Balancing distributed analytics' energy consumption using physics-inspired models

Brent Kraczek; Theodoros Salonidis; Prithwish Basu; Sayed Saghaian; Ali Sydney; Bongjun Ko; Tom Laporta; Kevin Chan; James Lambert

doi:10.1117/12.2304485

Balancing distributed analytics' energy consumption using physics-inspired models

Brent Kraczek
, Theodoros Salonidis
, Prithwish Basu
, Sayed Saghaian
, Ali Sydney
, Bongjun Ko
, Tom Laporta
, Kevin Chan
, James Lambert

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

Abstract

With the rise of small, networked sensors, the volume of data generated increasingly require curation by AI to analyze which events are of sufficient importance to report to human operators. We consider the ultimate limit of edge computing, when it is impractical to employ external resources for the curation, but individual devices have insufficient computing resources to perform the analytics themselves. In a previous paper we introduced a decenralized method that distributes the analytics over the network of devices, employing simulated annealing, based on physics-inspired Metropolis Monte Carlo. If the present paper we discuss the capability of this method to balance the energy consumption of the placement on a network of heterogeneous resources. We introduce the balanced utilization index (BUI), an adaptation of Jain's Fairness Index, to measure this balance.

Original language	English (US)
Title of host publication	Disruptive Technologies in Information Sciences
Editors	Misty Blowers, Russell D. Hall, Venkateswara R. Dasari
Publisher	SPIE
ISBN (Electronic)	9781510618152
DOIs	https://doi.org/10.1117/12.2304485
State	Published - 2018
Event	Disruptive Technologies in Information Sciences 2018 - Orlando, United States Duration: Apr 17 2018 → Apr 18 2018

Publication series

Name	Proceedings of SPIE - The International Society for Optical Engineering
Volume	10652
ISSN (Print)	0277-786X
ISSN (Electronic)	1996-756X

Other

Other	Disruptive Technologies in Information Sciences 2018
Country/Territory	United States
City	Orlando
Period	4/17/18 → 4/18/18

UN SDGs

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

SDG 7 Affordable and Clean Energy

All Science Journal Classification (ASJC) codes

Electronic, Optical and Magnetic Materials
Condensed Matter Physics
Computer Science Applications
Applied Mathematics
Electrical and Electronic Engineering

Access to Document

10.1117/12.2304485

Cite this

Kraczek, B., Salonidis, T., Basu, P., Saghaian, S., Sydney, A., Ko, B., Laporta, T., Chan, K., & Lambert, J. (2018). Balancing distributed analytics' energy consumption using physics-inspired models. In M. Blowers, R. D. Hall, & V. R. Dasari (Eds.), Disruptive Technologies in Information Sciences Article 1065206 (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 10652). SPIE. https://doi.org/10.1117/12.2304485

@inproceedings{d414df6446b944db95eca528a00afbd8,

title = "Balancing distributed analytics' energy consumption using physics-inspired models",

abstract = "With the rise of small, networked sensors, the volume of data generated increasingly require curation by AI to analyze which events are of sufficient importance to report to human operators. We consider the ultimate limit of edge computing, when it is impractical to employ external resources for the curation, but individual devices have insufficient computing resources to perform the analytics themselves. In a previous paper we introduced a decenralized method that distributes the analytics over the network of devices, employing simulated annealing, based on physics-inspired Metropolis Monte Carlo. If the present paper we discuss the capability of this method to balance the energy consumption of the placement on a network of heterogeneous resources. We introduce the balanced utilization index (BUI), an adaptation of Jain's Fairness Index, to measure this balance.",

author = "Brent Kraczek and Theodoros Salonidis and Prithwish Basu and Sayed Saghaian and Ali Sydney and Bongjun Ko and Tom Laporta and Kevin Chan and James Lambert",

