Cloud Infrastructure
Linux Performance Triage for Platform Engineers
Read latency, memory pressure, and disk stalls with tooling that fits SSH-first workflows.
- Duration
- 3 weeks
- Format
- Async labs with two live deep dives
- Skill focus
- Beginner
- Project intensity
- Medium
280,000 KRW
Informational tuition reference — no checkout on this static site.
Program narrative
A pragmatic tour of perf-oriented tooling on modern kernels, aimed at engineers who touch nodes but are not kernel developers. Labs include cgroup surprises, IO stack clues, and concise write-ups your incident channel will appreciate.
Included focus areas
- • Latency histogram interpretation without chart overload
- • cgroup v2 memory pressure signals
- • Disk stall patterns and when to escalate to storage teams
- • CPU scheduler basics tied to noisy neighbor cases
- • Safe data collection during incidents
- • Short postmortem sections that link evidence
- • Checklists for handoffs to vendor support
Outcomes you can show
- Produce a triage checklist adopted in a lab incident drill
- Capture a three-step remediation with measured impact
- Pair-present findings without relying on dashboard-only storytelling
Mentor of record
Marcus Reid
Senior DevOps instructor focused on measurable node behavior.
Participant notes
IO stack week mirrored a production mystery we had last quarter. Checklists are now in our wiki.
Straight answers
No; labs use VMs with injected slowdowns.