skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Design Space Exploration of Emerging Memory Technologies for Machine Learning Applications

Abstract

Memory design space exploration methods study memory systems’ performances and limitations before implementation. The computer memory design space has grown exponentially because of the enormous growth of memory types, memory controllers, and application software. Computer simulators are commonly used for memory design space exploration. However, complex memory simulations take an enormous amount of time. Hence, in this paper, we proposed a machine learning-based design space exploration method for dynamic random-access memory and non-volatile memory systems. We applied our method to the CosmoGAN and LeNet applications to predict the following six memory response parameters: (i) bandwidth, (ii) power, (iii) average latency, (iv) average total latency, (v) memory reads, and (vi) memory writes. Our experimental results show that machine learning models can predict memory response parameter values faster than simulations. We used support vector machine, random forest, and gradient boosting machine learning models. We observed that the support vector machine provides better performance for bandwidth, average latency, and average total latency. The random forest model works better for memory reads and writes. The gradient boosting model provides superior prediction performance for power. We provide a detailed discussion on learning curve characteristics, error analysis, and memory type recommendation.

Authors:
 [1]; ORCiD logo [1]; ORCiD logo [1]; ORCiD logo [1];  [1]
  1. ORNL
Publication Date:
Research Org.:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1807257
DOE Contract Number:  
AC05-00OR22725
Resource Type:
Conference
Resource Relation:
Conference: IEEE IPDPS 2021 - The Eleventh International Workshop on Accelerators and Hybrid Emerging Systems (AsHES) - Portland, Oregon, United States of America - 5/17/2021 4:00:00 AM-5/21/2021 4:00:00 AM
Country of Publication:
United States
Language:
English

Citation Formats

Hasan, S M Shamimul, Imam, Neena, Kannan, Ramakrishnan, Yoginath, Srikanth, and Kurte, Kuldeep. Design Space Exploration of Emerging Memory Technologies for Machine Learning Applications. United States: N. p., 2021. Web.
Hasan, S M Shamimul, Imam, Neena, Kannan, Ramakrishnan, Yoginath, Srikanth, & Kurte, Kuldeep. Design Space Exploration of Emerging Memory Technologies for Machine Learning Applications. United States.
Hasan, S M Shamimul, Imam, Neena, Kannan, Ramakrishnan, Yoginath, Srikanth, and Kurte, Kuldeep. 2021. "Design Space Exploration of Emerging Memory Technologies for Machine Learning Applications". United States. https://www.osti.gov/servlets/purl/1807257.
@article{osti_1807257,
title = {Design Space Exploration of Emerging Memory Technologies for Machine Learning Applications},
author = {Hasan, S M Shamimul and Imam, Neena and Kannan, Ramakrishnan and Yoginath, Srikanth and Kurte, Kuldeep},
abstractNote = {Memory design space exploration methods study memory systems’ performances and limitations before implementation. The computer memory design space has grown exponentially because of the enormous growth of memory types, memory controllers, and application software. Computer simulators are commonly used for memory design space exploration. However, complex memory simulations take an enormous amount of time. Hence, in this paper, we proposed a machine learning-based design space exploration method for dynamic random-access memory and non-volatile memory systems. We applied our method to the CosmoGAN and LeNet applications to predict the following six memory response parameters: (i) bandwidth, (ii) power, (iii) average latency, (iv) average total latency, (v) memory reads, and (vi) memory writes. Our experimental results show that machine learning models can predict memory response parameter values faster than simulations. We used support vector machine, random forest, and gradient boosting machine learning models. We observed that the support vector machine provides better performance for bandwidth, average latency, and average total latency. The random forest model works better for memory reads and writes. The gradient boosting model provides superior prediction performance for power. We provide a detailed discussion on learning curve characteristics, error analysis, and memory type recommendation.},
doi = {},
url = {https://www.osti.gov/biblio/1807257}, journal = {},
number = ,
volume = ,
place = {United States},
year = {2021},
month = {5}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share: