Parallel Computing

Use this skill to convert parallel performance work into reproducible scaling decisions.

Workflow

1. Define scaling objective and constraints.

• Capture workload shape, data size, and latency/throughput targets.
• Define hardware assumptions (core count, SMT policy, NUMA context).

2. Choose parallel model and partitioning.

• Select task/data/pipeline parallelism intentionally.
• Set chunk size and scheduling strategy to minimize overhead and imbalance.
• Define shared-state boundaries before coding.

3. Diagnose bottlenecks.

• Check lock contention, false sharing, synchronization frequency, and memory bandwidth pressure.
• Separate algorithmic limits from runtime/scheduler overhead.