NVIDIA releases Daviator GPU kernels to speed up VASP plane-wave DFT
NVIDIA has introduced Daviator, a library of GPU kernels that handles the k-point steps of plane-wave DFT calculations in batches instead of sequentially, plus a mixed precision option for its batched Davidson eigensolver. Against VASP 6.6.0 CPU and OpenACC baselines, NVIDIA reports up to 22x faster time-to-solution and up to 75% lower cost for a projected OMat24 dataset workload, and more than 85% shorter execution time for a 64-atom Bi-doped GaAs system. Both capabilities are available through the NVIDIA ALCHEMI VASP microservice by setting LDAVIATOR=True and LDAVIATOR_MIXED_PRECISION=1 in the INCAR file, with existing VASP inputs running unmodified. Daviator is currently limited to single-node execution.
Why it matters
Small-unit-cell, k-point-dense DFT jobs could previously run slower on GPUs than on CPUs, pushing researchers to route work by system size; NVIDIA says the batched kernels remove that decision. The mixed precision layer also lets an FP64-bound code recover throughput from FP32 units on GPUs where FP64 rates have stayed flat.
Sources
- PrimaryNVIDIA Developer
Written by AIToolMaps from the source above and checked against it sentence by sentence.