Parallelism and Performance

Understand Parthenon’s execution model across CPU/GPU backends, task-based parallelism, and performance tuning. Covers reductions, load balancing, and weak-scaling instrumentation.

• Parallelism
• Nested Parallelism
• Tasks
• Task-based reductions
• Load Balancing
• Running a Weak Scaling Test in Parthenon
• Performance Instrumentation