Match work to an execution model

15-minute session: read for 3 minutes, trace for 3, experiment for 7, and complete the check for 2. Spend longer whenever you need.

Independent operations over large arrays can expose parallelism. A serial dependency chain cannot be split merely by adding cores. Branch-heavy or tiny jobs may favor CPU execution after launch and transfer overhead. Treat these as hypotheses, not universal rules. Classify a parser, a large matrix multiply, and a small request handler, then state what you would measure.

Practice and evidence: Predict the example or draw its state transitions. Change one assumption, test the boundary, and record the actual result. Explain what this small model does not establish about a production system.

Practice and evidence

Keep three notes: your prediction, the actual result, and why they match or differ. The completion check grades understanding; the practice task builds skill.

Language reference: git.agilaos.com

Language reference: docs.nvidia.com