Huber Flores huber.flores@ee.oulu.fi Social-aware Hybrid Mobile Offloading A contribution for edge and fog computing? Huber Flores huber.flores@ee.oulu.fi.

Slides:



Advertisements
Similar presentations
Source: IEEE Pervasive Computing, Vol. 8, Issue.4, Oct.2009, pp. 14 – 23 Author: Satyanarayanan, M., Bahl, P., Caceres, R., Davies, N. Adviser: Chia-Nian.
Advertisements

Mostafa Ammar, School of Computer Science Georgia Institute of Technology Atlanta, GA Mobile Computing in Cirrus Clouds: Mobile Computing in Cirrus Clouds:
Understanding Human-Smartphone Concerns: A Study of Battery Life Denzil Ferreira, Anind K. Dey, Vassilis Kostakos Pervasive 2011.
Presented by Tao HUANG Lingzhi XU. Context Mobile devices need exploit variety of connectivity options as they travel. Operating systems manage wireless.
CrowdSearch: Exploiting Crowds for Accurate Real-Time Image Search on Mobile Phones Original work by Yan, Kumar & Ganesan Presented by Tim Calloway.
Mobile cloud computing: survey 1. Introduction  In recent years, applications targeted at mobile devices havs started becoming abundant with applications.
MOBILE CLOUD COMPUTING
GENI-related research activities of CSE, Aalto Zhonghong Ou Post-doc researcher Department of Computer Science and Engineering.
IEEE R lmap 23 Feb 2015.
A Unified Modeling Framework for Distributed Resource Allocation of General Fork and Join Processing Networks in ACM SIGMETRICS
Ways of thinking about ethics Duty ethics arise from an absolute set of values that apply in all circumstances (Ten commandments from the Christian Bible,
Children: Can they inspect it? Yes they can! Gavin Sim.
Secure Opportunistic Mobile Application Offload for Enterprise Networks Aaron Gember and Aditya Akella University of Wisconsin – Madison Abstract Application-independent.
Mobile Cloud
CrowdSearch: Exploiting Crowds for Accurate Real-Time Image Search on Mobile Phones Original work by Tingxin Yan, Vikas Kumar, Deepak Ganesan Presented.
Privacy and Security: Thinking About and Analyzing Privacy privacy and security 1 Research Topics in Ubiquitous Computing Ben Elgart thinking about and.
July 2013 Elastic Offloading by Dale Denis. Dale Denis The Elastic Offloading of Computationally Intensive Tasks to the Cloud to Augment the Computing.
Efficient Opportunistic Sensing using Mobile Collaborative Platform MOSDEN.
Application-Aware Traffic Scheduling for Workload Offloading in Mobile Clouds Liang Tong, Wei Gao University of Tennessee – Knoxville IEEE INFOCOM
Speaker: Yuen-Kuei Hsueh Mixed-Initiative Conflict Resolution for Context-aware Applications Choonsung Shin, Anind K. Dey, and Woontack Woo Proceedings.
CloudMAC: Moving MAC frames processing of the Sink to Cloud.
Q K-12 Blueprint Overview. 2 The K-12 Blueprint offers resources for education leaders involved in planning and implementing personalized learning.
Introduction to Mobile-Cloud Computing. What is Mobile Cloud Computing? an infrastructure where both the data storage and processing happen outside of.
Use of Spectrum Bands Above 24 GHz For Mobile Radio Services ‘5G’…
A Hierarchical Edge Cloud Architecture for Mobile Computing IEEE INFOCOM 2016 Liang Tong, Yong Li and Wei Gao University of Tennessee – Knoxville 1.
Edge Computing ——vision, challenges and promise. 物联网云计算.
Evolution of storage and data management
SAM Baseline Review Engagement
Seminar Announcement December 24, Saturday, 15:00-17:00, Room: A302, WNLO Title: Quality-of-Experience (QoE) and Power Efficiency Tradeoff for Fog Computing.
Energy-Aware Opportunistic Mobile Data Offloading
1.4 Wired and Wireless Networks
Behavioural Insights and the Change to Walking Program
Submission Title: [Discovery Latency] Date Submitted: [ ]
Methodology: Aspects: cost models, modelling of system, understanding of behaviour & performance, technology evolution, prototyping  Develop prototypes.
2 ATIS 5G OVERVIEW ATIS launched its 5G Ad Hoc in 2015 to advance regulatory imperatives, deliver an evolutionary path, address co-existence of technologies,
Quantifying the Impact of Edge Computing on Mobile Applications
Submission Title: [Discovery Latency] Date Submitted: [ ]
World-Leading Research with Real-World Impact!
© 2016 Global Market Insights, Inc. USA. All Rights Reserved Fuel Cell Market size worth $25.5bn by 2024 Beacon Technology Market to.
Introduction to Edge Computing
Quality-aware Aggregation & Predictive Analytics at the Edge
© 2016 Global Market Insights, Inc. USA. All Rights Reserved Beacon Technology Market to grow at 80% CAGR from 2017 to 2024: Global.
ISTE Workshop Research Methods in Educational Technology
Cloud Computing.
SAM Server Optimization Engagement
Script-less Automation: An Approach to Shift-Left.
SAM Infrastructure Optimization Engagement
Behavioural Insights and the Change to Walking Program
Thor: The Hybrid Online Repository
Exploratory Search Beyond the Query–Response Paradigm
Model’s core: Personalisation sub-model Model’s core: User sub-model
Submission Title: Technical proposal for PAC
Course Project Topics for CSE5469
Speaker: Jin-Wei Lin Advisor: Dr. Ho-Ting Wu
More on Estimation In general, effort estimation is based on several parameters and the model ( E= a + b*S**c ): Personnel Environment Quality Size or.
Smart Learning concepts to enhance SMART Universities in Africa
T IWORK Research topics
GENI Global Environment for Network Innovation
Sense-Aid: A Mobile Crowdsensing framework
IEEE MEDIA INDEPENDENT HANDOVER DCN:
IEEE MEDIA INDEPENDENT HANDOVER DCN:
Introducing Evaluation
and for the theatre community
IoT Requirements for Networking Protocols Sadoon Azizi Department of Computer Engineering and IT.
IEEE MEDIA INDEPENDENT HANDOVER DCN:
IEEE MEDIA INDEPENDENT HANDOVER DCN:
Kostas Kolomvatsos, Christos Anagnostopoulos
Dr. Kunal Mankodiya University of Rhode Island, USA
Carl-Fredrik Sørensen
AR and 5G In the Enterprise
Presentation transcript:

