Optimizing data allocation for loops on embedded systems with scratch-pad memory

Jun Zhang; Tan Deng; Qiuyan Gao; Qingfeng Zhuge; Edwin H.M. Sha

doi:10.1109/RTCSA.2012.20

Optimizing data allocation for loops on embedded systems with scratch-pad memory

Jun Zhang^*
, Tan Deng
, Qiuyan Gao
, Qingfeng Zhuge
, Edwin H.M. Sha

^*Corresponding author for this work

Research output: Contribution to conference › Paper › peer-review

9 Scopus citations

Abstract

Scratch Pad Memory (SPM), a software-controlled on-chip memory, is popular in embedded systems due to its many benefits. To efficiently manage SPM, many different data allocation algorithms are proposed. However, most of them cannot achieve optimal results. In this paper, we proposed a dynamic programming approach, Iterational Optimal Data Allocation (IODA) to allocate data for embedded systems with multiple types of memory units. According to the experimental results, the IODA algorithm lowered the energy consumption by 20.14% and 5.11% compared to a random memory allocation and a greedy algorithm, respectively. It also reduced the memory access time by 18.44% and 5.83% compared to a random memory allocation and a greedy algorithm, respectively.

Original language	English
Pages	184-191
Number of pages	8
DOIs	https://doi.org/10.1109/RTCSA.2012.20
State	Published - 2012
Externally published	Yes
Event	18th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2012 - Seoul, Korea, Republic of Duration: 19 Aug 2012 → 22 Aug 2012

Conference

Conference	18th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2012
Country/Territory	Korea, Republic of
City	Seoul
Period	19/08/12 → 22/08/12

Keywords

Data allocation
Loops
Optimization
Scratch-pad memory

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/RTCSA.2012.20

Cite this

@conference{9999ba4ee1c9473b96c4345d05c43a29,

title = "Optimizing data allocation for loops on embedded systems with scratch-pad memory",

abstract = "Scratch Pad Memory (SPM), a software-controlled on-chip memory, is popular in embedded systems due to its many benefits. To efficiently manage SPM, many different data allocation algorithms are proposed. However, most of them cannot achieve optimal results. In this paper, we proposed a dynamic programming approach, Iterational Optimal Data Allocation (IODA) to allocate data for embedded systems with multiple types of memory units. According to the experimental results, the IODA algorithm lowered the energy consumption by 20.14\% and 5.11\% compared to a random memory allocation and a greedy algorithm, respectively. It also reduced the memory access time by 18.44\% and 5.83\% compared to a random memory allocation and a greedy algorithm, respectively.",

keywords = "Data allocation, Loops, Optimization, Scratch-pad memory",

author = "Jun Zhang and Tan Deng and Qiuyan Gao and Qingfeng Zhuge and Sha, \{Edwin H.M.\}",

year = "2012",

doi = "10.1109/RTCSA.2012.20",

language = "英语",

pages = "184--191",

note = "18th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2012 ; Conference date: 19-08-2012 Through 22-08-2012",

}

TY - CONF

T1 - Optimizing data allocation for loops on embedded systems with scratch-pad memory

AU - Zhang, Jun

AU - Deng, Tan

AU - Gao, Qiuyan

AU - Zhuge, Qingfeng

AU - Sha, Edwin H.M.

PY - 2012

Y1 - 2012

N2 - Scratch Pad Memory (SPM), a software-controlled on-chip memory, is popular in embedded systems due to its many benefits. To efficiently manage SPM, many different data allocation algorithms are proposed. However, most of them cannot achieve optimal results. In this paper, we proposed a dynamic programming approach, Iterational Optimal Data Allocation (IODA) to allocate data for embedded systems with multiple types of memory units. According to the experimental results, the IODA algorithm lowered the energy consumption by 20.14% and 5.11% compared to a random memory allocation and a greedy algorithm, respectively. It also reduced the memory access time by 18.44% and 5.83% compared to a random memory allocation and a greedy algorithm, respectively.

AB - Scratch Pad Memory (SPM), a software-controlled on-chip memory, is popular in embedded systems due to its many benefits. To efficiently manage SPM, many different data allocation algorithms are proposed. However, most of them cannot achieve optimal results. In this paper, we proposed a dynamic programming approach, Iterational Optimal Data Allocation (IODA) to allocate data for embedded systems with multiple types of memory units. According to the experimental results, the IODA algorithm lowered the energy consumption by 20.14% and 5.11% compared to a random memory allocation and a greedy algorithm, respectively. It also reduced the memory access time by 18.44% and 5.83% compared to a random memory allocation and a greedy algorithm, respectively.

KW - Data allocation

KW - Loops

KW - Optimization

KW - Scratch-pad memory

UR - https://www.scopus.com/pages/publications/84869078607

U2 - 10.1109/RTCSA.2012.20

DO - 10.1109/RTCSA.2012.20

M3 - 论文

AN - SCOPUS:84869078607

SP - 184

EP - 191

T2 - 18th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2012

Y2 - 19 August 2012 through 22 August 2012

ER -

Optimizing data allocation for loops on embedded systems with scratch-pad memory

Abstract

Conference

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this