The adoption of the virtualization paradigm in both computing and networking domains portends a landscape of heterogeneous service capabilities and resources pervasively distributed and interconnected and deeply integ...
详细信息
ISBN:
(纸本)9781479978991
The adoption of the virtualization paradigm in both computing and networking domains portends a landscape of heterogeneous service capabilities and resources pervasively distributed and interconnected and deeply integrated through the 5G network infrastructure. In this service ecosystem, dynamic service demand can be flexibly and elastically accomplished by composing heterogeneous services provisioned over a distributed and virtualized resource infrastructure. Indeed, with the term Virtual functions we refer to virtual computing as well as network service capabilities (e.g., routers and middlebox functions provided as Virtual networkfunctions). In order to cope with the increasingly resource intensive demand, these virtual functions will be deployed in distributed clusters of small-scale datacenters typically located in current exchanges at the network edge and will supplement those deployed in traditional large cloud datacenters. In this work we formulate the problem of composing, computing and networking Virtual functions to select those nodes along the path that minimizes the overall latency (i.e. network and processing latency) in the above mentioned scenario. The optimization problem is formulated as a Resource Constrained Shortest Path problem on an auxiliary layered graph accordingly defined. The layered structure of the graph ensures that the order of VFs specified in the request is preserved. Additional constraints can be also taken into account in the graph construction phase. Finally, we provide a use case preliminary evaluation of the proposed model.
Virtualizing networkfunctions is becoming a major trend in today's research on cloud computing. Among networking elements, the Home Gateway appears to be one with the most diverse functions to handle and thus, wi...
详细信息
ISBN:
(纸本)9781467383868
Virtualizing networkfunctions is becoming a major trend in today's research on cloud computing. Among networking elements, the Home Gateway appears to be one with the most diverse functions to handle and thus, with great potential for virtualization. To this end, the paper proposes a solution to ease adoption by Service Providers of the latest breakthroughs in cloud computing technologies towards a virtualized Home Gateway. Although the NFV approach globally pretends bringing operational advantages in terms of CAPEX and OPEX, it is essential to prove them for Home Gateways scenarios where compatibility and versatility are strong requirements. To achieve this goal, we introduce the concept of Surrogate vNF, which makes Home Gateways NFV aware. The paper highlights a migration path towards full Home Gateway virtualization and proves its concept through a real implementation and a simulated evaluation on a practical use case related to video content distribution.
network function virtualization (NFV) is gaining a lot of interest as a flexible and cost-effective solution for replacing hardware-based, vendor-dependent middleboxes with software appliances running on general purpo...
详细信息
ISBN:
(纸本)9781509001804
network function virtualization (NFV) is gaining a lot of interest as a flexible and cost-effective solution for replacing hardware-based, vendor-dependent middleboxes with software appliances running on general purpose hardware in the cloud. This approach provides an unprecedented degree of flexibility with respect to conventional traffic processing based on middleboxes. In this manuscript a methodology is investigated for the design of general traffic steering policies to implement different service chains through virtual networkfunctions interconnected by a Software Defined networking (SDN) infrastructure. A proof-of-concept implementation on the OpenStack platform is then presented, to provide a practical example of the feasibility and degree of complexity of the proposed approach.
Over the last years, Wireless Local Area networks (WLANs) have been extensively deployed and have significantly evolved. However, the deployment of large-scale WLAN still has management issues. Moreover, while newer W...
详细信息
ISBN:
(纸本)9781467395267
Over the last years, Wireless Local Area networks (WLANs) have been extensively deployed and have significantly evolved. However, the deployment of large-scale WLAN still has management issues. Moreover, while newer WLAN technologies and services have been emerging at a prolific rate, the architecture of WLAN networks has largely been static and difficult to evolve. In this paper, we present a novel architecture for carrier-managed WLAN networks which leverages network function virtualization concepts and virtualization technology in general. It is based on a WLAN Cloudlet which offloads MAC layer processing from access points and consolidates networkfunctions and value-added services. All these functions and services are based on software instances. This brings more flexibility and adaptability and permits to easily implement new services while reducing CAPEX/OPEX and network equipment cost (e.g., access points).
network function virtualization (NFV) brings a cloud service automation paradigm to demand-driven elastic flexing of infrastructure resources. Thus, it is essential to characterize the impact of hardware and virtualiz...
详细信息
ISBN:
(纸本)9781467368841
network function virtualization (NFV) brings a cloud service automation paradigm to demand-driven elastic flexing of infrastructure resources. Thus, it is essential to characterize the impact of hardware and virtualization options on the virtual networkfunction (VNF) performance, and on the load on underlying infrastructure. In this paper, we present VNF characterization case studies with three sample open-source VNF platforms, the Clearwater IMS VNF and two intrusion detection system VNFs (Snort and Suricata). We demonstrate that VNF characterization is vital for optimizing VNF performance, as well as efficient utilization of infrastructure resources. We use the lessons learned from our case studies to design and implement a VNF characterization framework, NFV-VITAL, to characterize VNFs based on user preferences and available resources. We demonstrate how NFV-VITAL can automatically determine optimal configurations under different workloads with the three sample VNFs.
networkfunctions virtualization (NFV) consists of implementing networkfunctions as software applications that can run on general-purpose servers. This paper discusses the application of NFV to the Mobility Managemen...
