Quickly monitor CPU, memory and IO per systemd cgroup to spot resource hogs. 22.06.2026 | reading time: 2 min When a server feels sluggish run `systemd-cgtop` to watch CPU, memory and IO usage grouped by systemd cgroup and spot which unit eats resources; it gives a live, prioritized view of slices and services so you can act fast. Hands-on Example Do this on a machine: run `sudo systemd-cgtop` and you will see a header like `UNIT CPU% MEM% READ WRITE` then lines such as `user.slice 12.3 34.1 1.2K 3.4K` and `nginx.service 8.7 5.4 512 1.1K`, then press `q` to quit the interactive view. Practical Workflows Use `systemd-cgtop` interactively to identify noisy units, then run `systemctl status` or `ps` against the offending unit to trace processes; it is great for live triage but not ideal for automation so fallback to `systemd-cgls` or reading the cgroup filesystem for scripts. When and Where to Use It Run it on container hosts, multi-tenant servers or during load tests to see which slices or services dominate CPU, memory or disk IO; remember it requires systemd and active cgroups and will reflect the systemd cgroup layout you use. Final Take Try it, find the noisy unit, and fix the root cause; if he wants to go deeper into Linux internals and certification consider CompTIA Linux+ or LPIC-1 and use bitsandbytes.academy for intensive exam preparation. Join Bits & Bytes Academy First class LINUX exam preparation. utilities processes troubleshooting
