Optimal Virtual Network Function Placement and Resource Allocation in Multi-Cloud Service Function Chaining Architecture (1903.11550v1)

Abstract: Service Function Chaining (SFC) is the problem of deploying various network service instances over geographically distributed data centers and providing inter-connectivity among them. The goal is to enable the network traffic to flow smoothly through the underlying network, resulting in an optimal quality of experience to the end-users. Proper chaining of network functions leads to optimal utilization of distributed resources. This has been a de-facto model in the telecom industry with network functions deployed over underlying hardware. Though this model has served the telecom industry well so far, it has been adapted mostly to suit the static behavior of network services and service demands due to the deployment of the services directly over physical resources. This results in network ossification with larger delays to the end-users, especially with the data-centric model in which the computational resources are moving closer to end users. A novel networking paradigm, Network Function Virtualization (NFV), meets the user demands dynamically and reduces operational expenses (OpEx) and capital expenditures (CapEx), by implementing network functions in the software layer known as virtual network functions (VNFs). VNFs are then interconnected to form a complete end-to-end service, also known as service function chains (SFCs). In this work, we study the problem of deploying service function chains over network function virtualized architecture. Specifically, we study virtual network function placement problem for the optimal SFC formation across geographically distributed clouds. We set up the problem of minimizing inter-cloud traffic and response time in a multi-cloud scenario as an ILP optimization problem, along with important constraints such as total deployment costs and service level agreements (SLAs). We consider link delays and computational delays in our model.

• The paper introduces a novel Affinity-Based Allocation (ABA) heuristic designed for efficiently solving Virtual Network Function (VNF) placement problems in large multi-cloud Service Function Chaining (SFC) architectures.
• It utilizes an Integer Linear Programming (ILP) model for optimal solutions in smaller network topologies but highlights its computational limitations for large-scale applications.
• The ABA heuristic efficiently outperforms greedy methods like FFD by significantly reducing inter-cloud traffic and delays in large network configurations while achieving near-optimal placement.

Virtual Network Function Placement in Multi-Cloud Environments: An Analytical Perspective

The paper presents an analytical approach to optimize the placement of Virtual Network Functions (VNFs) in a multi-cloud framework, focusing on Service Function Chaining (SFC) architecture. The paper addresses the inherent challenges associated with deploying VNFs across geographically distributed cloud environments, with an emphasis on minimizing inter-cloud traffic and response times while adhering to constraints such as deployment costs and Service Level Agreements (SLAs).

At the core of this research is the Integer Linear Programming (ILP) model, which serves as a foundational technique for achieving optimal solutions for VNF placement. However, the authors acknowledge the scalability limitations of ILP concerning larger network topologies. Consequently, they introduce a novel Affinity-Based Allocation (ABA) heuristic as an alternative approach to efficiently solve placement problems in real-time scenarios for larger network settings.

Key Contributions and Findings

• ILP Model for Optimal Placement: The ILP model targets minimizing inter-cloud delays and operational costs by carefully placing VNFs to optimize traffic flows. Despite its potential accuracy, the complexity of ILP limits its practical application to networks with smaller clusters.
• Affinity-Based Allocation Heuristic: To circumvent the scalability issues of ILP, the authors propose the ABA heuristic. This method accounts for traffic affinities among VNFs, thereby reducing inter-cloud traffic and delays. It demonstrates a performance gap of less than 10% compared to the optimal ILP solution, ensuring quicker execution while maintaining solution quality.
• Comparison With Greedy Approach: The paper compares ABA with the widely adopted Greedy First-Fit Decreasing (FFD) method. While the FFD approach exhibits faster execution times, it considerably lacks the solution quality displayed by ABA, especially in environments with high traffic loads and complex SLAs.

Numerical Results

The research presents several performance evaluations demonstrating the efficacy of the ABA heuristic. Notably, the ABA approach results in reduced inter-cloud traffic and operational costs compared to the greedy FFD method. For instance, the ABA method achieved a 50% reduction in total delays in large topology setups compared to the FFD approach, exemplifying its suitability for larger networks with stringent SLA constraints.

Implications and Future Directions

This paper advances the computational strategies for virtual network function placement by introducing traffic-aware heuristics. The implications extend toward more efficient resource utilization in multi-cloud platforms, which is crucial as cloud-based services continue to expand in complexity and reach.

Future research could explore the integration of advanced machine learning techniques to dynamically adapt VNF placement based on real-time traffic patterns and SLA adjustments. Moreover, additional constraints such as energy consumption and security requirements could be incorporated into the optimization models to align with evolving industry standards. This trajectory will contribute to refining the deployment strategies in Network Function Virtualization (NFV) and further the practical implementation of resilient SFC architectures.

The analytical insights from this paper provide a valuable foundation for researchers and practitioners aiming to enhance VNF placement efficacy in multi-cloud environments. By leveraging the ABA heuristic, service providers can achieve better resource alignment with user expectations, ultimately improving service delivery and operational efficiency on a global scale.