Isbn: 9798180943231 - ai hardware engineering: designing gpus, tpus, and neural processing units for high-throughput machine learning workloads (5 risultati)

PAP. Condizione: New. New Book. Shipped from UK. Established seller since 2000.

PAP. Condizione: New. New Book. Shipped from UK. Established seller since 2000.

Paperback. Condizione: new. Paperback. Master the Architecture of Next-Generation AI SupercomputingAs deep learning models scale to hundreds of billions of parameters, the hardware executing these workloads has become the ultimate bottleneck. AI Hardware Engineering is the definitive engineering reference designed to bridge the gap between machine learning software and physical silicon. Written for hardware designers, system architects, VLSI engineers, and compiler developers, this book details the microarchitecture of the chips powering the AI revolution.This textbook provides a step-by-step deep dive into custom silicon design, from initial workload characterization to physical tape-out and validation. Readers will gain structural understanding of how mathematical primitives map directly to silicon gates, allowing them to optimize performance-per-watt metrics for massive language models and computer vision pipelines alike.What You Will Master: GPU Microarchitecture: Deep-dive into Streaming Multiprocessors, warp scheduling, tensor cores, and advanced register file architectures.Systolic Arrays & TPUs: Designing high-throughput matrix-multiplication engines, Google TPU architectures, and XLA compilation workflows.Edge NPUs: Architecting low-power neural processing units, Apple Neural Engine mechanics, and mobile accelerator bottlenecks.Advanced Memory Systems: Mastering HBM3, 3D-stacked DRAM, scratchpad SRAM, and high-performance Network-on-Chip (NoC) layouts.Hardware-Compiler Co-Design: Leveraging the MLIR compiler framework, CUDA software layers, and intermediate representations.Future Paradigms: Investigating wafer-scale engines, optical silicon photonics, and analog in-memory computing.Whether you are designing custom ASICs at a semiconductor giant, building custom hardware for hyper-scalers, or writing low-level compiler passes, this book delivers the mathematical rigor, architectural block diagrams, and performance roofline models required to build competitive ML accelerators. Empower your engineering career and bridge the gap between software and silicon. This item is printed on demand. Shipping may be from multiple locations in the US or from the UK, depending on stock availability.…

Condizione: New. Print on Demand.

Paperback. Condizione: new. Paperback. Master the Architecture of Next-Generation AI SupercomputingAs deep learning models scale to hundreds of billions of parameters, the hardware executing these workloads has become the ultimate bottleneck. AI Hardware Engineering is the definitive engineering reference designed to bridge the gap between machine learning software and physical silicon. Written for hardware designers, system architects, VLSI engineers, and compiler developers, this book details the microarchitecture of the chips powering the AI revolution.This textbook provides a step-by-step deep dive into custom silicon design, from initial workload characterization to physical tape-out and validation. Readers will gain structural understanding of how mathematical primitives map directly to silicon gates, allowing them to optimize performance-per-watt metrics for massive language models and computer vision pipelines alike.What You Will Master: GPU Microarchitecture: Deep-dive into Streaming Multiprocessors, warp scheduling, tensor cores, and advanced register file architectures.Systolic Arrays & TPUs: Designing high-throughput matrix-multiplication engines, Google TPU architectures, and XLA compilation workflows.Edge NPUs: Architecting low-power neural processing units, Apple Neural Engine mechanics, and mobile accelerator bottlenecks.Advanced Memory Systems: Mastering HBM3, 3D-stacked DRAM, scratchpad SRAM, and high-performance Network-on-Chip (NoC) layouts.Hardware-Compiler Co-Design: Leveraging the MLIR compiler framework, CUDA software layers, and intermediate representations.Future Paradigms: Investigating wafer-scale engines, optical silicon photonics, and analog in-memory computing.Whether you are designing custom ASICs at a semiconductor giant, building custom hardware for hyper-scalers, or writing low-level compiler passes, this book delivers the mathematical rigor, architectural block diagrams, and performance roofline models required to build competitive ML accelerators. Empower your engineering career and bridge the gap between software and silicon. This item is printed on demand. Shipping may be from our UK warehouse or from our Australian or US warehouses, depending on stock availability.…