Dynamic machine learning based matching of nonvolatile processor microarchitecture to harvested energy profile

Kaisheng Ma; Xueqing Li; Yongpan Liu; John Sampson; Yuan Xie; Vijaykrishnan Narayanan

doi:10.1109/ICCAD.2015.7372634

Dynamic machine learning based matching of nonvolatile processor microarchitecture to harvested energy profile

Kaisheng Ma, Xueqing Li, Yongpan Liu, John Sampson, Yuan Xie, Vijaykrishnan Narayanan

Computer Science and Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

25 Scopus citations

Abstract

Energy harvesting systems without an energy storage device have to efficiently harness the fluctuating and weak power sources to ensure the maximum computational progress. While a simpler processor enables a higher turn-on potential with a weak source, a more powerful processor can utilize more energy that is harvested. Earlier work shows that different complexity levels of nonvolatile microarchitectures provide best fit for different power sources, and even different trails within same power source. In this work, we propose a dynamic nonvolatile microarchitecture by integrating all non-pipelined (NP), N-stage-pipeline (NSP), and Out of Order (OoO) cores together. Neural network machine learning algorithms are also integrated to dynamically adjust the microarchitecture to achieve the maximum forward progress. This integrated solution can achieve forward progress equal to 2.4× of the baseline NP architecture (1.82× of an OoO core).

Original language	English (US)
Title of host publication	2015 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	670-675
Number of pages	6
ISBN (Electronic)	9781467383882
DOIs	https://doi.org/10.1109/ICCAD.2015.7372634
State	Published - Jan 5 2016
Event	34th IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015 - Austin, United States Duration: Nov 2 2015 → Nov 6 2015

Publication series

Name	2015 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015

Other

Other	34th IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015
Country/Territory	United States
City	Austin
Period	11/2/15 → 11/6/15

All Science Journal Classification (ASJC) codes

Computer Graphics and Computer-Aided Design

Access to Document

10.1109/ICCAD.2015.7372634

Cite this

Ma, K., Li, X., Liu, Y., Sampson, J., Xie, Y., & Narayanan, V. (2016). Dynamic machine learning based matching of nonvolatile processor microarchitecture to harvested energy profile. In 2015 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015 (pp. 670-675). Article 7372634 (2015 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCAD.2015.7372634

Ma, Kaisheng ; Li, Xueqing ; Liu, Yongpan et al. / Dynamic machine learning based matching of nonvolatile processor microarchitecture to harvested energy profile. 2015 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 670-675 (2015 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015).

@inproceedings{de200d99207f4c648b539bf52ee8d60d,

title = "Dynamic machine learning based matching of nonvolatile processor microarchitecture to harvested energy profile",

abstract = "Energy harvesting systems without an energy storage device have to efficiently harness the fluctuating and weak power sources to ensure the maximum computational progress. While a simpler processor enables a higher turn-on potential with a weak source, a more powerful processor can utilize more energy that is harvested. Earlier work shows that different complexity levels of nonvolatile microarchitectures provide best fit for different power sources, and even different trails within same power source. In this work, we propose a dynamic nonvolatile microarchitecture by integrating all non-pipelined (NP), N-stage-pipeline (NSP), and Out of Order (OoO) cores together. Neural network machine learning algorithms are also integrated to dynamically adjust the microarchitecture to achieve the maximum forward progress. This integrated solution can achieve forward progress equal to 2.4× of the baseline NP architecture (1.82× of an OoO core).",

author = "Kaisheng Ma and Xueqing Li and Yongpan Liu and John Sampson and Yuan Xie and Vijaykrishnan Narayanan",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; 34th IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015 ; Conference date: 02-11-2015 Through 06-11-2015",

year = "2016",

month = jan,

day = "5",

doi = "10.1109/ICCAD.2015.7372634",

language = "English (US)",

series = "2015 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "670--675",

booktitle = "2015 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015",

address = "United States",

}

Ma, K, Li, X, Liu, Y, Sampson, J, Xie, Y & Narayanan, V 2016, Dynamic machine learning based matching of nonvolatile processor microarchitecture to harvested energy profile. in 2015 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015., 7372634, 2015 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015, Institute of Electrical and Electronics Engineers Inc., pp. 670-675, 34th IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015, Austin, United States, 11/2/15. https://doi.org/10.1109/ICCAD.2015.7372634

Dynamic machine learning based matching of nonvolatile processor microarchitecture to harvested energy profile. / Ma, Kaisheng; Li, Xueqing; Liu, Yongpan et al.
2015 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015. Institute of Electrical and Electronics Engineers Inc., 2016. p. 670-675 7372634 (2015 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Dynamic machine learning based matching of nonvolatile processor microarchitecture to harvested energy profile

AU - Ma, Kaisheng

AU - Li, Xueqing

AU - Liu, Yongpan

AU - Sampson, John

AU - Xie, Yuan

AU - Narayanan, Vijaykrishnan

PY - 2016/1/5

Y1 - 2016/1/5

N2 - Energy harvesting systems without an energy storage device have to efficiently harness the fluctuating and weak power sources to ensure the maximum computational progress. While a simpler processor enables a higher turn-on potential with a weak source, a more powerful processor can utilize more energy that is harvested. Earlier work shows that different complexity levels of nonvolatile microarchitectures provide best fit for different power sources, and even different trails within same power source. In this work, we propose a dynamic nonvolatile microarchitecture by integrating all non-pipelined (NP), N-stage-pipeline (NSP), and Out of Order (OoO) cores together. Neural network machine learning algorithms are also integrated to dynamically adjust the microarchitecture to achieve the maximum forward progress. This integrated solution can achieve forward progress equal to 2.4× of the baseline NP architecture (1.82× of an OoO core).

AB - Energy harvesting systems without an energy storage device have to efficiently harness the fluctuating and weak power sources to ensure the maximum computational progress. While a simpler processor enables a higher turn-on potential with a weak source, a more powerful processor can utilize more energy that is harvested. Earlier work shows that different complexity levels of nonvolatile microarchitectures provide best fit for different power sources, and even different trails within same power source. In this work, we propose a dynamic nonvolatile microarchitecture by integrating all non-pipelined (NP), N-stage-pipeline (NSP), and Out of Order (OoO) cores together. Neural network machine learning algorithms are also integrated to dynamically adjust the microarchitecture to achieve the maximum forward progress. This integrated solution can achieve forward progress equal to 2.4× of the baseline NP architecture (1.82× of an OoO core).

UR - http://www.scopus.com/inward/record.url?scp=84964506908&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84964506908&partnerID=8YFLogxK

U2 - 10.1109/ICCAD.2015.7372634

DO - 10.1109/ICCAD.2015.7372634

M3 - Conference contribution

AN - SCOPUS:84964506908

T3 - 2015 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015

SP - 670

EP - 675

BT - 2015 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 34th IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015

Y2 - 2 November 2015 through 6 November 2015

ER -

Ma K, Li X, Liu Y, Sampson J, Xie Y, Narayanan V. Dynamic machine learning based matching of nonvolatile processor microarchitecture to harvested energy profile. In 2015 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015. Institute of Electrical and Electronics Engineers Inc. 2016. p. 670-675. 7372634. (2015 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2015). doi: 10.1109/ICCAD.2015.7372634

Dynamic machine learning based matching of nonvolatile processor microarchitecture to harvested energy profile

Abstract

Publication series

Other

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this