QMORA: A Q-Learning based Multi-objective Resource Allocation Scheme for NFV Orchestration

Fault tolerant placement of stateful VNFs and dynamic fault recovery in cloud networks

In Network Function Virtualization (NFV), traditional purpose-built proprietary middle-boxes are replaced with virtual network function (VNF) instances that are deployed over generic hardware; network traffic flows are steered through a sequence of VNF instances (i.e., that forms a service function chain) after resolving the VNF Forward Graph Embedding (FGE) problem. NFV eases the network management and fosters the agile deployment of network services, thus enabling a foremost solution to address issues of traditional enterprise networks. With the everlasting Internet traffic growth, in this work, we extend the provision of NFV upon Elastic Optical Networks (EONs) that deliver abundant elastic and fine-granular bandwidth for connections between VNFs. We systematically examine the integrated NFV-EON architecture, and investigate one key enabler, namely the FGE-EON problem, where one has to coordinate the allocation of computing resources (as in a typical FGE problem) and the assignment of optical bandwidth (i.e., spectrum allocation in EONs). The contributions of this work are multi-fold. The FGE-EON problem is proven to be NP-Complete for the first time. An Integer Linear Programming (ILP) model is developed to address FGE-EON problem of small instances. Despite its complexity, with novel modeling techniques, the proposed ILP model maintains a compact structure even after accounting for VNF sharing and VNF co-location together for the first time. For large scale problems, we design a heuristic algorithm based on layered graph that is shown to be near-optimal in our evaluations.

Middleboxes have become a vital part of modern networks by providing service functions such as content filtering, load balancing and optimization of network traffic. An ordered sequence of middleboxes composing a logical service is called service chain. Service Function Chaining (SFC) enables us to define these service chains. Recent optimization models of SFCs assume that the functionality of a middlebox is provided by a single software appliance, commonly known as Virtual Network Function (VNF). This assumption limits SFCs to the throughput of an individual VNF and resources of a physical machine hosting the VNF instance. Moreover, typical service providers offer VNFs with heterogeneous throughput and resource configurations. Thus, deploying a service chain with custom throughput can become a tedious process of stitching heterogeneous VNF instances. In this paper, we describe how we can overcome these limitations without worrying about underlying VNF configurations and resource constraints. This prospect is achieved by distributed deploying multiple VNF instances providing the functionality of a middlebox and modeling the optimal deployment of a service chain as a mixed integer programming problem. The proposed model optimizes host and bandwidth resources allocation, and determines the optimal placement of VNF instances, while balancing workload and routing traffic among these VNF instances. We show that this problem is NP-Hard and propose a heuristic solution called Kariz. Kariz utilizes a tuning parameter to control the trade-off between speed and accuracy of the solution. Finally, our solution is evaluated using simulations in data-center networks.

Network Function Virtualization (NFV) is a paradigm that virtualizes traditional network functions and instantiates Virtual Network Functions (VNFs) as software instances separate from hardware appliances. Service Chaining (SC), seen as one of the major NFV use cases, provides customized services to users by concatenating VNFs. A VNF placement model for SC that relaxes the visit order constraints of requested VNFs has been considered. Relaxing the VNF visit order constraints reduces the number of VNFs which need to be placed in the network. However, since the model does not permit any loop within an SC path, the efficiency of utilization of computation resources deteriorates in some topologies. This paper proposes a VNF placement model for SC which minimizes the cost for placing VNFs and utilizing link capacity while allowing both relaxation of VNF visit order constraints and configuration of SC paths including loops. The proposed model determines routes of requested SC paths, which can have loops, by introducing a logical layered network generated from an original physical network. This model is formulated as an Integer Linear Programming (ILP) problem. A heuristic algorithm is introduced for the case that the ILP problem is not tractable. Simulation results show that the proposed model provides SC paths with smaller cost compared to the conventional model.

