interrupts per second.
  • Storage controller issues: AHCI/NVMe driver problems show up as System Interrupts during disk I/O.
  • Network adapter driver bugs: Wi-Fi drivers are among the most common offenders, especially after Windows updates.

How to fix it

Work through these in order — diagnose first, then act.

Fix 1: Verify the reading and find the spiking device

Confirm it’s real, then let a proper tool name the device.

  1. Don’t trust Task Manager alone — download Process Explorer (Microsoft Sysinternals) and check the Interrupts pseudo-process, or use LatencyMon to measure DPC/interrupt load precisely.
  2. In LatencyMon, run it for 5 minutes at idle, then check the Drivers tab sorted by DPC count — the top driver is your prime suspect (often ndis.sys = network, storport.sys/stornvme.sys = storage, usbport.sys = USB).
  3. Note when it spikes: constant at idle, only during Wi-Fi activity, only when a USB device is plugged in, only on battery? The trigger narrows the device enormously.
  4. Check Device Manager for yellow warning icons — resolve those first regardless.
  5. If LatencyMon shows no driver with extreme counts but Task Manager shows high Interrupts, suspect hardware-level issues (Fix 6) rather than a driver bug.

Fix 2: Unplug USB devices one by one

The fastest hardware isolation test — takes five minutes.

  1. Unplug every USB device: hubs, external drives, docks, receivers, printers — everything.
  2. Watch System Interrupts for 2–3 minutes. Drops to normal? One of those devices (or its driver) is the storm source.
  3. Reconnect devices one at a time, waiting a minute between each, until the spike returns.
  4. For the offending device: try a different port, a different cable, update its driver — or replace it if it’s failing hardware (a dying external drive is a classic).
  5. Don’t forget internal-ish USB: some laptops route the webcam, fingerprint reader, or SD reader through internal USB — disable them in Device Manager one at a time as a test if external unplugging didn’t help.
  6. Docking stations deserve special suspicion — their USB/Ethernet/display combo drivers are interrupt-heavy when buggy.

Fix 3: Update the suspect drivers from the manufacturer

Generic Windows drivers are the usual bug source.

  1. Based on Fix 1’s findings, get the right driver: network issues → Wi-Fi/Ethernet driver; storage → chipset + Intel RST/NVMe; USB → chipset.
  2. Download from your laptop maker’s support site for your exact model: Dell , HP , Lenovo — maker drivers are tuned for your hardware.
  3. For Intel Wi-Fi/Bluetooth, the Intel Driver & Support Assistant often has newer fixes than the laptop maker.
  4. Install, reboot, and re-measure with LatencyMon — driver fixes show up immediately in the DPC counts.
  5. If the newest driver is the problem (started after a driver update), roll back: Device Manager > device > Properties > Driver > Roll Back Driver.

Fix 4: Update BIOS and chipset firmware

Interrupt routing lives partly in firmware.

  1. Check your current BIOS version (run msinfo32, look at BIOS Version/Date).
  2. Visit your maker’s support site and compare — BIOS updates frequently list “fixes system instability” or “improves device compatibility,” which often means interrupt handling.
  3. Install exactly per the maker’s instructions, with the charger connected and nothing else running. Do not interrupt a BIOS flash.
  4. Also install the latest chipset driver package — chipset drivers and BIOS work together on interrupt routing.
  5. Re-measure after the update. BIOS fixes for interrupt storms are common enough that this step resolves a meaningful share of stubborn cases.

Fix 5: Adjust power management

C-state and link-power transitions can trigger interrupt floods.

  1. Open powercfg.cpl, select High performance (or create it) as a test — if interrupts drop, power-state transitions were the trigger.
  2. In Device Manager, find the suspect device (network adapter, USB hubs) > Properties > Power Management tab > uncheck Allow the computer to turn off this device to save power as a test.
  3. For USB: Device Manager > Universal Serial Bus controllers > each USB Root Hub > Power Management > uncheck the power-saving box.
  4. Advanced: in the power plan settings, set PCI Express > Link State Power Management to Off as a diagnostic.
  5. If these help, you’ve found a workaround — but still pursue the driver/BIOS fix, because disabling power management costs battery life. The ideal end state is fixed drivers with power saving on.

Fix 6: Isolate failing hardware

When drivers and firmware are clean, the device itself is suspect.

  1. Run your maker’s hardware diagnostics (Dell SupportAssist, HP PC Hardware Diagnostics, Lenovo Vantage) — full component scan, especially storage and system board.
  2. Test the drive: manufacturer’s SSD toolbox health check, plus chkdsk C: /scan in an elevated terminal.
  3. For laptops with removable Wi-Fi cards (rare now) or accessible RAM, reseating can fix interrupt issues caused by poor contact — but on most modern ultrabooks, this means a service visit.
  4. Boot a Linux live USB: if interrupts are high there too, it’s definitively hardware — no Windows driver involved. If Linux is calm, it’s a Windows driver problem you haven’t found yet.
  5. Check temperatures too — overheating components error more, and errors generate interrupts. Clean dust from vents; repaste only if you’re experienced.

Fix 7: The targeted disable as last resort

When you’ve identified the device but can’t fix it yet.

  1. In Device Manager, you can Disable device on the offending non-critical hardware (fingerprint reader, SD card reader, secondary network adapter) to stop the storm while awaiting a driver fix or replacement.
  2. Never disable the storage controller or primary display adapter this way — you’ll lose the system.
  3. Check the maker’s forums for your exact model + “system interrupts” — some models have known-bad driver versions with community-confirmed good versions to roll back to.
  4. As a final software measure, an in-place Windows repair install can fix kernel-level corruption that mimics driver issues.
  5. If hardware is confirmed failing and the laptop is under warranty, contact the maker with your LatencyMon evidence — “DPC latency from stornvme.sys, diagnostics show drive errors” gets faster service than “laptop slow.”

Closing

System Interrupts high CPU is the kernel telling you a device is misbehaving — your job is translating that into which device. Measure with LatencyMon, unplug USB suspects, update maker drivers and BIOS, and test power management. The best fix is the one aimed at the actual interrupting device, found by diagnosis rather than guesswork. If the laptop is slow overall rather than just interrupt-bound, run the broader slow laptop diagnosis in parallel.

Frequently asked questions

What does System Interrupts mean in Task Manager?
System Interrupts isn’t a real process — it’s Task Manager’s placeholder for CPU time spent handling hardware interrupts and deferred procedure calls (DPCs). When a device needs attention, the CPU pauses normal work to service it. High ‘System Interrupts’ CPU means some hardware device or its driver is demanding excessive attention, usually due to a driver bug or failing hardware.
Is high System Interrupts CPU dangerous?
Not directly dangerous, but it signals something wrong: a misbehaving driver can waste 20% of your CPU indefinitely, draining battery and heating the laptop. In rare cases it indicates failing hardware generating spurious interrupts. It won’t damage the CPU, but the underlying cause deserves fixing — especially on a laptop where wasted CPU means wasted battery.
Why can't I just end the System Interrupts task?
Because there is no task to end — System Interrupts is a bookkeeping entry, not a process. Right-clicking it shows no useful options. The CPU time is being consumed inside the Windows kernel on behalf of hardware. The only fix is finding which device or driver is generating the interrupt storm and addressing that root cause.
What is a normal System Interrupts CPU percentage?
Under 1–2% at idle is normal; brief spikes to 5% during heavy I/O are fine. Sustained readings above 5% at idle, or anything above 10%, indicate a problem worth investigating. Note that Task Manager’s number can be misleading on some systems — confirm with Process Explorer or LatencyMon before chasing ghosts.