Performance

Review performance benchmarks, memory requirements, and scaling characteristics for NIM for BMD.

Benchmark Metric

The primary benchmark metric is the average time per molecular dynamics (MD) step per atom:

• P: Time per step per atom
• T: Total wall-clock time
• N: Number of atoms
• S: Number of simulation steps

This metric normalizes performance across system sizes and facilitates throughput comparison.

Batch Size Estimation

At startup, NIM for BMD automatically estimates the optimal batch size for each GPU by running benchmark calculations with representative structures.

INFO     | BMD:cuda:0 |  Atom count 40448 took 396.18 ms, 9.79 μs/atom (2 runs)
INFO     | BMD:cuda:0 |  Atom count 39516 took 391.91 ms, 9.92 μs/atom (2 runs)
INFO     | BMD:cuda:0 |  Atom count 41360 took 418.03 ms, 10.11 μs/atom (2 runs)
INFO     | BMD:cuda:0 |  Estimated batch size: 36876, max: 47136

Performance Results

The following benchmarks use the MACE-MPA-0 model for NVT simulations.

Model Characteristics

Different machine learning interatomic potential (MLIP) models exhibit varying computational characteristics. MACE-MPA-0 is the default model and is downloaded from Hugging Face at container startup. TensorNet and AIMNet2 generally offer faster inference and lower memory usage.

Ensemble Characteristics

Different ensembles introduce varying computational overhead.

Scaling Considerations

• System size: Larger systems typically achieve better GPU utilization, which improves time-per-atom efficiency.
• Multi-GPU: The NIM automatically distributes work across all available GPUs.
• Trajectory output: High-frequency trajectory saving (a small save_interval) can cause an I/O bottleneck during fast simulations.
• Barostat frequency: For NPT simulations, barostat_every controls the frequency of cell move attempts. Higher values reduce overhead.

Memory Requirements

Memory usage scales with the system size. The following approximate values apply to MACE models:

Benchmark Environment