Incidental computing on IoT nonvolatile processors

Kaisheng Ma, Xueqing Li, Jinyang Li, Yongpan Liu, Yuan Xie, John Morgan Sampson, Mahmut Kandemir, Vijaykrishnan Narayanan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

16 Citations (Scopus)

Abstract

Batteryless IoT devices powered through energy harvesting face a fundamental imbalance between the potential volume of collected data and the amount of energy available for processing that data locally. However, many such devices performsimilar operations across each new input record, which provides opportunities for mining the potential information in buffered historical data, at potentially lower effort, while processing new data rather than abandoning old inputs due to limited computational energy. We call this approach incidental computing, and highlight synergies between this approach and approximation techniques when deployed on a non-volatile processor platform (NVP). In addition to incidental computations, the backup and restore operations in an incidental NVP provide approximation opportunities and optimizations that are unique to NVPs. We propose a variety of incidental approximation approaches suited to NVPs, with a focus on approximate backup and restore, and approximate recomputation in the face of power interruptions. We perform RTL level evaluation for many frequently used workloads.We show that these incidental techniques provide an average of 4.2X more forward progress than precise NVP execution.

Original languageEnglish (US)
Title of host publicationMICRO 2017 - 50th Annual IEEE/ACM International Symposium on Microarchitecture Proceedings
PublisherIEEE Computer Society
Pages204-218
Number of pages15
ISBN (Electronic)9781450349529
DOIs
StatePublished - Oct 14 2017
Event50th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2017 - Cambridge, United States
Duration: Oct 14 2017Oct 18 2017

Publication series

NameProceedings of the Annual International Symposium on Microarchitecture, MICRO
VolumePart F131207
ISSN (Print)1072-4451

Other

Other50th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2017
CountryUnited States
CityCambridge
Period10/14/1710/18/17

Fingerprint

Energy harvesting
Internet of things

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture

Cite this

Ma, K., Li, X., Li, J., Liu, Y., Xie, Y., Sampson, J. M., ... Narayanan, V. (2017). Incidental computing on IoT nonvolatile processors. In MICRO 2017 - 50th Annual IEEE/ACM International Symposium on Microarchitecture Proceedings (pp. 204-218). (Proceedings of the Annual International Symposium on Microarchitecture, MICRO; Vol. Part F131207). IEEE Computer Society. https://doi.org/10.1145/3123939.3124533
Ma, Kaisheng ; Li, Xueqing ; Li, Jinyang ; Liu, Yongpan ; Xie, Yuan ; Sampson, John Morgan ; Kandemir, Mahmut ; Narayanan, Vijaykrishnan. / Incidental computing on IoT nonvolatile processors. MICRO 2017 - 50th Annual IEEE/ACM International Symposium on Microarchitecture Proceedings. IEEE Computer Society, 2017. pp. 204-218 (Proceedings of the Annual International Symposium on Microarchitecture, MICRO).
@inproceedings{22d3f6a3e7df4f06b7d3d9ceafcc6d12,
title = "Incidental computing on IoT nonvolatile processors",
abstract = "Batteryless IoT devices powered through energy harvesting face a fundamental imbalance between the potential volume of collected data and the amount of energy available for processing that data locally. However, many such devices performsimilar operations across each new input record, which provides opportunities for mining the potential information in buffered historical data, at potentially lower effort, while processing new data rather than abandoning old inputs due to limited computational energy. We call this approach incidental computing, and highlight synergies between this approach and approximation techniques when deployed on a non-volatile processor platform (NVP). In addition to incidental computations, the backup and restore operations in an incidental NVP provide approximation opportunities and optimizations that are unique to NVPs. We propose a variety of incidental approximation approaches suited to NVPs, with a focus on approximate backup and restore, and approximate recomputation in the face of power interruptions. We perform RTL level evaluation for many frequently used workloads.We show that these incidental techniques provide an average of 4.2X more forward progress than precise NVP execution.",
author = "Kaisheng Ma and Xueqing Li and Jinyang Li and Yongpan Liu and Yuan Xie and Sampson, {John Morgan} and Mahmut Kandemir and Vijaykrishnan Narayanan",
year = "2017",
month = "10",
day = "14",
doi = "10.1145/3123939.3124533",
language = "English (US)",
series = "Proceedings of the Annual International Symposium on Microarchitecture, MICRO",
publisher = "IEEE Computer Society",
pages = "204--218",
booktitle = "MICRO 2017 - 50th Annual IEEE/ACM International Symposium on Microarchitecture Proceedings",
address = "United States",

