Maximizing CNN accelerator efficiency through resource partitioning

Convolutional neural networks (CNNs) are revolutionizing machine learning, but they present significant computational challenges. Recently, many FPGA-based accelerators have been proposed to improve the performance and efficiency of CNNs. Current approaches construct a single processor that computes...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2017 ACM/IEEE 44th Annual International Symposium on Computer Architecture (ISCA) S. 535 - 547
Hauptverfasser: Yongming Shen, Ferdman, Michael, Milder, Peter
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: ACM 01.06.2017
Schlagworte:
Accelerator
Arrays
Convolutional codes
Convolutional Neural Network
Design methodology
Field programmable gate arrays
FPGA
Hardware
Throughput
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Convolutional neural networks (CNNs) are revolutionizing machine learning, but they present significant computational challenges. Recently, many FPGA-based accelerators have been proposed to improve the performance and efficiency of CNNs. Current approaches construct a single processor that computes the CNN layers one at a time; the processor is optimized to maximize the throughput at which the collection of layers is computed. However, this approach leads to inefficient designs because the same processor structure is used to compute CNN layers of radically varying dimensions. We present a new CNN accelerator paradigm and an accompanying automated design methodology that partitions the available FPGA resources into multiple processors, each of which is tailored for a different subset of the CNN convolutional layers. Using the same FPGA resources as a single large processor, multiple smaller specialized processors increase computational efficiency and lead to a higher overall throughput. Our design methodology achieves 3.8x higher throughput than the state-of-the-art approach on evaluating the popular AlexNet CNN on a Xilinx Virtex-7 FPGA. For the more recent SqueezeNet and GoogLeNet, the speedups are 2.2x and 2.0x.
DOI:10.1145/3079856.3080221