With the constant demand for connectivity at an all-time high, Network Service Providers (NSPs) are required to optimize their networks to cope with rising capital and operational expenditures required to meet the growing connectivity demand. A solution to this challenge was presented through Network Function Virtualization (NFV). As network complexity increases and futuristic networks take shape, NSPs are required to incorporate an increasing amount of operational efficiency into their NFV-enabled networks. One such technique is Machine Learning (ML), which has been applied to various entities in NFV-enabled networks, most notably in the NFV Orchestrator. While traditional ML provides tremendous operational efficiencies, including real-time and high-volume data processing, challenges such as privacy, security, scalability, transferability, and concept drift hinder its widespread implementation. Through the adoption of Advanced Intelligence techniques such as Reinforcement Learning and Federated Learning, NSPs can leverage the benefits of traditional ML while simultaneously addressing the major challenges traditionally associated with it. This work presents the benefits of adopting these advanced techniques, provides a list of potential use cases and research topics, and proposes a bottom-up micro-functionality approach to applying these methods of Advanced Intelligence to NFV Management and Orchestration.

With the emerging paradigm of Virtual Network and Network Function Virtualization (NFV), the data center operator can flexibly manage the network to reduce Operating Expenditures (OPEX) and Capital Expenditures (CAPEX). Some new issues including the assignment of Virtual Network Functions (VNFs) and Service Function Chain (SFC) scheduling should be considered to apply in the communication. Mathematically, the SFC scheduling process can be formulated as a queuing system, and every server installed with VNF instances can be regarded as a service node, where batched requests are parallel submitted and processed one after another. In the management of SFC, the cloud service provider needs to resolve the problem of network congestion. Latency and throughput are two important but contradictory indexes in the network management, and both of them are deserved to be optimized during the scheduling of SFC requests. In this paper, different from most existing studies, we focus on the service rate control problem in the scheduling of SFC requests. Firstly, we formulate this problem as an integer programming problem. Through a cooperative game approach, a Nash bargaining based model is proposed to jointly optimize the latency and throughput, which is proven to provide fair performance guarantees by both theoretical analysis and simulation. To improve the scalability of the proposed algorithm, we also design a polynomial two-phase heuristic to perform Pareto optimization. Simulation evaluation shows that the proposed algorithm can implement balanced traffic scheduling and avoid excessive server latency caused by network congestion.

The discovery process of OPC Unified Architecture (OPC-UA) is well defined but lacks flexibility when implemented in large networks. While the standard provides concepts for local and global discovery services, the core assumption is that all clients and servers are aware of the discovery server(s) in the network. This dependency is a barrier for Plug-and-Play installations and is overcome by self announces via the multicast Domain Name System (mDNS), however, this scales badly with an increasing number of network nodes and deployed servers. We propose a discovery approach based on Software-Defined Networking (SDN) and Network Function Virtualization (NFV), that installs a plug-and-play mechanism for global OPC-UA discovery in modern virtualized networks.

Joint optimization of Service Chain Graph Design and Mapping in NFV-enabled networks

Network Function Virtualization (NFV) has promoted the implementation of network functions from expensive hardwares to software middleboxes. These software middleboxes, also called Virtual Network Functions (VNFs), are executed on switch-connected servers. Efficiently deploying such VNFs on servers is challenging because the traffic rate of flows must be fully processed by their requested VNFs when they reach destinations, and the deployed positions of VNFs are restricted by the server capacity. In addition, each network function offers non-uniform VNF models (types) with different configurations of processing volumes and costs. This paper focuses on minimizing the total cost of deploying VNFs for providing a specific network function to all flows in tree-structured networks. First, we prove the NP-hardness of non-uniform VNF deployment in a tree topology and propose a dynamic programming based solution with a pseudo-polynomial time complexity. Then we narrow it down to three simplified cases by focusing on either uniform VNFs or the linear line topology. Specifically, three algorithms are introduced: an improved dynamic programming based algorithm for deploying uniform VNFs in a tree topology, a performance-guaranteed algorithm for deploying non-uniform VNFs in a linear line topology, and an optimal greedy algorithm for deploying uniform VNFs in a linear line topology. Additionally, we generalize our approach to a case of deploying a service chain, which consists of multiple network functions applied to flows in a specific order. We propose two solutions: one is optimal but time-consuming while another is heuristic but efficient. Extensive simulations are conducted to evaluate our algorithms.