note = "Publisher Copyright: {\textcopyright} 2018 SPIE.; Disruptive Technologies in Information Sciences 2018 ; Conference date: 17-04-2018 Through 18-04-2018",

year = "2018",

doi = "10.1117/12.2304485",

language = "English (US)",

series = "Proceedings of SPIE - The International Society for Optical Engineering",

publisher = "SPIE",

editor = "Misty Blowers and Hall, \{Russell D.\} and Dasari, \{Venkateswara R.\}",

booktitle = "Disruptive Technologies in Information Sciences",

address = "United States",

}

Kraczek, B, Salonidis, T, Basu, P, Saghaian, S, Sydney, A, Ko, B, Laporta, T, Chan, K & Lambert, J 2018, Balancing distributed analytics' energy consumption using physics-inspired models. in M Blowers, RD Hall & VR Dasari (eds), Disruptive Technologies in Information Sciences., 1065206, Proceedings of SPIE - The International Society for Optical Engineering, vol. 10652, SPIE, Disruptive Technologies in Information Sciences 2018, Orlando, United States, 4/17/18. https://doi.org/10.1117/12.2304485

Balancing distributed analytics' energy consumption using physics-inspired models. / Kraczek, Brent; Salonidis, Theodoros; Basu, Prithwish et al.
Disruptive Technologies in Information Sciences. ed. / Misty Blowers; Russell D. Hall; Venkateswara R. Dasari. SPIE, 2018. 1065206 (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 10652).

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

TY - GEN

T1 - Balancing distributed analytics' energy consumption using physics-inspired models

AU - Kraczek, Brent

AU - Salonidis, Theodoros

AU - Basu, Prithwish

AU - Saghaian, Sayed

AU - Sydney, Ali

AU - Ko, Bongjun

AU - Laporta, Tom

AU - Chan, Kevin

AU - Lambert, James

PY - 2018

Y1 - 2018

N2 - With the rise of small, networked sensors, the volume of data generated increasingly require curation by AI to analyze which events are of sufficient importance to report to human operators. We consider the ultimate limit of edge computing, when it is impractical to employ external resources for the curation, but individual devices have insufficient computing resources to perform the analytics themselves. In a previous paper we introduced a decenralized method that distributes the analytics over the network of devices, employing simulated annealing, based on physics-inspired Metropolis Monte Carlo. If the present paper we discuss the capability of this method to balance the energy consumption of the placement on a network of heterogeneous resources. We introduce the balanced utilization index (BUI), an adaptation of Jain's Fairness Index, to measure this balance.

AB - With the rise of small, networked sensors, the volume of data generated increasingly require curation by AI to analyze which events are of sufficient importance to report to human operators. We consider the ultimate limit of edge computing, when it is impractical to employ external resources for the curation, but individual devices have insufficient computing resources to perform the analytics themselves. In a previous paper we introduced a decenralized method that distributes the analytics over the network of devices, employing simulated annealing, based on physics-inspired Metropolis Monte Carlo. If the present paper we discuss the capability of this method to balance the energy consumption of the placement on a network of heterogeneous resources. We introduce the balanced utilization index (BUI), an adaptation of Jain's Fairness Index, to measure this balance.

UR - https://www.scopus.com/pages/publications/85049225449

UR - https://www.scopus.com/pages/publications/85049225449#tab=citedBy

U2 - 10.1117/12.2304485

DO - 10.1117/12.2304485

M3 - Conference contribution

AN - SCOPUS:85049225449

T3 - Proceedings of SPIE - The International Society for Optical Engineering

BT - Disruptive Technologies in Information Sciences

A2 - Blowers, Misty

A2 - Hall, Russell D.

A2 - Dasari, Venkateswara R.

PB - SPIE

T2 - Disruptive Technologies in Information Sciences 2018

Y2 - 17 April 2018 through 18 April 2018

ER -

Balancing distributed analytics' energy consumption using physics-inspired models

Abstract

Publication series

Other

UN SDGs

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this