Network Operators’ networks are populated with a large and
increasing variety of proprietary hardware appliances. To launch a new network
service often requires yet another variety and finding the space and power to
accommodate these boxes is becoming increasingly difficult; compounded by the
increasing costs of energy, capital investment challenges and the rarity of
skills necessary to design, integrate and operate increasingly complex
hardware-based appliances. Moreover, hardware-based appliances rapidly reach
end of life, requiring much of the procure- design-integrate-deploy
Network Functions Virtualization aims to address these
problems by leveraging standard IT virtualization technology to consolidate
many network equipment types onto industry standard high volume servers,
switches and storage, which could be located in Datacenters, Network Nodes and
in the end user premises. We believe Network Functions Virtualization is
applicable to any data plane packet processing and control plane function in
fixed and mobile network infrastructures.
We would like to emphasize that we see Network Functions
Virtualization as highly complementary to Software Defined Networking (SDN).
These topics are mutually beneficial but are not dependent on each other.
Network Functions can be virtualized and deployed without an SDN being required
and vice-versa.
Virtualizing Network Functions could potentially offer
many benefits including, but not limited to:
- Reduced equipment costs and reduced power consumption through consolidating equipment and exploiting the economies of scale of the IT industry.
- Increased speed of Time to Market by minimizing the typical network operator cycle of innovation. Economies of scale required to cover investments in hardware-based functionalities are no longer applicable for software-based development, making feasible other modes of feature evolution. Network Functions Virtualization should enable network operators to significantly reduce the maturation cycle.
- Availability of network appliance multi-version and multi-tenancy, which allows use of a single platform for different applications, users and tenants. This allows network operators to share resources across services and across different customer bases.
- Targeted service introduction based on geography or customer sets is possible. Services can be rapidly scaled up/down as required.
- Enables a wide variety of eco-systems and encourages openness. It opens the virtual appliance market to pure software entrants, small players and academia, encouraging more innovation to bring new services and new revenue streams quickly at much lower risk.
To leverage these benefits, there are a number of technical challenges
which need to be addressed:
- Achieving high performance virtualized network appliances which are portable between different hardware vendors, and with different hypervisors.
- Achieving co-existence with bespoke hardware based network platforms whilst enabling an efficient migration path to fully virtualized network platforms which re-use network operator OSS/BSS. OSS/BSS development needs to move to a model in-line with Network Functions Virtualization and this is where SDN can play a role.
- Managing and orchestrating many virtual network appliances (particularly alongside legacy management systems) while ensuring security from attack and misconfiguration.
- Network Functions Virtualization will only scale if all of the functions can be automated.
- Ensuring the appropriate level of resilience to hardware and software failures.
- Integrating multiple virtual appliances from different vendors. Network operators need to be able to “mix & match” hardware from different vendors, hypervisors from different vendors and virtual appliances from different vendors without incurring significant integration costs and avoiding lock-in.
Solutions to these technical challenges are available, or
could be made available, it recommend that the IT and Networks industries
combine their complementary expertise and resources in a joint collaborative
effort to reach broad agreement on
standardized approaches and common architectures which address these technical
challenges, and which are interoperable and have economies of scale.
To accelerate progress, a new network operator-led
Industry Specification Group (ISG) with open membership is being setup under
the auspices of ETSI to work through the technical challenges for Network
Functions Virtualization as outlined in this white paper. The formal creation
process of this ETSI
ISG has been already started.
Introduction
Network operators’ networks are populated with a large and
increasing variety of proprietary hardware appliances. To launch a new network
service often requires yet another variety and finding the space and power to
accommodate these boxes is becoming increasingly difficult; compounded by the
increasing costs of energy, capital investment challenges and the rarity of
skills necessary to design, integrate and operate increasingly complex
hardware-based appliances. Moreover, hardware-based appliances rapidly reach
end of life, requiring much of the procure- design-integrate-deploy cycle to be
repeated with little or no revenue benefit. Worse, hardware lifecycles are
becoming shorter as technology and services innovation accelerates, inhibiting
the roll out of new revenue earning network services and constraining
innovation in an increasingly Network-centric
connected world.
Definition
Network Functions Virtualization aims to transform the way
that network operators architect networks by evolving standard IT
virtualization technology to consolidate many network equipment types onto
industry standard high volume servers, switches and storage, which could be
located in Datacenters, Network Nodes and in the end user premises, as
illustrated in Figure 1. It involves the implementation of network functions in
software that can run on a range of industry standard server hardware, and that
can be moved to, or instantiated in, various locations in the network as
required, without the need for installation of new equipment.
 |
| Figure 1: Vision for Network Functions Virtualization |
Relationship with Software Defined
Networks (SDN)
As
shown in Figure 2, Network Functions Virtualization is highly complementary to
Software Defined Networking (SDN), but not dependent on it (or vice-versa).