5G networks represent a paradigm shift from their 4G predecessors, engineered to cater to the burgeoning demands of densely populated areas and the high data throughput requirements of the Internet of Things (IoT) and smart city infrastructures. Beyond enhancing fundamental network attributes such as reliability, availability, and security, 5G introduces architectural innovations to accommodate these advanced needs. This paper delves into the architectural nuances of 5G, focusing primarily on the Standalone (SA) deployment method, distinguished by its utilization of network slicing. This approach leverages cutting-edge virtual network technologies, including Software-Defined Networking (SDN) and Network Function Virtualization (NFV), to forge virtual network slices tailored to diverse consumer needs. Such customization is imperative for meeting the expansive and varied connectivity demands characteristic of 5G networks.To navigate the complexities of 5G architecture, this study advocates for the adoption of Reference Architectures (RAs), which provide a high-level abstraction framework for dissecting and understanding intricate systems. By proposing a model that outlines the standalone architecture of 5G networks, our research lays the groundwork for evolving this model into a comprehensive RA for 5G. This foundational work is instrumental in equipping system designers with the tools necessary for generalizing systems to manage design and deployment challenges effectively, as well as the heterogeneity inherent in system components. Additionally, our proposed model facilitates the establishment of a common vocabulary, promoting a coherent understanding of component interactions and offering a unified system perspective. Ultimately, this paper contributes a pivotal model that not only enhances our comprehension of 5G network architecture but also paves the way for future advancements in network design and deployment, promising to significantly impact the development of smart cities and IoT ecosystems.Keywords: 5G, 5G architecture, Reference architecture, patterns. Network slicing, Network Function Virtualization, Virtual Network Function.

With the increasing number of providers at a rapid speed in the cloud computing market. Technologies are developing into more complex and interconnected entities. Virtual Network Functions (VNFs), Service Function Chains (SFC), and multi-cloud environments have appeared as emerging developments, which enable Virtual services to be hosted dynamically over multiple clouds, it can be done with greater flexibility in efficiency and cost optimization. On the other hand, the growing demand for low-latency network services is progressively demanding new and creative placement techniques. The problem of optimal Network Function (NF) provision, which involves deciding the optimal position of VNF, on which nodes these functions must be located, and how VNF instances are assigned to Service Function Chains (SFCs), is considered in this work. to active optimal utilization of the resource while meeting the end-to-end latency of the service chain. With large-scale web server workloads, we test our algorithm in the simulation environment by comparing it with the existing state-of-the-art solutions [1], that are unaware of the critical application. The result shows that the proposed algorithm has more advantages than the state-of-the-art as around 85.9% of critical requests are served from a single cloud data center.

The rapid development of network function virtualization (NFV) enables a communication network to provide in-network services using virtual network functions (VNFs) deployed on general IT hardware. While existing studies on NFV focused on how to provision VNFs from the provider's perspective, little is done about how to validate the provisioned resources from the user's perspective. In this work, we take a first step towards this problem by developing an inference framework designed to “look into” the NFV network. Our framework infers the structure and state of the overlay formed by VNF instances, ingress/egress points of measurement flows, and critical points on their paths (branching/joining points). Our solution only uses external observations such as the required service chains and the end-to-end performance measurements. Besides the novel application scenario, our work also fundamentally advances the state of the art on topology inference by considering (i) general topologies with general measurement paths, and (ii) information of service chains. Our evaluations show that the proposed solution significantly improves both the reconstruction accuracy and the inference accuracy over existing solutions, and service chain information is critical in revealing the structure of the underlying topology.

