Systems and efficient execution · MLD-0188
Compute kernel
A low-level routine executed across processor threads to perform an operation.
Original reference page · detailed expansion pending. Original definitions, examples, notes and citations are preserved below. New examples are illustrative, not benchmark results.
Original example
Implement a matrix multiplication on a GPU.
Original caveat
A compute kernel is not a kernel-method similarity function.
Context and aliases
Level: Advanced. Stages: Training · Inference · Deployment.
Tags: DL · LLM · SLM · Systems.
Aliases: GPU kernel.
Related concepts
Concepts linking here
Original references
- PyTorch Fully Sharded Data Parallel ↗ · Consulted
- PyTorch documentation ↗ · Reference link
- vLLM documentation ↗ · Consulted
- Original reference URL ↗