[Paper Review] Toward Cloud-based Vehicular Networks with Efficient Resource Management

Rong Yu, Yan Zhang+3 more|arXiv (Cornell University)|Aug 28, 2013

Vehicular Ad Hoc Networks (VANETs)10 references4 citations

TL;DR

This paper proposes a hierarchical cloud-based vehicular network architecture integrating vehicular, roadside, and central clouds to enable efficient sharing of computation, storage, and bandwidth resources. It employs a game-theoretic approach for optimal cloud resource allocation and a resource reservation scheme to minimize virtual machine dropping during mobility-induced migration, achieving a significant reduction in service dropping rates.

ABSTRACT

In the era of Internet of Things, all components in intelligent transportation systems will be connected to improve transport safety, relieve traffic congestion, reduce air pollution and enhance the comfort of driving. The vision of all vehicles connected poses a significant challenge to the collection and storage of large amounts of traffic-related data. In this article, we propose to integrate cloud computing into vehicular networks such that the vehicles can share computation resources, storage resources and bandwidth resources. The proposed architecture includes a vehicular cloud, a roadside cloud, and a central cloud. Then, we study cloud resource allocation and virtual machine migration for effective resource management in this cloud-based vehicular network. A game-theoretical approach is presented to optimally allocate cloud resources. Virtual machine migration due to vehicle mobility is solved based on a resource reservation scheme.

Motivation & Objective

• To address the challenge of limited onboard resources in vehicles by enabling resource sharing through cloud integration.
• To design a resilient, hierarchical cloud architecture that supports mobile vehicles with scalable and low-latency services.
• To ensure service continuity and reliability in dynamic vehicular environments due to vehicle mobility.
• To optimize cloud resource allocation among competing virtual machines using game theory.
• To reduce virtual machine migration failure rates through proactive resource reservation.

Proposed method

• Proposes a three-layer cloud architecture: vehicular cloud (onboard), roadside cloud (nearby infrastructure), and central cloud (remote data centers).
• Models cloud resource allocation as a non-cooperative game among virtual machines to achieve optimal resource distribution.
• Uses a continuous-time Markov chain to model system states, capturing arrivals and departures of local and migrated VMs.
• Introduces a resource reservation scheme to reserve capacity for incoming migrated VMs, minimizing blocking and dropping.
• Derives steady-state probabilities of system states to compute blocking and dropping rates under different resource configurations.
• Solves an optimization problem to minimize dropping rate subject to a maximum allowable blocking rate constraint.

Experimental results

Research questions

• RQ1 How can cloud resources be efficiently allocated among competing virtual machines in a mobile vehicular environment?
• RQ2 What is the optimal amount of reserved resources needed to minimize VM dropping during migration due to vehicle mobility?
• RQ3 How does resource reservation impact the blocking and dropping rates of virtual machines in roadside clouds?
• RQ4 What is the trade-off between resource utilization and service reliability in cloud-based vehicular networks?
• RQ5 How does the game-theoretic approach ensure fairness and efficiency in resource allocation under dynamic conditions?

Key findings

• The proposed resource reservation scheme significantly reduces the dropping rate of migrated virtual machines, especially under high load conditions.
• Simulation results show that with resource reservation, the dropping rate of migrated VMs is substantially lower compared to non-reserved configurations.
• The optimization framework successfully minimizes the dropping rate while keeping the blocking rate within the specified constraint.
• The system achieves a balance between resource utilization and service reliability by reserving capacity for incoming VMs.
• The game-theoretic approach enables efficient and stable resource allocation, improving overall system performance.
• The Markov chain model accurately captures the dynamics of VM arrivals and departures, enabling precise computation of performance metrics.