JavaScript parece estar deshabilitado en tu navegador. Para la mejor experiencia en nuestro sitio, asegúrate de activar JavaScript en tu navegador.

SKU/Artículo: AMZ-B0G2KFGTKY

FPGAs for Developers: A Software Engineer's Guide to Thinking in Hardware and Building Custom Accelerators.

Format:

Kindle

Hardcover

Kindle

Paperback

Detalles del producto

Disponibilidad:
Fuera de stock

Peso con empaque:
0.94 kg

Devolución:
No

Condición
Nuevo

Producto de:
Amazon

Viaja desde
USA

Sobre este producto

For a long time now, software developers enjoyed a "free lunch." Our code magically became faster with every new CPU generation. That era is over. We've slammed into the "power wall," and the general-purpose CPU, once our greatest ally, is now our biggest bottleneck. When you've optimized your algorithm, refactored your loops, and profiled your code, but your application is still too slow... what do you do next? You stop trying to optimize your software for the hardware. You build new hardware for your software. This book is the bridge for software engineers to the world of custom hardware. It's not about becoming an electrical engineer; it's about leveraging your existing C/C++ skills to build blazingly fast, custom accelerators using High-Level Synthesis (HLS). You'll learn to see FPGAs (Field-Programmable Gate Arrays) not as intimidating, low-level silicon, but as a "blank slate" you can forge into a perfect, parallel machine designed to run your specific algorithm at the speed of light. This book is a journey to "think in hardware" a new mental model that will permanently change how you solve problems. You will learn:The CPU vs. The "Sea of Logic": Deconstruct why a CPU is a sequential, "jack-of-all-trades" and how an FPGA is a parallel, "master-of-one" machine.The Mindset Shift: Master the core hardware patterns. Learn why a software if statement is a physical data selector (a MUX) and why your trusty for loop is a hardware design choice between a time-based FSM, a high-throughput "factory" (Pipeline), or a massive "brute-force" parallel circuit (Unroll).The C++ "Bridge" (HLS): Use High-Level Synthesis to translate your C/C++ functions directly into hardware blueprints. You'll master the "magic wand" pragmas (#pragma HLS PIPELINE, DATAFLOW, UNROLL) to command the compiler to build the exact architecture you've designed in your head.The End-to-End System: Go from a single kernel to a true "System-on-a-Chip" (SoC). You'll learn to run Linux on the built-in ARM "boss" processor while commanding your custom-built "factory" on the FPGA fabric, connecting them with high-speed AXI busses, DMAs, and software drivers.This book is for C++, C, and Python developers, data scientists, and systems engineers who have hit a performance wall and are ready for a new, more powerful tool. By the end of this journey, you won't just be a software developer. You'll be an Accelerator Architect. You will have the rare and valuable skill to look at a "hot loop" in a profiler, make the call to move it to hardware, and personally build the custom circuit that delivers the 100x or 1000x speedup your application demands. It's time to stop waiting for faster CPUs. It's time to build your own.

Fuera de stock

Selecciona otra opción o busca otro producto.

Medios de pago

Aceptamos múltiples medios de pago para tu comodidad

Tarjetas prepagas, débito y crédito