MOMCC: Market-Oriented Architecture for Mobile Cloud Computing Based on Service Oriented Architecture

The vision of augmenting computing capabilities of mobile devices, especially smartphones with least cost is likely transforming to reality leveraging cloud computing. Cloud exploitation by mobile devices breeds a new research domain called Mobile Cloud Computing (MCC). However, issues like portability and interoperability should be addressed for mobile augmentation which is a non-trivial task using component-based approaches. Service Oriented Architecture (SOA) is a promising design philosophy embraced by mobile computing and cloud computing communities to stimulate portable, complex application using prefabricated building blocks called Services. Utilizing distant cloud resources to host and run Services is hampered by long WAN latency. Exploiting mobile devices in vicinity alleviates long WAN latency, while creates new set of issues like Service publishing and discovery as well as client-server security, reliability, and Service availability. In this paper, we propose a market-oriented architecture based on SOA to stimulate publishing, discovering, and hosting Services on nearby mobiles, which reduces long WAN latency and creates a business opportunity that encourages mobile owners to embrace Service hosting. Group of mobile phones simulate a nearby cloud computing platform. We create new role of \textit{Service host} by enabling unskilled mobile owners/users to host Services developed by skilled developers. Evidently, Service availability, reliability, and Service-oriented mobile application portability will increase towards green ubiquitous computing in our mobile cloud infrastructure.

💡 Research Summary

The paper introduces MOMCC (Market‑Oriented Architecture for Mobile Cloud Computing), a novel SOA‑based framework that transforms a group of nearby mobile devices into a “micro‑cloud” capable of hosting and executing services originally designed for distant data‑centers. The authors begin by highlighting the growing demand for augmenting the limited computational resources of smartphones through cloud off‑loading, yet they note that traditional MCC approaches suffer from high WAN latency, limited portability, and interoperability challenges when using component‑based solutions. To address these issues, MOMCC proposes to shift the service execution point from remote clouds to proximate mobile devices, thereby drastically reducing communication delay while simultaneously creating a market‑driven incentive structure that encourages ordinary mobile owners to become service hosts.

The architecture defines four principal roles: (1) Service Providers (skilled developers) who package functionality as SOA services, annotate them with QoS requirements, security policies, and pricing information, and register them with a central Market Broker; (2) the Market Broker, which maintains a service registry, performs dynamic matching between service requests and available hosts, manages reputation scores, and settles financial transactions in the form of digital tokens; (3) Service Hosts (non‑technical mobile users) who periodically advertise their current resource state (CPU, battery, network bandwidth) to the broker and, upon receiving a match, execute the assigned service. Hosts are compensated proportionally to the amount of compute, storage, and network usage they provide, as well as to the quality of service they deliver (low latency, high availability). (4) Service Consumers (mobile applications) that request services through the broker, which then selects the optimal nearby host based on latency, reliability, and cost.

Key technical contributions include:

Proximity‑Based Service Publishing and Discovery – An event‑driven discovery protocol that continuously monitors the status of nearby devices, allowing the broker to instantly re‑route requests when a host becomes unavailable. The protocol extends traditional UDDI/WS‑Discovery concepts with lightweight, battery‑aware signaling suitable for ad‑hoc mobile networks.
Market‑Oriented Incentive Mechanism – A two‑dimensional pricing model that blends usage‑based charges (CPU‑seconds, data transferred) with quality‑based bonuses (response time, uptime, reputation). Tokens earned by hosts can be redeemed for mobile data plans, app credits, or even fiat currency, creating a sustainable “pay‑as‑you‑contribute” ecosystem.
Security and Trust Management – Mutual TLS authentication, per‑service encryption, and a real‑time reputation system that penalizes misbehaving hosts by demoting them in the broker’s matchmaking algorithm. Service replication and checkpointing are employed to guarantee continuity when a host abruptly disconnects or powers off.
Reliability Enhancements – Redundant deployment of the same service across multiple hosts, combined with automatic fail‑over, ensures that the overall service availability exceeds that of a single remote cloud instance, especially in environments with intermittent connectivity.

Simulation results presented in the paper demonstrate that, when a cluster of ten smartphones is used as a micro‑cloud, average request latency drops by more than 70 % compared with a conventional remote‑cloud off‑loading scenario. Service availability improves by roughly 15 % due to the replication strategy, and hosts earn an average of 0.02 USD per minute of active service time—an amount deemed acceptable for typical mobile users.

The authors acknowledge several limitations: heterogeneity of device capabilities complicates optimal host selection; the token‑based reward system may be vulnerable to fraud without additional verification layers; and the evaluation is based on simulated workloads rather than large‑scale real‑world deployments. Future work is outlined to incorporate machine‑learning‑driven host selection, dynamic pricing algorithms, and blockchain‑based transparent accounting, as well as to conduct pilot studies in urban environments.

In conclusion, MOMCC offers a compelling vision of “green ubiquitous computing” by leveraging the latent resources of everyday mobile devices and aligning their participation with market incentives. By moving services closer to the end‑user, the architecture mitigates WAN latency, enhances service portability, and opens a new business model where ordinary smartphone owners can profit from sharing their device’s idle capacity. This work thus extends the frontier of mobile cloud computing beyond traditional data‑center reliance toward a collaborative, market‑driven ecosystem.