Energy Cost Models of Smartphones forTask Offloading to the Cloud
Smartphones have unique constraints
Need to reduce the energy consumption
Many methodologies and techniques have been proposed in literature
Cloud computing
Task offloading
2. Energy Cost Models of
Smartphones for
Task Offloading to the
Cloud
ASWIN VP
CTAMECS042
3. INTRODUCTION
• Smartphones have unique constraints
• Need to reduce the energy consumption
• Many methodologies and techniques have been proposed in
literature
• Cloud computing
• Task offloading
4. INTRODUCTION
• Mobile device can save energy by offloading heavy tasks to the
cloud
• Cloud executes the tasks
• Cloud provides the mobile device with the results
• There are two scenarios:
• Execute the task locally (S1)
• Offload the task to the cloud (S2)
• Offloading is only beneficial if E(S2) < E(S1)
5. LITERATURE
REVIEW
DevScope: A Nonintrusive and Online
Power Analysis Tool for Smartphone
Hardware Components[1]
• Online approach
• Battery Monitoring Unit
• Generating a dynamic power model
• Autonomous power modeling tool for smartphones called
DevScope
6. LITERATURE
REVIEW
• Advantages
• Automatic and online smartphone power modeling
techniques
• Generating a dynamic power model
• Disadvantages
• BMU can’t trace events that are shorter than BMU
update rate
• Not accurate and not extendible for modeling
7. LITERATURE
REVIEW
Understanding the Challenges in Mobile
Computation Offloading to Cloud[2]
• Saving Battery Consumption
• Dynamic Task Offloading
• Low Cost Cloud Services
• Mobile Cloud Computing
• Application Characteristics
8. LITERATURE
REVIEW
• Advantages
• Easy to understand the challenges
• Easy to find out the impacts of offloading
• Disadvantages
• No Security
• Not understanding the efficiency
10. LITERATURE
REVIEW
• Advantages
• Saves energy of handled devices
• Disadvantages
• Only takes into account the impact of the torrent traffic
pattern
• Does not consider the computation cost of the given task
12. TECHNOLOGIES
USED
3G (THIRD GENERATION)
• Mobile Telecommunication Technology
• International Mobile Telecommunication standard
• Wireless Voice Telephony
• Mobile Internet Access
• Fixed Wireless Internet Access
• Higher Data Rates
13. TECHNOLOGIES
USED
4G (FOURTH GENERATION)
• Mobile Telecommunication Technology
• International Mobile Telecommunication standard
• IP Telephony
• 300 Mbits/s
• 3D Television
• Cloud Computing
14. TECHNOLOGIES
USED
Wireless LAN (WLAN)
• Wireless computer network
• Wireless distribution method
• Limited Area
• Home , School , Computer Lab and Office Building
• IEEE 802.11
• Easy to handle
• 600 Mbits/s
15. WORKING
• Model consist of Two parts
• Smartphone
• Cloud Computing
• Smartphones connected to Internet by WLAN,3G,4G
• CC part consists of cloud data center and cloud provider
17. WORKING
• Network Related Application
• Internet protocols
• Network Interfaces
• Energy Cost
• Task data is available on the smartphone itself
• Task data is available in the cloud
• Four scenarios related to the location of the task data
20. WORKING
• Uploading files to the cloud
• Downloading from the cloud
• Energy implications
• HTTP and FTP protocols at the application level
• 3G and WLAN communications at the wireless interface level
• The total energy consumed in a smartphone is
• E(Total)=E(WNI)+E(OS)
22. WORKING
• Network type
• Amount of Task data
• Energy Models
• WLAN Analytical Energy Model
• File Download Case
• File Upload Case
• Mobile Data Analytical Energy Model
• Energy Models For The 3g/4g
23. ADVANTAGES
• Estimate accurately the energy consumed during the network
activities of task offloading.
• Allow smartphones to make correct offloading decisions.
• Predicts the energy consumption.
24. DISADVANTAGES
• Need to know accurate amount of transferrable data
• Model is not effective in the case of small tasks
• Need to get correct system parameters from the system
25. FUTURE
ENHANCEMENTS
• Enhanced from mobile devices to each and every other
computational electronic device
• Providing security to the tasks being uploaded to the cloud
26. CONCLUSION
• The proposed energy models of WLAN, 3G, and 4G interfaces allow
smartphones to make correct offloading decisions
• This models not only help for task offloading but also opens new
door for energy solutions that require predicting the energy
consumption
27. REFERENCES
• [1] W. Jung, C. Kang, C. Yoon, D. Kim, and H. Cha, DevScope: A
nonintrusive and online power analysis tool for smartphone
hardware components, in Proc. 8th IEEE/ACM/IFIP Int. Conf.
Hardw./Softw. Codesign Syst Synth. (CODES+ISSS), Oct. 2012
• [2] Understanding the Challenges in Mobile Computation
Offloading to Cloud through Experimentation Authors: Joshi.P,
Kumar R,Kumar S Date: 16 May 2015
• [3] I. Kelenyi “CloudTorrentEnergy-efcient BitTorrent content
sharing for mobile devices via cloud services” in Proc. 7th IEEE
Consum. Commun. Netw. Conf. (CCNC), Jan. 2010, pp. 12