Network Functions Virtualization can be implemented without a SDN being
required, although the two concepts and solutions can be combined and
potentially greater value accrued.
 |
| Figure 2: Network Functions Virtualization Relationship with SDN |
Network Functions Virtualization goals can be achieved
using non-SDN mechanisms, relying on the techniques currently in use in many datacenters.
But approaches relying on the separation of the control and data forwarding
planes as proposed by SDN can enhance performance, simplify compatibility with
existing deployments, and facilitate operation and maintenance procedures.
Network Functions Virtualization is able to support SDN by
providing the infrastructure upon which the SDN software can be run.
Furthermore, Network Functions Virtualization aligns closely with the SDN
objectives to use commodity servers and switches. ETSI
ISG work closely with Open Networking Foundation (ONF) combine
NFV & SDN
Fields of Application and
Use Cases
Network Functions Virtualization is applicable to any data
plane packet processing and control plane function in mobile and fixed
networks. Potential examples that can be listed include (not in any particular
order):
- Switching elements: BNG, CG-NAT, routers.
- Mobile network nodes: HLR/HSS, MME, SGSN, GGSN/PDN-GW, RNC, Node B, eNode B.
- Functions contained in home routers and set top boxes to create virtualized home environments.
- Tunneling gateway elements: IPSec/SSL VPN gateways.
- Traffic analysis: DPI, QoE measurement.
- Service Assurance, SLA monitoring, Test and Diagnostics.
- NGN signaling: SBCs, IMS.
- Converged and network-wide functions: AAA servers, policy control and charging platforms.
- Application-level optimization: CDNs, Cache Servers, Load Balancers, Application Accelerators.
- Security functions: Firewalls, virus scanners, intrusion detection systems, spam protection
Further studies are required to identify those network
functions for which virtualization yields most benefits. Use cases of interest
include (not in any particular order):
- A software-based DPI, providing advanced traffic analysis and multi-dimensional reporting, and showing the possibility of making off-the-shelf hardware work at actual line rates. Software-based DPI can be pervasively deployed in the network, providing much better analysis capabilities, as well as simpler mechanisms for deployment, update, testing, and to scale it to changing workloads.
- IP node implementations, supporting - for example, but not
limited to: CG-NAT and BRAS capabilities on standard high-end servers, offering
the opportunity for an effective re-use of hardware as the demand for such
capabilities evolves.
- The virtualization of services and capabilities that presently require dedicated hardware appliances on customer premises (home environment to small branch office to large corporate premises), including but not restricted to: firewall, web security, IPS/IDS, WAN acceleration and optimization, and router functions. The virtualization of the home environment including routers, hubs and set top boxes would potentially enable a simpler and seamless migration to IPv6, reduce energy consumption and avoid successive hardware updates as broadband applications and services evolve.
- The virtualization of Content Distribution Networks (CDN), with the initial goal of extending and scaling Content Delivery Services more easily, and also with the objective of maximizing hardware re-use in PoPs by being able to install other Service Delivery Applications (e.g. Web Acceleration) on demand. Virtualization of CDNs will also allow the hosting of CDN services from potential business partners, like external CDN providers.
- The virtualization of a mobile core network targeting at a more cost efficient production environment, which allows network operators to cope with the increasing traffic demand in mobile networks, and leading to better resource utilization (including energy savings), more flexible network management (no need to change hardware for nodes’ upgrades), hardware consolidation, easier multi-tenancy support and faster configuration of new services. Network Functions Virtualization in mobile networks can also be used to create core network instances optimized for specific services, e.g. for Machine-to-Machine communications (M2M).
- Co-ordinated implementation of cloud and networking for enterprises, allowing on-demand services to be offered and providing capital efficiency for enterprise customers and network operators.
- Hybrid fiber-DSL nodes are located deep in the external network in street cabinets, underground and on poles. These nodes must be very low power consumption and very low/zero maintenance to be economic. Virtualization could be used to reduce hardware complexity at the remote node, saving energy and providing an enhanced degree of future proofing as services evolve. These remote nodes could more economically provide both fixed and wireless access if key functions were virtualized on a common platform.
- Network Functions Virtualization can also be used to provide an efficient production environment which can commonly be used by different applications, users and tenants, thus supporting the coexistence of several versions and variants of a network service (including test versions and beta versions).
