High-Performance Data Mapping for BNNs on PCM-based Integrated Photonics

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shahroodi, Taha, Cardoso, Raphael, Wong, Stephan, Bosio, Alberto, O'Connor, Ian, Hamdioui, Said
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929230226391040
author Shahroodi, Taha
Cardoso, Raphael
Wong, Stephan
Bosio, Alberto
O'Connor, Ian
Hamdioui, Said
author_facet Shahroodi, Taha
Cardoso, Raphael
Wong, Stephan
Bosio, Alberto
O'Connor, Ian
Hamdioui, Said
contents State-of-the-Art (SotA) hardware implementations of Deep Neural Networks (DNNs) incur high latencies and costs. Binary Neural Networks (BNNs) are potential alternative solutions to realize faster implementations without losing accuracy. In this paper, we first present a new data mapping, called TacitMap, suited for BNNs implemented based on a Computation-In-Memory (CIM) architecture. TacitMap maximizes the use of available parallelism, while CIM architecture eliminates the data movement overhead. We then propose a hardware accelerator based on optical phase change memory (oPCM) called EinsteinBarrier. Ein-steinBarrier incorporates TacitMap and adds an extra dimension for parallelism through wavelength division multiplexing, leading to extra latency reduction. The simulation results show that, compared to the SotA CIM baseline, TacitMap and EinsteinBarrier significantly improve execution time by up to ~154x and ~3113x, respectively, while also maintaining the energy consumption within 60% of that in the CIM baseline.
format Preprint
id arxiv_https___arxiv_org_abs_2401_17724
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle High-Performance Data Mapping for BNNs on PCM-based Integrated Photonics
Shahroodi, Taha
Cardoso, Raphael
Wong, Stephan
Bosio, Alberto
O'Connor, Ian
Hamdioui, Said
Hardware Architecture
Emerging Technologies
State-of-the-Art (SotA) hardware implementations of Deep Neural Networks (DNNs) incur high latencies and costs. Binary Neural Networks (BNNs) are potential alternative solutions to realize faster implementations without losing accuracy. In this paper, we first present a new data mapping, called TacitMap, suited for BNNs implemented based on a Computation-In-Memory (CIM) architecture. TacitMap maximizes the use of available parallelism, while CIM architecture eliminates the data movement overhead. We then propose a hardware accelerator based on optical phase change memory (oPCM) called EinsteinBarrier. Ein-steinBarrier incorporates TacitMap and adds an extra dimension for parallelism through wavelength division multiplexing, leading to extra latency reduction. The simulation results show that, compared to the SotA CIM baseline, TacitMap and EinsteinBarrier significantly improve execution time by up to ~154x and ~3113x, respectively, while also maintaining the energy consumption within 60% of that in the CIM baseline.
title High-Performance Data Mapping for BNNs on PCM-based Integrated Photonics
topic Hardware Architecture
Emerging Technologies
url https://arxiv.org/abs/2401.17724