GPU kernel
A GPU kernel is a program or function executed on a graphics processing unit to perform a computation, often in parallel across many threads.
GPU Programming with Triton
A GPU kernel is a function launched on a GPU to perform computation in parallel.
- How it is used
- Uses custom GPU kernels to accelerate specialized operations or fuse work when framework-level optimization is insufficient.
- How deep it goes
- Explains launch, execution hierarchy, and memory roles, then works through vector addition and two-dimensional tile addressing.
- Examples
- Demonstrates vector addition with a host wrapper and masked accesses, and extends block indexing to matrix tiles.
- What you can do after
- Apply GPU kernels via Triton to implement and launch vector addition and reason about two-dimensional tile indexing.
LLMs in Production
A kernel is a small program or function executed on a parallel processor to perform a computational operation.
- How it is used
- Supports model compilation and deployment tuning by mapping neural-network operations onto efficient processor routines.
- How deep it goes
- Explains kernel selection and tuning factors, but stops short of implementation code or detailed hardware execution mechanics.
- Examples
- Uses convolution algorithms and fused convolution-bias-ReLU operations to illustrate alternative kernels and graph optimization.
- What you can do after
- Evaluate/understand how hardware and tensor characteristics influence kernel selection and deployed model performance.
Fast Python: High performance techniques for large datasets
- How it is used
- Uses a GPU entry point to distribute data-processing computations across threads.
- How deep it goes
- Explains thread positions, launch configuration, and bounds checks, while deferring shared-memory algorithms as more advanced.
- Examples
- Demonstrates doubling array elements and profiling Mandelbrot implementations.
- What you can do after
- Apply it via Numba's CUDA launch syntax to update array elements across GPU threads.
Read how Fast Python: High performance techniques for large datasets explains GPU kernel →
Deep Learning with JAX
A compiled low-level executable unit for performing an operation on target hardware, not a convolution kernel.
- How it is used
- Uses weighted matrices to blur, denoise, or sharpen images through convolution.
- How deep it goes
- Explains windowed multiplication, summation, channel handling, and clipping, but does not cover compiled executable kernels.
- Examples
- Demonstrates blur, Gaussian denoising, sharpening, per-channel color convolution, and comparisons of noisy, blurred, restored, and original images.
- What you can do after
- Apply it via 2D convolution to construct blur, denoising, and sharpening image filters.
What to know first
- threadA thread is a distinct sequence of execution with its own control flow and state, managed or scheduled by a processor, runtime, or operating system.
- CPUA central processing unit (CPU) is a computer processor that executes program instructions and performs general-purpose computational operations.
- CUDACUDA is NVIDIA’s parallel computing platform and programming model for performing general-purpose computations on NVIDIA GPUs.
- PythonPython is a high-level, general-purpose, dynamically typed programming language with readable syntax and support for multiple programming paradigms.
Related concepts
- kernel fusionproduced by · Kernel fusion is an optimization that combines multiple computational operations into fewer execution units or passes, reducing overhead and often intermediate data movement.
- entry pointkind of · An entry point is a designated starting location or interface through which a program, component, process, or interaction begins.
- PyTorchuses · PyTorch is an open-source machine-learning framework that provides tensor computation, automatic differentiation, and tools for developing and training neural networks.
- Tritonproduced by · Triton is a programming language and compiler for writing and compiling high-performance GPU kernels.
- maskuses · A mask is a representation or mechanism that selectively identifies, includes, excludes, reveals, or obscures specified parts of data, a system, or a computation.
Questions about GPU kernel
What is GPU kernel?
A GPU kernel is a program or function executed on a graphics processing unit to perform a computation, often in parallel across many threads.
Which Manning books explain GPU kernel?
GPU Programming with Triton, LLMs in Production, Fast Python: High performance techniques for large datasets and Deep Learning with JAX.
What should I know before learning GPU kernel?
Manning books that teach GPU kernel usually introduce thread, CPU, CUDA and Python first.
What is the difference between GPU kernel and device function?
A GPU kernel is a program or function executed on a graphics processing unit to perform a computation, often in parallel across many threads. A device function is a GPU-compiled function that is callable from device-side code rather than directly launched from the host.
What is GPU kernel a kind of?
GPU kernel is a kind of entry point. An entry point is a designated starting location or interface through which a program, component, process, or interaction begins.
What is GPU kernel also called?
GPU kernel is also called CUDA kernel or kernel function.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.



