A Highly Parallel DRAM Architecture to Mitigate Large Access Latency and Improve Energy Efficiency of Modern DRAM Systems

Modern Dynamic Random Access Memory (DRAM) banks are characterized by their ability to work in parallel, enabling concurrent servicing of multiple memory accesses through the interleaved DRAM banks. This attractive feature is supported in modern DRAM systems by employing large-sized pages to exploit...

Full description

Saved in:
Bibliographic Details
Main Authors: Tareq A. Alawneh, Ahmed A. M. Sharadqh, Ashraf Al Sharah, Emad Awada, Jawdat S. Alkasassbeh, Ayman Y. Al-Rawashdeh, Aws Al-Qaisi
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10778429/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850126913076461568
author Tareq A. Alawneh
Ahmed A. M. Sharadqh
Ashraf Al Sharah
Emad Awada
Jawdat S. Alkasassbeh
Ayman Y. Al-Rawashdeh
Aws Al-Qaisi
author_facet Tareq A. Alawneh
Ahmed A. M. Sharadqh
Ashraf Al Sharah
Emad Awada
Jawdat S. Alkasassbeh
Ayman Y. Al-Rawashdeh
Aws Al-Qaisi
author_sort Tareq A. Alawneh
collection DOAJ
description Modern Dynamic Random Access Memory (DRAM) banks are characterized by their ability to work in parallel, enabling concurrent servicing of multiple memory accesses through the interleaved DRAM banks. This attractive feature is supported in modern DRAM systems by employing large-sized pages to exploit the locality that exists in the row-buffers. However, one of the key limiting factors for fully utilizing the Bank-Level Parallelism (BLP) feature is the increased contention among processing cores in multi-core systems. This may result in inefficient utilization of the available localities in row-buffers. This, in turn, leads to a degradation in DRAM performance and significant energy wastage. This is due to the activation of large pages only to access a small amount of data, typically 64 bytes. As a result, the likelihood of encountering critical power and performance timing constraints increases. In this article, we propose a highly parallel DRAM architecture designed to mitigate the adverse effects of increased memory contention in multi-core systems on DRAM performance and energy efficiency. Specifically, the DRAM architecture introduced in this study incorporates cost-effective modifications that reduce resource sharing among DRAM pages within a sub-array. This design enables concurrent access to wordlines of varying sizes belonging to different DRAM pages within the same sub-array of a memory bank. This enhancement in the utilization of the allocated bank’s row-buffer space significantly improves the performance and energy efficiency of DRAM systems by further enhancing intra- and inter-sub-array level parallelism. This also further relaxes critical DRAM timing restrictions and facilitates access to finer DRAM page granularities. Our experimental results for quad-core multi-program workloads showed that the DRAM architecture proposed in this study provides significant improvements in average memory access latency and overall DRAM energy consumption compared to the baseline, outperforming previously proposed mechanisms.
format Article
id doaj-art-7e65b2dca40546ca9315845d9e1359e1
institution OA Journals
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-7e65b2dca40546ca9315845d9e1359e12025-08-20T02:33:48ZengIEEEIEEE Access2169-35362024-01-011218299818302310.1109/ACCESS.2024.351217610778429A Highly Parallel DRAM Architecture to Mitigate Large Access Latency and Improve Energy Efficiency of Modern DRAM SystemsTareq A. Alawneh0https://orcid.org/0000-0002-2400-1599Ahmed A. M. Sharadqh1https://orcid.org/0000-0002-5327-0288Ashraf Al Sharah2https://orcid.org/0000-0002-4848-8282Emad Awada3https://orcid.org/0000-0003-3271-7105Jawdat S. Alkasassbeh4https://orcid.org/0000-0003-4573-1328Ayman Y. Al-Rawashdeh5Aws Al-Qaisi6https://orcid.org/0000-0002-1929-5668Electrical Engineering Department, Faculty of Engineering Technology, Al-Balqa Applied University, Amman, JordanElectrical Engineering Department, Faculty of Engineering Technology, Al-Balqa Applied University, Amman, JordanElectrical Engineering Department, Faculty of Engineering Technology, Al-Balqa Applied University, Amman, JordanElectrical Engineering Department, Faculty of Engineering Technology, Al-Balqa Applied University, Amman, JordanElectrical Engineering Department, Faculty of Engineering Technology, Al-Balqa Applied University, Amman, JordanElectrical Engineering Department, Faculty of Engineering Technology, Al-Balqa Applied University, Amman, JordanCollege of Engineering and Technology, American University of the Middle East, Eqaila, KuwaitModern Dynamic Random Access Memory (DRAM) banks are characterized by their ability to work in parallel, enabling concurrent servicing of multiple memory accesses through the interleaved DRAM banks. This attractive feature is supported in modern DRAM systems by employing large-sized pages to exploit the