Dernière sortie
GPU Programming using Rust and CUDA
C++ has been the go-to for GPU programming for almost 20 years. Can Rust do the job, and how well?This book is all about getting hands-on with different toolchains that connect Rust to NVIDIA hardware. There's RustaCUDA for safe host-side control, the Rust-CUDA project for writing kernels in pure Rust, and NVIDIA's experimental cuda-oxide compiler with its typed launches and async execution graphs.
We're going to build one Cargo workspace that keeps on growing. It'll include device queries, launch planning, Rust-written kernels, memory optimization, parallel reductions and scans, multi-stream pipelines, matrix multiplication benchmarked against cuBLAS, a Monte Carlo option pricer validated against a closed formula, and a complete batched inference application measured against a Python baseline.
We'll check every result against a CPU reference, and the reports will give accurate numbers, including where libraries outperform hand-written kernels and where experimental toolchains are still a work in progress. Key LearningsLaunch, synchronize, and verify GPU kernels with ownership-managed device memory. Write real CUDA kernels using Rust-CUDA and cuda-oxide. Plan grids, blocks, and warps for 2D workloads.
Accelerate transfer speeds with pinned memory and coalesced access patterns. Build race-free thread cooperation using shared memory, barriers, and atomics. Overlap transfers with computation using streams, events, and async Rust pipelines. Optimize matrix multiplication and benchmark against cuBLAS ceiling. Wrap CUDA C library safely with handles, error enums, and Drop. Ship complete batched GPU inference application against Python baselines.
Diagnose performance with Nsight Systems, Nsight Compute, and compute-sanitizer. Table of ContentNew Beneficiary of GPU ComputingThinking in ThreadsCommanding GPUWriting GPU KernelsCleaner Kernels with cuda-oxideMastering GPU MemoryMaking Threads CooperateKeeping GPU BusyDelivering Real MathBorrowing NVIDIA's MuscleShipping Complete GPU ApplicationProving Performance
We're going to build one Cargo workspace that keeps on growing. It'll include device queries, launch planning, Rust-written kernels, memory optimization, parallel reductions and scans, multi-stream pipelines, matrix multiplication benchmarked against cuBLAS, a Monte Carlo option pricer validated against a closed formula, and a complete batched inference application measured against a Python baseline.
We'll check every result against a CPU reference, and the reports will give accurate numbers, including where libraries outperform hand-written kernels and where experimental toolchains are still a work in progress. Key LearningsLaunch, synchronize, and verify GPU kernels with ownership-managed device memory. Write real CUDA kernels using Rust-CUDA and cuda-oxide. Plan grids, blocks, and warps for 2D workloads.
Accelerate transfer speeds with pinned memory and coalesced access patterns. Build race-free thread cooperation using shared memory, barriers, and atomics. Overlap transfers with computation using streams, events, and async Rust pipelines. Optimize matrix multiplication and benchmark against cuBLAS ceiling. Wrap CUDA C library safely with handles, error enums, and Drop. Ship complete batched GPU inference application against Python baselines.
Diagnose performance with Nsight Systems, Nsight Compute, and compute-sanitizer. Table of ContentNew Beneficiary of GPU ComputingThinking in ThreadsCommanding GPUWriting GPU KernelsCleaner Kernels with cuda-oxideMastering GPU MemoryMaking Threads CooperateKeeping GPU BusyDelivering Real MathBorrowing NVIDIA's MuscleShipping Complete GPU ApplicationProving Performance
Les livres de Maris Fenlor
25,49 €