To establish a network service in network functions virtualization (NFV) networks, the orchestrator addresses the challenge of service chaining and virtual network function placement (SC-VNFP) by mapping virtual network functions (VNFs) and virtual links onto physical nodes and links. Unlike traditional networks, network operators in NFV networks must contend with both hardware and software failures in order to ensure resilient network services, as NFV networks consist of physical nodes and software-based VNFs. To guarantee network service quality in NFV networks, the existing work has proposed an approach for the SC-VNFP problem that considers VNF diversity and redundancy. VNF diversity splits a single VNF into multiple lightweight replica instances that possess the same functionality as the original VNF, which are then executed in a distributed manner. VNF redundancy, on the other hand, deploys backup instances with standby mode on physical nodes to prepare for potential VNF failures. However, the existing approach does not adequately consider the tradeoff between resource efficiency and service availability in the context of VNF diversity and redundancy. In this paper, we formulate the SC-VNFP problem with VNF diversity and redundancy as a two-step integer linear program (ILP) that adjusts the balance between service availability and resource efficiency. Through numerical experiments, we demonstrate the fundamental characteristics of the proposed ILP, including the tradeoff between resource efficiency and service availability.

In the coming smart-home era, more and more household electrical appliances are generating more and more sensor data and transmitting them over the home networks, which are often connected to Internet through Point-to-Point Protocol over Ethernet (PPPoE) for desirable authentication and accounting. However, according to our knowledge, high-speed commercial home PPPoE router is still absent for a home network environment. In this paper, we first introduce and evaluate our programmable platform FLARE-DPDK for ease of programming network functions. Then we introduce our effort to build a compact 10Gbps software FLARE PPPoE router on a commercial mini-PC. In our implementation, the control plane is implemented with Linux PPPoE software for authentication-like signaling control. The data plane is implemented over FLARE-DPDK platform, where we get packets from physical network interfaces directly bypassing Linux kernel and distribute packets to multiple CPU cores for data processing in parallel. We verify our software PPPoE router in both lab and production network environment. The experimental results show that our FLARE software PPPoE router can achieve much higher throughput than a commercial PPPoE router tested in a production environment.

Many studies have revealed that the performance of software-based Virtual Network Functions (VNFs) is insufficient for mission-critical networks. Scaling-out approaches, such as auto-scaling of VNFs, could handle a huge amount of traffic; however, the exponential traffic growth confronts us the limitations of both expandability of physical resources and complexity of their management. In this paper, we propose a fast datapath processing method called Packet Aggregation Flow (PA-Flow) that is based on hop-by-hop packet aggregation for more efficient Service Function Chaining (SFC). PA-Flow extends a notion of existing intra-node packet aggregation toward network-wide packet aggregation, and we introduce following three novel features. First, packet I/O overheads at intermediate network devices including NFV-nodes are mitigated by reduction of packet amount. Second, aggregated packets are further aggregated as going through the service chain in a hop-by-hop manner. Finally, next-hop aware packet aggregation is realized using OpenFlow-based flow tables. PA-Flow is designed to be available with various VNF forms (e.g. VM/container/baremetal-based) and virtual I/O technologies (e.g. vhost-user/SR-IOV), and its implementation does not bring noticeable delay for aggregation. We conducted two evaluations: (i) a baseline evaluation for understanding fundamental performance characteristics of PA-Flow (ii) a simulation-based SFC evaluation for proving PA-Flow's effect in a realistic environment. The results showed that throughput of short packet forwarding was improved by 4 times. Moreover, the total number of packets was reduced by 93% in a large-scale SFC.

Network Functions Virtualization (NFV) is expected to provide network systems that offer significantly lower cost and greatly flexibility to network service providers and their users. Unfortunately, it is extremely difficult to implement Virtualized Network Functions (VNFs) that can equal the performance of Physical Network Functions. To realize NFV systems that have adequate performance, it is critical to accurately grasp VNF workload. In this paper, we focus on the virtual firewall as a representative VNF. The workload of the virtual firewall is mostly determined by firewall rule processing and the Access Control List (ACL) configurations. Therefore, we first reveal the major factors influencing the workload of the virtual firewall and some issues of monitoring CPU load as a traditional way of understanding the workload of virtual firewalls through preliminary experiments. Additionally, we propose a new workload metric for the virtual firewall that is derived by mathematical models of the firewall workload in consideration of the packet processing in each rule and the ACL configurations. Furthermore, we show the effectiveness of the proposed workload metric through various experiments.

Mobile virtual network operators (MVNOs) are mobile operators without their own infrastructure or government issued spectrum licenses. They purchase spectrum resources from primary mobile network operators (MNOs) to provide communication services under their own brands. MVNOs are expected to play an important role in mobile network markets, as this will increase the competition in retail markets and help to meet the demand of niche markets. However, with the rapidly increasing demand of mobile data traffic, efficient utilization of the limited spectrum resources owned by MVNOs has become an important issue. We propose here a resource sharing mechanism between MVNOs against the background of network functions virtualization (NFV). The proposed mechanism enables MVNOs to improve their quality of service (QoS) by sharing spectrum resources with each other. A nash bargaining solution based decision strategy is also devised to ensure the fairness of resource sharing. Extensive numerical evaluation results validate the effectiveness of the proposed models and mechanisms.

Due to the recent network service market trends, network infrastructure providers must make their network infrastructures tolerant of network service complexity and swift at providing new network services. To achieve this, we first make a design decision for the single domain network infrastructure in which we use network virtualization and separate the network service control and management from the network infrastructure and leave the resource connectivity control and management in the network infrastructure so that the infrastructure can maintain simplicity and the network service can become complex and be quickly provided. Along with the decision, we construct an architecture of the network infrastructure and a network management model. The management model defines a slice as being determined by abstracted resource requirements and restructures the roles and planes from the viewpoint of network infrastructure usability so that network service requesters can manage network resources freely and swiftly in an abstract manner within the authorities the network infrastructure operator provides. We give the details of our design and implementation for a network virtualization management system along with the model. We deployed and evaluated our designed and implemented management system on the Japan national R&E testbed (JGN-X) to confirm the feasibility of our management system design and discuss room for improvement in terms of response time and scalability towards practical use. We also investigated certain cases of sophisticated network functions to confirm that the infrastructure can accept these functions without having to be modified.

The use of computing resources on network is becoming active in the Internet and private networks. OpenFlow/Software-Defined Networking (SDN) is drawing attention as a method to control network virtualization for the cloud computing services and other carrier services. This paper introduces examples of OpenFlow/SDN technologies applied to commercial cloud services. Various activities to expand coverage over commercial carrier networks are also mentioned.