详细信息
ISBN:
(纸本)9781467395601
networkfunctions virtualization (NFV) consists of implementing networkfunctions as software applications that can run on general-purpose servers. This paper discusses the application of NFV to the Mobility Management Entity (MME), a control plane entity in the Evolved Packet Core (EPC). With the convergence of cloud computing and mobile networks, conventional architectures of network elements need to be re-designed in order to fully harness benefits such as scalability and elasticity. To this end, we design and implement a distributed MME with the three-tier architecture common to web applications. We deploy the components of the distributed MME on two separate OpenStack clouds and evaluate the latency of the attach procedure. We find that the placement of the components within the data center significantly affects the attach latency.
The introduction of network function virtualization (NFV) in carrier-grade networks promises improved operations in terms of flexibility, efficiency, and manageability. NFV is an approach to combine network and comput...
详细信息
ISBN:
(纸本)9781509001804
The introduction of network function virtualization (NFV) in carrier-grade networks promises improved operations in terms of flexibility, efficiency, and manageability. NFV is an approach to combine network and compute virtualizations together. However, network and compute resource domains expose different virtualizations and programmable interfaces. In the UNIFY framework[1] we introduced a joint compute and networkvirtualization and programming. In this paper we analyze how the joint virtualization and control of UNIFY can be applied to a OpenStack-OpenDaylight Data Center. We show through an implementation, that the proposed unified recurring control has clear advantages compared to standard separated cloud and WAN orchestration approaches.
By moving network appliance functionality from proprietary hardware to software, network function virtualization promises to bring the advantages of cloud computing to network packet processing. However, the evolution...
详细信息
ISBN:
(纸本)9781450338349
By moving network appliance functionality from proprietary hardware to software, network function virtualization promises to bring the advantages of cloud computing to network packet processing. However, the evolution of cloud computing (particularly for data analytics) has greatly benefited from application-independent methods for scaling and placement that achieve high efficiency while relieving programmers of these burdens. NFV has no such general management solutions. In this paper, we present a scalable and application-agnostic scheduling framework for packet processing, and compare its performance to current approaches.
network function virtualization (NFV) allows Internet Service Providers (ISPs) to implement key function modules, such as, BRAS (Broadband Remote Access Server), IMS (Internet Multimedia System), etc. in virtual machi...
详细信息
ISBN:
(纸本)9781467397773
network function virtualization (NFV) allows Internet Service Providers (ISPs) to implement key function modules, such as, BRAS (Broadband Remote Access Server), IMS (Internet Multimedia System), etc. in virtual machines in a cloud environment. One of the key problems in NFV implementation is the placement of virtual machines (VMs) in clouds managed by different cloud service providers each with its own management interface. It would be helpful if the clients can implement their policies in a multi-cloud environment using a single interface. Our proposed solution is a modular multi-cloud management system called OpenADN that provides a common interface for resource allocation in a multi-cloud environment. The solution is also applicable to non-ISP applications, such as, banking, financial, and other sectors that need to use globally distributed multi-cloud resources. This paper presents a brief overview of the OpenADN architecture. The key feature of OpenADN is that multiple tenants can share the resources and all resource owners keep complete control over their resources. The data plane module of OpenADN is called OpenADN (Open Application Delivery network). OpenADN has been implemented and brief details of implementation are also presented in this paper.
An increasing demand is seen from enterprises to host and dynamically manage middlebox services in public clouds in order to leverage the same benefits that networkfunctions provide in traditional, in-house deploymen...
详细信息
ISBN:
(纸本)9781467368841
An increasing demand is seen from enterprises to host and dynamically manage middlebox services in public clouds in order to leverage the same benefits that networkfunctions provide in traditional, in-house deployments. However, today's public clouds provide only a limited view and programmability for tenants that challenges flexible deployment of transparent, software-defined networkfunctions. Moreover, current virtual networkfunctions can't take full advantage of a virtualized cloud environment, limiting scalability and fault tolerance. In this paper we review and evaluate the current infrastructural limitations imposed by public cloud providers and present the design and implementation of GNFC, a cloud-based network function virtualization (NFV) framework that gives tenants the ability to transparently attach stateless, container-based networkfunctions to their services hosted in public clouds. We evaluate the proposed system over three public cloud providers (Amazon EC2, Microsoft Azure and Google Compute Engine) and show the effects on end-to-end latency and throughput using various instance types for NFV hosts.
暂无评论