Diagnose Storage I/O Performance and Identify the Workload
Together AISoftware Engineer · Technical Screen
Last updated
Diagnose storage slowness using device metrics, process attribution, workload shape, cache and durability effects, and measurements appropriate to the storage stack.
A service is slow and storage I/O is suspected. Explain how you would use system metrics such as iostat, identify the responsible workload, and decide whether the storage path is the bottleneck.
Constraints & Assumptions
No device type or filesystem is specified. Interpret queueing, latency, and utilization in the context of the actual storage stack rather than assuming one threshold applies to every disk or SSD.
Clarifying Questions
Are reads or writes slow? Is the pattern random or sequential? What are the request size, concurrency, caching, and durability requirements? Does the application wait on sync or metadata operations?
Follow-up questions
- Why is a high utilization percentage not sufficient to prove saturation on every device? How would you separate application queueing from device service time? What changes when storage is remote?
Solution walkthrough is for members
Get the full walkthrough with reference code and complexity analysis, plus Kitty Coach at 300 messages a day. Hints, the code runner and community answers stay free.