locality that exists in the row-buffers. However, one of the key limiting factors for fully utilizing the Bank-Level Parallelism (BLP) feature is the increased contention among processing cores in multi-core systems. This may result in inefficient utilization of the available localities in row-buffers. This, in turn, leads to a degradation in DRAM performance and significant energy wastage. This is due to the activation of large pages only to access a small amount of data, typically 64 bytes. As a result, the likelihood of encountering critical power and performance timing constraints increases. In this article, we propose a highly parallel DRAM architecture designed to mitigate the adverse effects of increased memory contention in multi-core systems on DRAM performance and energy efficiency. Specifically, the DRAM architecture introduced in this study incorporates cost-effective modifications that reduce resource sharing among DRAM pages within a sub-array. This design enables concurrent access to wordlines of varying sizes belonging to different DRAM pages within the same sub-array of a memory bank. This enhancement in the utilization of the allocated bank’s row-buffer space significantly improves the performance and energy efficiency of DRAM systems by further enhancing intra- and inter-sub-array level parallelism. This also further relaxes critical DRAM timing restrictions and facilitates access to finer DRAM page granularities. Our experimental results for quad-core multi-program workloads showed that the DRAM architecture proposed in this study provides significant improvements in average memory access latency and overall DRAM energy consumption compared to the baseline, outperforming previously proposed mechanisms.https://ieeexplore.ieee.org/document/10778429/DRAM architectureDRAM performancememory interferencerow activationbank prechargepage conflict
spellingShingle Tareq A. Alawneh
Ahmed A. M. Sharadqh
Ashraf Al Sharah
Emad Awada
Jawdat S. Alkasassbeh
Ayman Y. Al-Rawashdeh
Aws Al-Qaisi
A Highly Parallel DRAM Architecture to Mitigate Large Access Latency and Improve Energy Efficiency of Modern DRAM Systems
IEEE Access
DRAM architecture
DRAM performance
memory interference
row activation
bank precharge
page conflict
title A Highly Parallel DRAM Architecture to Mitigate Large Access Latency and Improve Energy Efficiency of Modern DRAM Systems
title_full A Highly Parallel DRAM Architecture to Mitigate Large Access Latency and Improve Energy Efficiency of Modern DRAM Systems
title_fullStr A Highly Parallel DRAM Architecture to Mitigate Large Access Latency and Improve Energy Efficiency of Modern DRAM Systems
title_full_unstemmed A Highly Parallel DRAM Architecture to Mitigate Large Access Latency and Improve Energy Efficiency of Modern DRAM Systems
title_short A Highly Parallel DRAM Architecture to Mitigate Large Access Latency and Improve Energy Efficiency of Modern DRAM Systems
title_sort highly parallel dram architecture to mitigate large access latency and improve energy efficiency of modern dram systems
topic DRAM architecture
DRAM performance
memory interference
row activation
bank precharge
page conflict
url https://ieeexplore.ieee.org/document/10778429/
work_keys_str_mv AT tareqaalawneh ahighlyparalleldramarchitecturetomitigatelargeaccesslatencyandimproveenergyefficiencyofmoderndramsystems
AT ahmedamsharadqh ahighlyparalleldramarchitecturetomitigatelargeaccesslatencyandimproveenergyefficiencyofmoderndramsystems
AT ashrafalsharah ahighlyparalleldramarchitecturetomitigatelargeaccesslatencyandimproveenergyefficiencyofmoderndramsystems
AT emadawada ahighlyparalleldramarchitecturetomitigatelargeaccesslatencyandimproveenergyefficiencyofmoderndramsystems
AT jawdatsalkasassbeh ahighlyparalleldramarchitecturetomitigatelargeaccesslatencyandimproveenergyefficiencyofmoderndramsystems
AT aymanyalrawashdeh ahighlyparalleldramarchitecturetomitigatelargeaccesslatencyandimproveenergyefficiencyofmoderndramsystems
AT awsalqaisi ahighlyparalleldramarchitecturetomitigatelargeaccesslatencyandimproveenergyefficiencyofmoderndramsystems
AT tareqaalawneh highlyparalleldramarchitecturetomitigatelargeaccesslatencyandimproveenergyefficiencyofmoderndramsystems
AT ahmedamsharadqh highlyparalleldramarchitecturetomitigatelargeaccesslatencyandimproveenergyefficiencyofmoderndramsystems
AT ashrafalsharah highlyparalleldramarchitecturetomitigatelargeaccesslatencyandimproveenergyefficiencyofmoderndramsystems
AT emadawada highlyparalleldramarchitecturetomitigatelargeaccesslatencyandimproveenergyefficiencyofmoderndramsystems
AT jawdatsalkasassbeh highlyparalleldramarchitecturetomitigatelargeaccesslatencyandimproveenergyefficiencyofmoderndramsystems
AT aymanyalrawashdeh highlyparalleldramarchitecturetomitigatelargeaccesslatencyandimproveenergyefficiencyofmoderndramsystems
AT awsalqaisi highlyparalleldramarchitecturetomitigatelargeaccesslatencyandimproveenergyefficiencyofmoderndramsystems