Abstract
Conventional mobile computation offloading relies on offline prefetching that fetches user-specific data to the cloud prior to computing. For computing depending on real-time inputs, the offline operation can result in fetching large volumes of redundant data over wireless channels and unnecessarily consumes mobile-transmission energy. To address this issue, we propose the novel technique of online prefetching for a large-scale program with numerous tasks, which seamlessly integrates task-level computation prediction and real-time prefetching within the program runtime. The technique not only reduces mobile-energy consumption by avoiding excessive fetching but also shortens the program runtime by parallel fetching and computing enabled by prediction. By modeling the sequential task transition in an offloaded program as a Markov chain, stochastic optimization is applied to design the online-fetching policies to minimize mobile-energy consumption for transmitting fetched data over fading channels under a deadline constraint. The optimal policies for slow and fast fading are shown to have a similar threshold-based structure that selects candidates for the next task by applying a threshold on their likelihoods and furthermore uses them controlling the corresponding sizes of prefetched data. In addition, computation prediction for online prefetching is shown theoretically to always achieve energy reduction.
| Original language | English |
|---|---|
| Title of host publication | 2017 IEEE International Conference on Communications, ICC 2017 |
| Editors | Merouane Debbah, David Gesbert, Abdelhamid Mellouk |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| ISBN (Electronic) | 9781467389990 |
| DOIs | |
| Publication status | Published - 2017 Jul 28 |
| Event | 2017 IEEE International Conference on Communications, ICC 2017 - Paris, France Duration: 2017 May 21 → 2017 May 25 |
Other
| Other | 2017 IEEE International Conference on Communications, ICC 2017 |
|---|---|
| Country | France |
| City | Paris |
| Period | 17/5/21 → 17/5/25 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Computer Networks and Communications
- Electrical and Electronic Engineering
Cite this
}
Energy efficient mobile computation offloading via online prefetching. / Ko, Seung Woo; Huang, Kaibin; Kim, Seong-Lyun; Chae, Hyukjin.
2017 IEEE International Conference on Communications, ICC 2017. ed. / Merouane Debbah; David Gesbert; Abdelhamid Mellouk. Institute of Electrical and Electronics Engineers Inc., 2017. 7997341.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
TY - GEN
T1 - Energy efficient mobile computation offloading via online prefetching
AU - Ko, Seung Woo
AU - Huang, Kaibin
AU - Kim, Seong-Lyun
AU - Chae, Hyukjin
PY - 2017/7/28
Y1 - 2017/7/28
N2 - Conventional mobile computation offloading relies on offline prefetching that fetches user-specific data to the cloud prior to computing. For computing depending on real-time inputs, the offline operation can result in fetching large volumes of redundant data over wireless channels and unnecessarily consumes mobile-transmission energy. To address this issue, we propose the novel technique of online prefetching for a large-scale program with numerous tasks, which seamlessly integrates task-level computation prediction and real-time prefetching within the program runtime. The technique not only reduces mobile-energy consumption by avoiding excessive fetching but also shortens the program runtime by parallel fetching and computing enabled by prediction. By modeling the sequential task transition in an offloaded program as a Markov chain, stochastic optimization is applied to design the online-fetching policies to minimize mobile-energy consumption for transmitting fetched data over fading channels under a deadline constraint. The optimal policies for slow and fast fading are shown to have a similar threshold-based structure that selects candidates for the next task by applying a threshold on their likelihoods and furthermore uses them controlling the corresponding sizes of prefetched data. In addition, computation prediction for online prefetching is shown theoretically to always achieve energy reduction.
AB - Conventional mobile computation offloading relies on offline prefetching that fetches user-specific data to the cloud prior to computing. For computing depending on real-time inputs, the offline operation can result in fetching large volumes of redundant data over wireless channels and unnecessarily consumes mobile-transmission energy. To address this issue, we propose the novel technique of online prefetching for a large-scale program with numerous tasks, which seamlessly integrates task-level computation prediction and real-time prefetching within the program runtime. The technique not only reduces mobile-energy consumption by avoiding excessive fetching but also shortens the program runtime by parallel fetching and computing enabled by prediction. By modeling the sequential task transition in an offloaded program as a Markov chain, stochastic optimization is applied to design the online-fetching policies to minimize mobile-energy consumption for transmitting fetched data over fading channels under a deadline constraint. The optimal policies for slow and fast fading are shown to have a similar threshold-based structure that selects candidates for the next task by applying a threshold on their likelihoods and furthermore uses them controlling the corresponding sizes of prefetched data. In addition, computation prediction for online prefetching is shown theoretically to always achieve energy reduction.
UR - http://www.scopus.com/inward/record.url?scp=85028297591&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85028297591&partnerID=8YFLogxK
U2 - 10.1109/ICC.2017.7997341
DO - 10.1109/ICC.2017.7997341
M3 - Conference contribution
AN - SCOPUS:85028297591
BT - 2017 IEEE International Conference on Communications, ICC 2017
A2 - Debbah, Merouane
A2 - Gesbert, David
A2 - Mellouk, Abdelhamid
PB - Institute of Electrical and Electronics Engineers Inc.
ER -