}

Ma, K, Li, X, Li, J, Liu, Y, Xie, Y, Sampson, JM, Kandemir, M & Narayanan, V 2017, Incidental computing on IoT nonvolatile processors. in MICRO 2017 - 50th Annual IEEE/ACM International Symposium on Microarchitecture Proceedings. Proceedings of the Annual International Symposium on Microarchitecture, MICRO, vol. Part F131207, IEEE Computer Society, pp. 204-218, 50th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2017, Cambridge, United States, 10/14/17. https://doi.org/10.1145/3123939.3124533

Incidental computing on IoT nonvolatile processors. / Ma, Kaisheng; Li, Xueqing; Li, Jinyang; Liu, Yongpan; Xie, Yuan; Sampson, John Morgan; Kandemir, Mahmut; Narayanan, Vijaykrishnan.

MICRO 2017 - 50th Annual IEEE/ACM International Symposium on Microarchitecture Proceedings. IEEE Computer Society, 2017. p. 204-218 (Proceedings of the Annual International Symposium on Microarchitecture, MICRO; Vol. Part F131207).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Incidental computing on IoT nonvolatile processors

AU - Ma, Kaisheng

AU - Li, Xueqing

AU - Li, Jinyang

AU - Liu, Yongpan

AU - Xie, Yuan

AU - Sampson, John Morgan

AU - Kandemir, Mahmut

AU - Narayanan, Vijaykrishnan

PY - 2017/10/14

Y1 - 2017/10/14

N2 - Batteryless IoT devices powered through energy harvesting face a fundamental imbalance between the potential volume of collected data and the amount of energy available for processing that data locally. However, many such devices performsimilar operations across each new input record, which provides opportunities for mining the potential information in buffered historical data, at potentially lower effort, while processing new data rather than abandoning old inputs due to limited computational energy. We call this approach incidental computing, and highlight synergies between this approach and approximation techniques when deployed on a non-volatile processor platform (NVP). In addition to incidental computations, the backup and restore operations in an incidental NVP provide approximation opportunities and optimizations that are unique to NVPs. We propose a variety of incidental approximation approaches suited to NVPs, with a focus on approximate backup and restore, and approximate recomputation in the face of power interruptions. We perform RTL level evaluation for many frequently used workloads.We show that these incidental techniques provide an average of 4.2X more forward progress than precise NVP execution.

AB - Batteryless IoT devices powered through energy harvesting face a fundamental imbalance between the potential volume of collected data and the amount of energy available for processing that data locally. However, many such devices performsimilar operations across each new input record, which provides opportunities for mining the potential information in buffered historical data, at potentially lower effort, while processing new data rather than abandoning old inputs due to limited computational energy. We call this approach incidental computing, and highlight synergies between this approach and approximation techniques when deployed on a non-volatile processor platform (NVP). In addition to incidental computations, the backup and restore operations in an incidental NVP provide approximation opportunities and optimizations that are unique to NVPs. We propose a variety of incidental approximation approaches suited to NVPs, with a focus on approximate backup and restore, and approximate recomputation in the face of power interruptions. We perform RTL level evaluation for many frequently used workloads.We show that these incidental techniques provide an average of 4.2X more forward progress than precise NVP execution.

UR - http://www.scopus.com/inward/record.url?scp=85034080348&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85034080348&partnerID=8YFLogxK

U2 - 10.1145/3123939.3124533

DO - 10.1145/3123939.3124533

M3 - Conference contribution

AN - SCOPUS:85034080348

T3 - Proceedings of the Annual International Symposium on Microarchitecture, MICRO

SP - 204

EP - 218

BT - MICRO 2017 - 50th Annual IEEE/ACM International Symposium on Microarchitecture Proceedings

PB - IEEE Computer Society

ER -

Ma K, Li X, Li J, Liu Y, Xie Y, Sampson JM et al. Incidental computing on IoT nonvolatile processors. In MICRO 2017 - 50th Annual IEEE/ACM International Symposium on Microarchitecture Proceedings. IEEE Computer Society. 2017. p. 204-218. (Proceedings of the Annual International Symposium on Microarchitecture, MICRO). https://doi.org/10.1145/3123939.3124533