BF16 tensor cores on GPUs emulate FP32 SGEMM with superior performance, power efficiency, and numerical accuracy compared to native FP32, including a library implementation that handles denormals.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.DC 2representative citing papers
Presents a tensor-parallel distributed MPS method with block-cyclic partitioning and pivoted QR that emulates Google's RCS benchmark at bond dimension 16384 on 32 nodes, claiming three orders of magnitude better accuracy than prior methods.
citing papers explorer
-
Exceeding the Numerical and Performance Characteristics of IEEE-754 SGEMM with BFloat16 Tensor Cores on GPUs for Scientific Computing
BF16 tensor cores on GPUs emulate FP32 SGEMM with superior performance, power efficiency, and numerical accuracy compared to native FP32, including a library implementation that handles denormals.
-
Tensor-Parallel Emulation of Quantum Circuits with Block-Cyclic Distributed Matrix Product States
Presents a tensor-parallel distributed MPS method with block-cyclic partitioning and pivoted QR that emulates Google's RCS benchmark at bond dimension 16384 on 32 nodes, claiming three orders of magnitude better accuracy than prior methods.