Dynamic energy-aware cloudlet-based mobile cloud computing model for green computing

Dynamic energy-aware cloudlet-based mobile cloud computing model for green computing Anjana H.L.W.[anjanalunuvila@gmail.com]Key words: Mobile Cloud computing, Green Computing, Dynamic Program, Energy-aware, Cloudlets

AuthorsKeke Gai Meikang Qiu Hui Zhao Lixin Tao Ziliang Zong

OutlineIntroductionProposed solutionDECM algorithm Experiments and resultsConclusion

IntroductionMobile cloud computingCloudletWhy energy-aware?Previous researches

4

Mobile cloud computingMobile computingMobile internetCloud computing

Combing the advantages of multiple techniques allows users to offload data processing and storage to the cloud- based servers5

CloudletMobility-enhanced small-scale cloud datacenter that is located at the edge of the Internet.

Supporting resource-intensive and interactive mobile applications by providing powerful computing resources to mobile devices with lower latency6

Why energy-aware ?

Energy over consumptionsWireless communications are weakKeeping searching wireless signals can dry out the power of mobile devices, which may cause unexpected energy wasteGreen computing : minimize energy waste

Previous researchesEnergy aware scheduling by Zhu and his teamCloudlet for efficient data allocationImprovement of cloud service quality from architectural to operating levelsAdmission control policy

Not consider about mobility

Proposed solutionDynamic energy-aware cloudlet-based mobile cloud computingmodel (DEMC)Dynamic cloudlet (DCL)DEMC operation

9

Dynamic energy-aware cloudlet-based mobile cloud computing model (DEMC) Green computing Assign, manage and optimize the cloud-based infrastructure usage and servicesDynamic programming to assist cloudlets

Dynamic CloudletThree missionsLeveraging dynamic programming to determine and predict physical machines on the cloud for current and future use. Calculations of simple applications and quickly respond to end- users. Predict whether users should switch to other cloudlets and enable a real-time suggestion for cloudlets switching.

DEMC operationSearching the nearest CDLs Do calculations and predict other potential cloudlet services according to the service contentsSearch the cloud servers for better service

DECM algorithm DefinitionRecursive formulation

DefinitionEnergy cost minimization (EMC)Total energy consumption - CTotal(T,x) Ei(t,x) : energy consumption among mobile users, cloudlets, and cloud servers specifically occurred on one routePi(t,x) : performance level localized on one route x

Target for green computing

Minimize CTotal(T,x)

In wire-less communicationAmong set of cloudletssum of the products of energy cost for each model and performance percentage levels 14

Dynamic switching between cloudlets that can have better service offeringsMinimum energy consumption with quality services can derive from the recursive formulation.

Performance of each cloudlet is symbolized by CostCDL Recursive formulation

EnergyCloudlet(CDLi,t(Mn)), represents the total energy costs with the service performance at each cloudletTotal cost function is f(Pi(t,x), Ei(t,x))Energy cost that occurred while switching cloudlets is smallRecursive formulation

Experiments and resultsSimulation Two method routsFour method routs

Simulation Work has been examined on the simulator named DECM-Sim that is developed by authors The simulator is designed to calculate input data and generate the results with using the dynamic programmingThe minimum latency time for each route is defined as a full level performance that is marked as 1.0Smart phones with Android 4.4.2 and up operating system, which support at least 300 Mbps max speedHTC One (M8), Samsung Galaxy Note 3, and Samsung Galaxy Note 4.

Two method routs(M1 & M2)Energy consumptions and performances between different cloudlet-based modules.

Two method routs cont..Minimum total costs with computed performance capacities under various timing constraints for a DECM.

Two method routs cont..The result of DECM approach focusing on energy- aware performance

X time constraint Y energy/power

Two method routs cont..Four scenarios defined by the varied timing constraints

Four method routs(M1,M2,M3 & M4)Energy consumptions and performances

Four method routs cont..Minimum total costs under various timing constraints

Four method routs cont..Energy consumptions for different routes with full service performances at four timing constraints

Four method routs cont..Energy consumptions for different routes with full service performances at four timing constraints

Four method routs cont..Energy consumptions for different routes with full service performances at four timing constraints

ConclusionContributionFuture research

Contributions

This research is the first attempt on the functionality of cloudlets in order to achieve energy-aware performances in the dynamic networking environmentThe experimental results proved that this approach was an effective mechanism that could enable mobile users to address green IT with in a dynamic complicated wireless environment.

Future research

DECM applications in industrial environmentBuild up structured connections among CDLs to strengthen communications between cloud servers and mobile devices

Thank you!