Data Orchestration in Deep Learning Accelerators - Hyoukjun Kwon,Ananda Samajdar,Tushar Krishna,Michael Pellauer,Angshuman Parashar
Hyoukjun Kwon, Ananda Samajdar, Tushar Krishna, Michael Pellauer, Angshuman Parashar
-20% with code BOOKS
Shipping in 12-18 days
30-day return policy
This Synthesis Lecture focuses on techniques for efficient data orchestration within DNN accelerators. The End of Moore's Law, coupled with the increasing growth in deep learning and other AI applications has led to the emergence of custom Deep Neural Network (DNN) accelerators for energy-efficient inference on edge devices. Modern DNNs have millions of hyper parameters and involve billions of computations; ... Full description
You May Also Like
TOP
If Anyone Builds It, Everyone Dies: The Case Against Superintelligent AI
Eliezer Yudkowsky, Nate Soares
-20% with code BOOKS
In stock at our supplier
TOP
The Design of Everyday Things
-20% with code BOOKS
In stock at our supplier
TOP
Open Circuits: The Inner Beauty of Electronic Components
Eric Schlaepfer, Windell Oskay
-20% with code BOOKS
In stock at our supplier
TOP
Get started with MicroPython on Raspberry Pi Pico: The Official Raspberry Pi Pico Guide
-20% with code BOOKS
In stock at our supplier
TOP
Practical Electronics for Inventors
-20% with code BOOKS
In stock at our supplier
TOP
Nucleo Boards Programming with the STM32CubeIDE: Hands-on in more than 50 projects
-20% with code BOOKS
In stock at our supplier
TOP
Data-Driven Science and Engineering: Machine Learning, Dynamical Systems, and Control
Steven L. Brunton, J. Nathan Kutz
-20% with code BOOKS
In stock at our supplier
TOP
Fundamentals of Aerodynamics ISE
Christopher Cadou, John D. Anderson
-20% with code BOOKS
In stock at our supplier
TOP
Black Holes: The Key to Understanding the Universe
-20% with code BOOKS
In stock at our supplier
TOP
KiCad Like A Pro - Fundamentals and Projects: Getting started with the world's best open-source PCB tool
-20% with code BOOKS
In stock at our supplier
TOP
Foundations of Flavor: The Noma Guide to Fermentation: Including Step-By-Step Information on Making and Cooking with: Koji, Kombuchas, Shoyus, Misos, Vinegars, Garums, Lacto-Ferments and black Fruits Vegetables
-20% with code BOOKS
In stock at our supplier
TOP
Make: Electronics: Learning by Discovery: A hands-on primer for the new electronics enthusiast
-20% with code BOOKS
In stock at our supplier
TOP
The Idea Factory: Bell Labs and the Great Age of American Innovation
-20% with code BOOKS
In stock at our supplier
TOP
Flowers For Algernon: A Modern Literary Classic
-20% with code BOOKS
In stock at our supplier
TOP
Professional SMD Assembly: Practical Low-Cost Methods for Reliable PCB Production
-20% with code BOOKS
In stock at our supplier
TOP
Blood in the Machine: The Origins of the Rebellion Against Big Tech
-20% with code BOOKS
In stock at our supplier
TOP
Learning the Art of Electronics: A Hands-On Lab Course
Thomas C. Hayes, Paul Horowitz, David Abrams
-20% with code BOOKS
In stock at our supplier
TOP
Raspberry Pi 5 for Radio Amateurs: Program and Build Raspberry Pi 5 Based Ham Station Utilities with the RTL-SDR
-20% with code BOOKS
In stock at our supplier
Description
This Synthesis Lecture focuses on techniques for efficient data orchestration within DNN accelerators. The End of Moore's Law, coupled with the increasing growth in deep learning and other AI applications has led to the emergence of custom Deep Neural Network (DNN) accelerators for energy-efficient inference on edge devices. Modern DNNs have millions of hyper parameters and involve billions of computations; this necessitates extensive data movement from memory to on-chip processing engines. It is well known that the cost of data movement today surpasses the cost of the actual computation; therefore, DNN accelerators require careful orchestration of data across on-chip compute, network, and memory elements to minimize the number of accesses to external DRAM. The book covers DNN dataflows, data reuse, buffer hierarchies, networks-on-chip, and automated design-space exploration. It concludes with data orchestration challenges with compressed and sparse DNNs and future trends. The target audience is students, engineers, and researchers interested in designing high-performance and low-energy accelerators for DNN inference.
More Information
| Author | Hyoukjun Kwon, Ananda Samajdar, Tushar Krishna, Michael Pellauer, Angshuman Parashar |
|---|---|
| Publisher | Springer International Publishing |
| Series | Synthesis Lectures on Computer Architecture |
| Release year | 2020 |
| Cover type | Softcover |
| EAN | 9783031006395 |