TY - GEN
T1 - Feasibility of multi-tenancy on intermittent power
AU - Patoukas, Dimitris
AU - Yýldýrým, Kasim Sinan
AU - Majid, Amjad Yousef
AU - Hester, Josiah David
AU - Pawełczak, Przemyslaw
N1 - Publisher Copyright:
© ENSsys 2018 - Proceedings of the 2018 International Workshop on Energy Harvesting and Energy-Neutral Sensing Systems, Part of SenSys.
Copyright:
Copyright 2019 Elsevier B.V., All rights reserved.
PY - 2018/11/4
Y1 - 2018/11/4
N2 - Energy harvesting and battery-free sensing devices show great promise for revolutionizing computing in the home, in the wild, and on the body. The promise of cheap, dense, and ubiquitous sensing technology brings new applications for the Internet of Things. However, the future programming model is blurry and complex. With a potential for trillions of devices, and thousands of devices per person on earth, programming languages and associated operating systems must be usable, flexible, and resource efficient. Because of the thousands of applications and fine grained differences in requirements, multi-tenancy may be a part of the solution to solving this programming model crisis. This paper explores the energy and resources costs, feasibility, and motivation for multi-tenancy on these tiniest of computing devices-namely the difficulties in scheduling tasks fairly, efficiently, and simply. Because of intermittent power, resources and energy must be mostly devoted towards user tasks, we implement a rudimentary operating system with low overhead to conduct experiments and test time-sharing and scheduling protocols. We close with a discussion on challenges to implementing a multi-tenant run-time on battery-free tags, and proposals for future work.
AB - Energy harvesting and battery-free sensing devices show great promise for revolutionizing computing in the home, in the wild, and on the body. The promise of cheap, dense, and ubiquitous sensing technology brings new applications for the Internet of Things. However, the future programming model is blurry and complex. With a potential for trillions of devices, and thousands of devices per person on earth, programming languages and associated operating systems must be usable, flexible, and resource efficient. Because of the thousands of applications and fine grained differences in requirements, multi-tenancy may be a part of the solution to solving this programming model crisis. This paper explores the energy and resources costs, feasibility, and motivation for multi-tenancy on these tiniest of computing devices-namely the difficulties in scheduling tasks fairly, efficiently, and simply. Because of intermittent power, resources and energy must be mostly devoted towards user tasks, we implement a rudimentary operating system with low overhead to conduct experiments and test time-sharing and scheduling protocols. We close with a discussion on challenges to implementing a multi-tenant run-time on battery-free tags, and proposals for future work.
UR - http://www.scopus.com/inward/record.url?scp=85058421386&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85058421386&partnerID=8YFLogxK
U2 - 10.1145/3279755.3279761
DO - 10.1145/3279755.3279761
M3 - Conference contribution
AN - SCOPUS:85058421386
T3 - ENSsys 2018 - Proceedings of the 2018 International Workshop on Energy Harvesting and Energy-Neutral Sensing Systems, Part of SenSys 2018
SP - 26
EP - 31
BT - ENSsys 2018 - Proceedings of the 2018 International Workshop on Energy Harvesting and Energy-Neutral Sensing Systems, Part of SenSys 2018
A2 - Bhaskar, Krishnamachari
A2 - Ramachandran, Gowri Sankar
PB - Association for Computing Machinery, Inc
T2 - 6th International Workshop on Energy Harvesting and Energy Neutral Sensing Systems, ENSsys 2018, co-located with ACM SenSys 2018
Y2 - 4 November 2018
ER -