Extends magnum.np to the first Python-native multi-GPU micromagnetic framework via PyTorch Distributed, reporting 7.0x speedup on 8 GPUs for demagnetization and 6.8x on CPU with MPI.
https://pubs.aip.org/aip/apm/article-pdf/doi/10.1063/5
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Magnum.np.distributed: Accelerating Finite Difference Micromagnetic Simulations with Multiple GPUs
Extends magnum.np to the first Python-native multi-GPU micromagnetic framework via PyTorch Distributed, reporting 7.0x speedup on 8 GPUs for demagnetization and 6.8x on CPU with MPI.