Network Function Virtualization (NFV) is a promising network architecture concept, in which virtualization technologies are employed to manage networking functions via software as opposed to having to rely on hardware to handle these functions. By shifting dedicated, hardware-based network function processing to software running on commoditized hardware, NFV has the potential to make the provisioning of network functions more flexible and cost-effective, to mention just a few anticipated benefits. Despite consistent initial efforts to make NFV a reality, little has been done towards efficiently placing virtual network functions and deploying service function chains (SFC). With respect to this particular research problem, it is important to make sure resource allocation is carefully performed and orchestrated, preventing over- or under-provisioning of resources and keeping end-to-end delays comparable to those observed in traditional middlebox-based networks. In this paper, we formalize the network function placement and chaining problem and propose an Integer Linear Programming (ILP) model to solve it. Additionally, in order to cope with large infrastructures, we propose a heuristic procedure for efficiently guiding the ILP solver towards feasible, near-optimal solutions. Results show that the proposed model leads to a reduction of up to 25% in end-to-end delays (in comparison to chainings observed in traditional infrastructures) and an acceptable resource over-provisioning limited to 4%. Further, we demonstrate that our heuristic approach is able to find solutions that are very close to optimality while delivering results in a timely manner.

Middleboxes have become an integral part of modern enterprise and data center networks. They are used for realizing various performance and security objectives. Most middlboxes (e.g., firewalls, Intrusion Detection Systems (IDSs), Network Address Translators (NATs), etc.) are dedicated hardware appliances. However, recent advancements in cloud and virtualization technologies have fueled the concept of Virtual Middleboxes or Virtual Network Functions (VNFs) along with a new research field known as Network Function Virtualization (NFV). This area of research has gained a lot of traction from both industry and academia. Although much progress has been made in NFV technology, a crucial component for realizing the primary objective of NFV is still missing – a management and orchestration [7] system that conforms to the principles of NFV: open source, open API and standardized software solutions. Without this feature, network operators may end-up with the same situation of vendor lock-in as with proprietary hardware middleboxes. A number of recent proposals like Stratos [8], OpenNF [9], and Split/Merge [10], strive to fulfill the requirements for VNF management and orchestration. However, they propose incompatible northbound APIs. What is really needed is a standardized API that is flexible enough to express a wide range of NFV management and orchestration operations. History shows that standardization efforts usually take a long time and often are futile. Hence, we take a different approach, and propose to use an existing, well known, standardized interface for NFV management and orchestration: the Linux file system interface.

The integration of software defined networking (SDN) and network function virtualization (NFV) is a promising approach to support flexible service deployment and provisioning. In particular, NFV enables a network service to be represented as one network service function chain, which is composed of several virtual network functions in a specific order. However, with regard to service provisioning in multidomain SDN networks, virtual network functions are deployed in several different SDN domains due to limited resource capacity of each SDN domain. It results in the network service deployment complexity increase and causes additional operating cost. On the other hand, most current network service deployment proposals are limited to static user request scenarios. How to effectively deploy network service across multiple SDN domains to satisfy dynamic service demands of end users is a challenging problem. To address the aforementioned challenge, in this paper, we study the problem of dynamic network service deployment across multiple SDN domains. Specifically, we first design a novel service deployment framework to support dynamic deployment and optimization of virtual network function in multidomain SDN networks. Then, we formulate the problem as a multiobjective optimization problem with the target of jointly minimizing resource consumption cost and operating cost. Furthermore, we transform the multiobjective optimization problem into a single objective optimization model and present an efficient heuristic service deployment approach to solve it by leveraging Dijkstra algorithm. The simulation results demonstrate that the proposed service deployment method is efficient and outperforms comparison methods in terms of service rejection ratio, resource cost, and operating cost.