NID: Processing binary convolutional neural network in commodity DRAM

Jaehyeong Sim, Hoseok Seol, Lee Sup Kim

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

6 Scopus citations

Abstract

Recent large-scale CNNs suffer from a severe memory wall problem as their number of weights range from tens to hundreds of millions. Processing in-memory (PIM) and binary CNN have been proposed to alleviate the number of memory accesses and footprints, respectively. By combining the two separate concepts, we propose a novel processing in-DRAM framework for binary CNN, called NID, where dominant convolution operations are processed using in-DRAM bulk bitwise operations. We first identify the problem that the bitcount operations with only bulk bitwise AND/OR/NOT incur significant overhead in terms of delay when the size of kernels gets larger. Then, we not only optimize the performance by efficiently allocating inputs and kernels to DRAM banks for both convolutional and fully-connected layers through design space explorations, but also mitigate the overhead of bitcount operations by splitting kernels into multiple parts. Partial sum accumulations and tasks of the other layers such as max-pooling and normalization layers are processed in the peripheral area of DRAM with negligible overheads. In results, our NID framework achieves 19X-36X performance and 9X-14X EDP improvements for convolutional layers, and 9X-17X performance and 1.4X-4.5X EDP improvements for fully-connected layers over previous PIM technique in four large-scale CNN models.

Original languageEnglish
Title of host publication2018 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2018 - Digest of Technical Papers
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781450359504
DOIs
StatePublished - 5 Nov 2018
Event37th IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2018 - San Diego, United States
Duration: 5 Nov 20188 Nov 2018

Publication series

NameIEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD
ISSN (Print)1092-3152

Conference

Conference37th IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2018
Country/TerritoryUnited States
CitySan Diego
Period5/11/188/11/18

Fingerprint

Dive into the research topics of 'NID: Processing binary convolutional neural network in commodity DRAM'. Together they form a unique fingerprint.

Cite this