Memory Access Characteristics of Neural Network Workloads and Their Implications

Soyeon Park, Hyokyung Bahn

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

9 Scopus citations

Abstract

With the recent advances in machine learning and many-core computing technologies, neural networks are widely used in various service domains of the 4th industrial revolution. As the data set of neural network is becoming increasingly large, it is important to analyze the memory access characteristics of neural network workloads. In this paper, we perform a comprehensive analysis of memory access behaviors in four types of neural network configurations, i.e., CNN (convolutional neural networks), RNN (recurrent neural networks), DNN (deep neural networks), and ANN (artificial neural networks). From this analysis, we observe the following characteristics, which are quite different from traditional desktop and smartphone memory accesses. First, we analyze the access bias of memory locations and find that most memory accesses occur in a certain limited memory locations. Second, the identity of these hot locations is the data and heap regions, which account for over 90% of total memory accesses. Third, the bias of memory access in neural network workloads is relatively weaker than other desktop or smartphone workloads, specially for write operations. Fourth, write operations account for about twice of read operations regardless of neural network types. Fifth, in predicting re-access likelihood, temporal locality provides better information than access frequency in read operations, but combining the two properties is necessary for accurate estimation in write operations. We anticipate that the analysis of this study will be a good guideline for designing an efficient memory system for neural network workloads.

Original languageEnglish
Title of host publicationProceedings of IEEE Asia-Pacific Conference on Computer Science and Data Engineering, CSDE 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665453059
DOIs
StatePublished - 2022
Event2022 IEEE Asia-Pacific Conference on Computer Science and Data Engineering, CSDE 2022 - Gold Coast, Australia
Duration: 18 Dec 202220 Dec 2022

Publication series

NameProceedings of IEEE Asia-Pacific Conference on Computer Science and Data Engineering, CSDE 2022

Conference

Conference2022 IEEE Asia-Pacific Conference on Computer Science and Data Engineering, CSDE 2022
Country/TerritoryAustralia
CityGold Coast
Period18/12/2220/12/22

Bibliographical note

Funding Information:
ACKNOWLEDGMENT This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No. 2019R1A2C1009275) and also by the ICT R&D program of MSIT/IITP (2020-0-00121, development of data improvement and dataset correction technology based on data quality assessment).

Publisher Copyright:
© 2022 IEEE.

Keywords

  • access bias
  • artificial intelligence
  • machine learning
  • memory access characteristics
  • neural network

Fingerprint

Dive into the research topics of 'Memory Access Characteristics of Neural Network Workloads and Their Implications'. Together they form a unique fingerprint.

Cite this