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

Conference Paper (2024)
Author(s)

Taha Shahroodi (TU Delft - Computer Engineering)

Raphael Cardoso (École Centrale de Lyon)

Stephan Wong (TU Delft - Computer Engineering)

A. Bosio (École Centrale de Lyon)

Ian O'Connor (École Centrale de Lyon)

Said Hamdioui (TU Delft - Computer Engineering)

Research Group
Computer Engineering
More Info
expand_more
Publication Year
2024
Language
English
Research Group
Computer Engineering
ISBN (print)
979-8-3503-4860-6
ISBN (electronic)
978-3-9819263-8-5
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

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 ∼ 154× and ∼ 3113×, respectively, while also maintaining the energy consumption within 60% of that in the CIM baseline.

Files

High-Performance_Data_Mapping_... (pdf)
(pdf | 1.22 Mb)
- Embargo expired in 10-12-2024
License info not available