TY - JOUR
T1 - Elasticity Debt Analytics Exploitation for Green Mobile Cloud Computing
T2 - An Equilibrium Model
AU - Skourletopoulos, Georgios
AU - Mavromoustakis, Constandinos X.
AU - Mastorakis, George
AU - Batalla, Jordi Mongay
AU - Song, Houbing
AU - Sahalos, John N.
AU - Pallis, Evangelos
PY - 2019/3/1
Y1 - 2019/3/1
N2 - Mobile cloud computing is the model to ubiquitously access a shared pool of cloud computing resources, data, and services on-demand. This paper introduces the elasticity debt analytics paradigm as a solution concept for the resource provisioning problem in mobile cloud computing environments, guaranteeing the quality of service requirements. A novel green-centric, game theoretic approach to minimizing the elasticity debt on mobile cloud-based service level is proposed, investigating the mobile cloud offloading case. The decision to offload a mobile device user's task on cloud affects the level of elasticity debt minimization for the provided services. The modeling for the computation of the processing time, energy, and overhead in mobile opportunistic offloading is presented. A utility-driven elasticity debt and profit quantification approach is also examined for maximization of resource utilization, exploiting the hidden Markov model. The problem is formulated as an elasticity debt quantification game, elaborating on an incentive mechanism to predict elasticity debt, mitigate the risk of service over-utilization, achieve scalability, and optimize cloud resource provisioning. The experimental results prove the effectiveness of the equilibrium model, which allocates the mobile device user requests to high elasticity debt-level services and facilitates the elasticity debt minimization for green mobile cloud computing environments.
AB - Mobile cloud computing is the model to ubiquitously access a shared pool of cloud computing resources, data, and services on-demand. This paper introduces the elasticity debt analytics paradigm as a solution concept for the resource provisioning problem in mobile cloud computing environments, guaranteeing the quality of service requirements. A novel green-centric, game theoretic approach to minimizing the elasticity debt on mobile cloud-based service level is proposed, investigating the mobile cloud offloading case. The decision to offload a mobile device user's task on cloud affects the level of elasticity debt minimization for the provided services. The modeling for the computation of the processing time, energy, and overhead in mobile opportunistic offloading is presented. A utility-driven elasticity debt and profit quantification approach is also examined for maximization of resource utilization, exploiting the hidden Markov model. The problem is formulated as an elasticity debt quantification game, elaborating on an incentive mechanism to predict elasticity debt, mitigate the risk of service over-utilization, achieve scalability, and optimize cloud resource provisioning. The experimental results prove the effectiveness of the equilibrium model, which allocates the mobile device user requests to high elasticity debt-level services and facilitates the elasticity debt minimization for green mobile cloud computing environments.
KW - Elasticity debt analytics
KW - game theory
KW - green cloud computing
KW - hidden Markov model
KW - mobile cloud computing
KW - utility computing
UR - http://www.scopus.com/inward/record.url?scp=85067532119&partnerID=8YFLogxK
U2 - 10.1109/TGCN.2018.2890034
DO - 10.1109/TGCN.2018.2890034
M3 - Article
AN - SCOPUS:85067532119
SN - 2473-2400
VL - 3
SP - 122
EP - 131
JO - IEEE Transactions on Green Communications and Networking
JF - IEEE Transactions on Green Communications and Networking
IS - 1
M1 - 8590820
ER -
