GPU processing clusters
Graphics processing clusters and texture processing clusters are physical groupings of resources inside NVIDIA GPUs. They sit above individual streaming multiprocessors in architecture diagrams.
How it works
The exact grouping and attached graphics resources vary by device. A CUDA programmer usually schedules blocks through a grid rather than selecting a physical TPC or GPC. Supported thread-block clusters add cooperative execution guarantees, but are a separate programming construct.
Work through an example
Use a chip diagram to understand shared resources and the architecture guide to determine legal cooperation. Two boxes drawn near each other do not establish that their blocks may synchronize.
Where the term can mislead
A physical GPC, a CUDA thread-block cluster and a cluster of networked GPU servers describe three different scopes.
Connect the idea
Take it into practice
Continue with the existing concept, apply it to an interview problem, or follow a course lesson. Access requirements are shown before you open a page.
New to GPUs? Start with the free execution-model concept.
Sources and scope
Examples here are teaching calculations, not measured GPU performance. Consult the linked documentation for the exact device, instruction and tool version. Source review: 25 October 2026.