Showing cs.MSShow all
2 papers · 1 filter
cs.MS2026
Multiple Double Arithmetic on NVIDIA Tensor Cores
Howard Chen, Jan Verschelde
A multiple double is an unevaluated sum of doubles. An NVIDIA tensor core is a specialized high performance compute core for matrix multiplication. The Ampere A100, released in 202…
cs.MS2026
Multiword Arithmetic and Parallel Computing
Jan Verschelde
In many applications, the precision by the available hardware arithmetic is insufficient to guarantee accurate results. Multiword arithmetic is a special type of multiprecision ari…