Back to Glossary

Specialized processing units in NVIDIA GPUs designed for matrix operations common in deep learning.

Tensor Cores perform mixed-precision matrix multiply-and-accumulate calculations in a single operation. They provide significant speedups for deep learning training and inference. Found in NVIDIA's Volta, Turing, Ampere, and Hopper architectures.

Explore More Terms

RDMA

Remote Direct Memory Access - a technology that allows direct memory access between computers without involving the CPU.

Qubit

The fundamental unit of quantum information, analogous to a classical bit but capable of existing in superposition states.

GitOps

An operational framework that uses Git repositories as the single source of truth for infrastructure and application configurations.

NVMe

Non-Volatile Memory Express - a specification for accessing solid-state storage over PCIe.

CUDA

NVIDIA's parallel computing platform and programming model for general computing on GPUs.

cgroups

Control groups - a Linux kernel feature for limiting, accounting, and isolating resource usage of process collections.