Benefits of Network Functions Virtualization
We believe the application of Network Functions Virtualization
brings many benefits to network operators, contributing to a dramatic change in
the telecommunications industry landscape. Benefits we foresee include (not in
any particular order):
- Reduced equipment costs and reduced power consumption through consolidating equipment and exploiting the economies of scale of the IT industry (~9.5M Servers shipped in 2011 compared with ~1.5M routers forecast for 2012).
- Increased velocity of Time to Market by minimizing the typical network operator cycle of innovation. Economies of scale required to cover investments in hardware-based functionalities are no longer applicable for software-based development, making feasible other modes of feature evolution. Network Functions Virtualization should enable network operators to significantly reduce the maturation cycle.
- The possibility of running production, test and reference facilities on the same infrastructure provides much more efficient test and integration, reducing development costs and time to market.
- Targeted service introduction based on geography or customer sets is possible. Services can be rapidly scaled up/down as required. In addition, service velocity is improved by provisioning remotely in software without any site visits required to install new hardware.
- Enabling a wide variety of eco-systems and encouraging openness. It opens the virtual appliance market to pure software entrants, small players and academia, encouraging more innovation to bring new services and new revenue streams quickly at much lower risk.
- Optimizing network configuration and/or topology in near
real time based on the actual traffic/mobility patterns and service demand. For
example, optimization of the location & assignment of resources to network
functions automatically and in near real time could provide protection against
failures without engineering full 1+1 resiliency.
- Supporting multi-tenancy thereby allowing network operators to provide tailored services and connectivity for multiple users, applications or internal systems or other network operators, all co-existing on the same hardware with appropriate secure separation of administrative domains.
- Reduced energy consumption by exploiting power management
features in standard servers and storage, as well as workload consolidation and location optimization. For example, relying on virtualization techniques it would be possible to concentrate the workload on a smaller number of servers during off-peak hours (e.g. overnight) so that all the other servers can be switched off or put into an energy saving mode.
- Improved operational efficiency by taking advantage of the higher uniformity of the physical network platform and its homogeneity to other support platforms:
Ø IT
orchestration mechanisms provide automated installation, scaling-up and
scaling- out of capacity, and re-use of Virtual Machine (VM) builds.
Ø Eliminating
the need for application-specific hardware. The skills base across the
industry for operating standard high volume IT servers is much larger and less
fragmented than for today’s telecom-specific network equipment.
Ø Reduction
in variety of equipment for planning & provisioning. Assuming tools are
developed for automation and to deal with the increased software complexity of
virtualization.
Ø Option
to temporarily repair failures by automated re-configuration and moving network
workloads onto spare capacity using IT orchestration mechanisms. This could be
used to reduce the cost of 24/7 operations by mitigating failures automatically.
Ø The
potential to gain more efficiency between IT and Network Operations.
Ø The
potential to support in-service software upgrade (ISSU) with easy reversion by
installing the new version of a Virtualized Network Appliance (VNA) as a new
Virtual Machine (VM). Assuming traffic can be transferred from the old VM to
the new VM without interrupting service. For some applications it may be
necessary to synchronize the state of the new VM with the old VM.
The Changing Telecoms Industry
Landscape
Although Network Functions Virtualization brings many
advantages to the telecommunications industry it is likely to transform the
vendor landscape. Each player will need to position/re-position itself in the
new Network Functions Virtualization market.
This is not as disruptive as it may seem because network
equipment vendors already implement some of their solutions by combining their
proprietary software with industry standard hardware and software components,
but in a proprietary way. Enabling their proprietary software to run on
industry standard hardware in a standardized way may be a significant
opportunity for existing players because their software and networking know-how
is where the real value is in many cases. Some major industry players are
already moving in this direction by offering virtualized versions of their
products. The challenge for network operators is how to migrate their
operations and skill base to a software based networking environment while
carefully re-targeting investment to maximize re- use of existing systems and
processes.
Enablers for Network Functions Virtualization
Several recent technology developments make the goals of
Network Functions Virtualization achievable. This section describes these
enablers and briefly discusses relevance.
Cloud
Computing
Network Functions Virtualization will leverage modern
technologies such as those developed for cloud computing. At the core of these
cloud technologies are virtualization mechanisms: hardware virtualization by
means of hypervisors, as well as the usage of virtual Ethernet switches (e.g.
vswitch) for connecting traffic between virtual machines and physical
interfaces. For communication-oriented functions, high-performance packet
processing is available through high-speed multi-core CPUs with high I/O
bandwidth, the use of smart Ethernet NICs for load sharing and TCP Offloading,
and routing packets directly to Virtual Machine memory, and poll-mode Ethernet
drivers (rather than interrupt driven, for example Linux NAPI and Intel’s
DPDK).