Huber Flores huber.flores@ee.oulu.fi Social-aware Hybrid Mobile Offloading A contribution for edge and fog computing? Huber Flores huber.flores@ee.oulu.fi Hello, good day, my name is Huber Flores, I am postdoc at University of Oulu I conduct research about mobile and cloud computing, specifically about mobile offloading. Today I am glad to present you the results about a study we conducted to design a mobile offloading framework which unifies different offloading models So the title of my presentations is Social-aware Hybrid Mobile Offloading And the discussion around it … is to know whether or not the approach can be exploited in practice

Roadmap Background Problem statement Studies towards the solution Mobile offloading Problem statement Opportunistic spectrum Studies towards the solution Hybrid system Lessons learned and experiences Discussion Conclusions I will talk briefly about mobile offloading and its multiple models. I will also highlight the opportunistic spectrum of each of them Later, I will present the hybrid system along with the challenges that we overcome for building it. Next, some results about the experimental evaluation in the wild. Lastly, some conclusions and insights about the exploitation of the system in edge and fog computing

Mobile offloading Opportunistic augmentation of resources Processing Storage etc… It’s just the opportunistic process of moving one computational or storage task from one place to another. Naturally, when the mobile context to offload the task is good enough, such that outsourcing the task represents less effort for the device rather than processing by itself. Usually, low power rely on this approach to release the resources from computational processing, such that it is possible to improve performance and save energy. The potential of mobile offloading is already well-known in the community. Thus, the effort is towards approaches that improve the availability of the offloading support. [IEEE Communications 2015] Flores, H., Hui, P., Tarkoma, S., Li, Y., Srirama, S., & Buyya, R. (2015). Mobile code offloading: from concept to practice and beyond. IEEE Communications Magazine, 53(3), 80-88.

Mobile offloading models Cloudlet Scalability Remote cloud Latency in the communication Device-to-Device (D2D) Social participation There are different offloading models. The opportunistic spectrum is influenced by many drawbacks. Cloudlet Rich and nearby servers Remote cloud Cloud infrastructure Device to device communication User’s device

Hybrid offloading system Increasing the offloading spectrum We propose a hybrid system that combines all the offloading models in order to augment the offloading spectrum But the real question here is Can this system be really exploited in the wild? What are the considerations for this system to work?

Hybrid offloading system Addressability of the user’s device Privacy Stability based on user’s mobility Duration and frequency Social participation How valuable is batterylife is for the user? Now you may think…. Why do we need to address this challenges within the system is communication is increasing….

Study: proximal infrastructure Even the device is present, it is not discoverable, so how much really D2D can deliver? If we let the mobile operator handle this, many concerns will arise Who is the owner of the device? [WMSC-UbiComp 2016] Flores, H., Sharma, R., Ferreira D., Lou, C., Kostakos, V. Tarkoma, Hui, P., Li, Y. (2016) Social-aware Device-to-Device Communication: A contribution for edge and fog computing?, UbiComp Workshop on Mobile and Situated Crowdsourcing , 2016.

Results

Results

Study: stability region [PMC 2016] Flores, H., Sharma, R., Ferreira D., Kostakos, V. Tarkoma, S. Manner, J., Hui, P., Li, Y. (2016) Social-aware Hybrid Mobile Offloading, Pervasive and Mobile Computing Journal, 2016.

Results

Study: monetary assesment of battery life Battery is money!

Monetary assesment of battery life [CHI 2016] Hosio, S., Ferreira, D., Goncalves, J., van Berkel, Flores, H., ... & Kostakos, V. (2016). Monetary Assessment of Battery Life on Smartphones. In Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems (pp. 1869-1880). ACM.

Monetary assesment of battery life

Social-aware hybrid offloading

Social-aware hybrid offloading Peers Super-peers Credit and reputation Propagation of updates and consistency, which we need to explore further.

Social-aware hybrid offloading

Mobile offloading (3G/3G LTE) DNA ELISA SONERA 1000 samples to characterize the average latency experienced by a user. Notice that I am showing average. I should be showing median because people is not measure in averages, but even though is the average, we can see relative low latency.

Conclusions The offloading spectrum increases substancially An extra layer of complexity and overhead is introduced when handling social participation.

questions