Systems Performance Review

Use this skill to reason about bottlenecks before the project burns time on blind optimization.

Core Workflow

1. State the workload, SLOs, hardware targets, and failure envelope.
2. Draw the critical path and concurrency model.
3. Build a budget table for latency, bandwidth, memory, storage, and power.
4. Identify queueing, serialization, synchronization, and data-movement hotspots.
5. Propose measurement points before proposing optimizations.
6. Rank fixes by expected impact, risk, and implementation cost.

Execution Rules

1. Do not optimize without a budget.
2. Separate steady-state behavior from tail behavior.
3. Check batching, caching, affinity, backpressure, retries, and failure recovery.
4. For accelerators and edge systems, include host-device transfer and power or thermal constraints.
5. Be explicit about whether a claim is analytical, measured, or inferred.

Output Contract

Return: