[Paper Review] Green Energy Aware Avatar Migration Strategy in Green Cloudlet Networks

Xiang Sun, Nirwan Ansari+1 more|arXiv (Cornell University)|Sep 11, 2015

Caching and Content Delivery22 references7 citations

TL;DR

This paper proposes the Green-energy aware Avatar Migration (GEAR) strategy for Green Cloudlet Networks (GCN) to minimize on-grid energy consumption by intelligently migrating Avatars among cloudlets based on real-time green energy generation and energy demand. GEAR reduces on-grid energy use by 35% compared to the Follow Me Avatar (FAR) strategy while maintaining strict latency constraints via mixed-integer linear programming and Branch and Bound optimization.

ABSTRACT

We propose a Green Cloudlet Network (\emph{GCN}) architecture to provide seamless Mobile Cloud Computing (\emph{MCC}) services to User Equipments (\emph{UE}s) with low latency in which each cloudlet is powered by both green and brown energy. Fully utilizing green energy can significantly reduce the operational cost of cloudlet providers. However, owing to the spatial dynamics of energy demand and green energy generation, the energy gap among different cloudlets in the network is unbalanced, i.e., some cloudlets' energy demands can be fully provided by green energy but others need to utilize on-grid energy (i.e., brown energy) to satisfy their energy demands. We propose a Green-energy awarE Avatar migRation (\emph{GEAR}) strategy to minimize the on-grid energy consumption in GCN by redistributing the energy demands via Avatar migration among cloudlets according to cloudlets' green energy generation. Furthermore, GEAR ensures the Service Level Agreement (\emph{SLA}) in terms of the maximum Avatar propagation delay by avoiding Avatars hosted in the remote cloudlets. We formulate the GEAR strategy as a mixed integer linear programming problem, which is NP-hard, and thus apply the Branch and Bound search to find its sub-optimal solution. Simulation results demonstrate that GEAR can save on-grid energy consumption significantly as compared to the Follow me AvataR (\emph{FAR}) migration strategy, which aims to minimize the propagation delay between an UE and its Avatar.

Motivation & Objective

• Address the imbalance in green energy generation and energy demand across distributed cloudlets in Mobile Cloud Computing (MCC) networks.
• Minimize on-grid (brown) energy consumption in Green Cloudlet Networks (GCN) by redistributing computational workloads via Avatar migration.
• Ensure Service Level Agreement (SLA) compliance by limiting maximum Avatar propagation delay to maintain low-latency MCC services.
• Optimize energy utilization in cloudlets powered by hybrid green and on-grid energy sources, especially under spatial and temporal dynamics of energy supply and demand.

Proposed method

• Formulate the GEAR strategy as a mixed-integer linear programming (MILP) problem to optimize Avatar migration decisions.
• Use Branch and Bound search to compute a sub-optimal solution for the NP-hard MILP problem.
• Model cloudlet energy dynamics based on real-time solar radiation data and varying UE density across urban and rural areas.
• Integrate SDN-based control plane to enable flexible, low-latency communication paths between Avatars and UEs.
• Balance energy demand across cloudlets by migrating Avatars from high-demand, low-green-energy cloudlets to low-demand, high-green-energy cloudlets.
• Incorporate constraints to limit Avatar propagation delay and ensure QoS compliance, preventing remote hosting.

Experimental results

Research questions

• RQ1 How can Avatar migration be optimized to minimize on-grid energy consumption in Green Cloudlet Networks?
• RQ2 To what extent can energy demand imbalance across cloudlets be mitigated through intelligent Avatar migration?
• RQ3 How does the spatial distribution of green energy generation (e.g., urban vs. rural) affect the effectiveness of energy-aware migration strategies?
• RQ4 What is the trade-off between minimizing on-grid energy use and maintaining low propagation delay in Avatar-based MCC systems?
• RQ5 How does increasing UE density impact the scalability and energy savings of the GEAR strategy compared to conventional approaches?

Key findings

• GEAR reduces on-grid energy consumption by 35% compared to the Follow Me Avatar (FAR) strategy over a 24-hour period under standard conditions.
• The energy savings from GEAR increase with higher UE density up to 1,100 UEs, as energy demand imbalance becomes more pronounced and GEAR effectively balances the load.
• When UE density exceeds 1,100, energy savings plateau because green energy is already fully utilized, and additional demand forces reliance on on-grid power.
• In scenarios with spatially varying solar radiation (e.g., urban areas receiving 0–30% less solar input than rural areas), GEAR saves significantly more on-grid energy than FAR as the energy gap widens.
• During periods of low or no green energy generation (e.g., 8–9 a.m.), both GEAR and FAR show similar on-grid energy use, confirming that GEAR’s advantage emerges only when green energy is available.
• GEAR maintains strict SLA compliance by avoiding remote cloudlet hosting, ensuring maximum propagation delay remains within acceptable limits.