GPU Glossary
GPU Glossary
/device-hardware/load-store-unit
?
Something seem wrong?
Or want to contribute?

Click this button to
let us know on GitHub.

What is a Load/Store Unit?

LSU

The Load/Store Units (LSUs) dispatch requests to load or store data to the memory subsystems of the GPU.

Most importantly for CUDA programmers , they interact directly with the Streaming Multiprocessor 's on-chip SRAM L1 data cache and indirectly with the off-chip, on-device global RAM that respectively implement the lowest and highest levels of the memory hierarchy in the CUDA programming model .

Modal LogoBuilding on GPUs? We know a thing or two about it.

Modal is an ergonomic Python SDK wrapped around a global GPU fleet. Deploy serverless AI workloads instantly without worrying about quota requests, driver compatibility issues, or managing bulky ML dependencies.

Deploy on GPUs