Cloud infrastructures provide methods to enhance resource
availability and usage by means of orchestration and management mechanisms,
applicable to the automatic instantiation of virtual appliances in the network,
to the management of resources by assigning virtual appliances to the correct
CPU core, memory and interfaces, to the re-initialization of failed VMs, to
snapshot VM states and the migration of VMs.
Finally, the availability of open APIs for management and
data plane control, like OpenFlow, OpenStack, OpenNaaS or OGF’s NSI, provide an
additional degree of integration of Network Functions Virtualization and cloud
infrastructure.
Industry
Standard High Volume Servers
The use of industry standard high volume servers is a key
element in the economic case for Network Functions Virtualization. Network
Functions Virtualization leverages the economies of scale of the IT industry.
An industry standard high volume server is a server built using standardized IT
components (for example x86 architecture) and sold in the millions. A common feature of industry standard high
volume servers is that there is competitive supply of the sub components which are interchangeable inside the server.
We believe that Network Appliances which depend on the
development of bespoke Application Specific Integrated Circuits (ASICs) will
become increasingly uncompetitive against general purpose processors as the
cost of developing ASICs increases exponentially with decreasing feature
size.[5] Merchant silicon will still be applicable for commodity functions
implemented at scale, and ASICs will still be applicable for some types of very
high throughput applications.
Challenges for Network Functions Virtualization
There are a number of challenges to implement Network
Functions Virtualization which need to be addressed by the community interested
in accelerating progress. How this effort could be progressed is described
later in this document. Challenges we have identified are (not in any
particular order):-
Ø Portability/Interoperability. The ability to load and execute virtual appliances in
different but standardized datacenter environments, provided by different
vendors for different operators. The challenge is to define a unified interface
which clearly decouples the software instances from the underlying hardware, as
represented by virtual machines and their hypervisors. Portability and
Interoperability is very important as it creates different ecosystems for
virtual appliance vendors and datacenter vendors, while both ecosystems are
clearly coupled and depend on each other. Portability also allows the operator
the freedom to optimize the location and required resources of the virtual
appliances without constraints.
Ø Performance Trade-Off. Since the Network Functions Virtualization approach is
based on industry standard hardware (i.e. avoiding any proprietary hardware
such as acceleration engines) a probable decrease in performance has to be
taken into account. The challenge is how to keep the performance degradation as
small as possible by using appropriate hypervisors and modern software
technologies, so that the effects on latency, throughput and processing
overhead are minimized. The available performance of the underlying platform
needs to be clearly indicated, so that virtual appliances know what they can
get from the hardware. The authors of the white paper believe that using the
right technology choice will allow virtualization not only of network control
functions but also data/user plane functions.
Ø Migration and co-existence of
legacy & compatibility with existing platforms. Implementations of Network Functions Virtualization must
co-exist with network operators’ legacy network equipment and be compatible
with their existing Element Management Systems, Network Management Systems, OSS
and BSS, and potentially existing IT orchestration systems if Network Functions
Virtualization orchestration and IT orchestration are to converge. The Network
Functions Virtualization architecture must support a migration path from
today’s proprietary physical network appliance based solutions to more open
standards based virtual network appliance solutions. In other words, Network
Functions Virtualization must work in a hybrid network composed of classical
physical network appliances and virtual network appliances. Virtual appliances
must therefore use existing North Bound Interfaces (for management &
control) and interwork with physical appliances implementing the same
functions.
Ø Management and Orchestration. A consistent management and orchestration architecture is
required. Network Functions Virtualization presents an opportunity, through the
flexibility afforded by software network appliances operating in an open and standardized
infrastructure, to rapidly align management and orchestration North Bound
Interfaces to well defined standards and abstract specifications. This will
greatly reduce the cost and time to integrate new virtual appliances into a
network operator’s operating environment. Software Defined Networking (SDN)
further extends this to streamlining the integration of packet and optical
switches into the system e.g. a virtual appliance or Network Functions Virtualization
orchestration system may control the forwarding behaviors of physical switches
using SDN.
Ø Automation. Network Functions Virtualization will only scale if all
of the functions can be automated. Automation of process is paramount to
success.
Ø Security & Resilience. Network operators need to be assured that the security,
resilience and availability of their networks are not impaired when virtualized
network functions are introduced. Our initial expectations are that Network
Functions Virtualization improves network resilience and availability by
allowing network functions to be recreated
on demand after a failure. A virtual appliance should be as secure as a
physical appliance if the infrastructure, especially the hypervisor and its
configuration, is secure. Network operators will be seeking tools to control
and verify hypervisor configurations.
They will also require security certified hypervisors and virtual
appliances.
Ø Network Stability. Ensuring stability of the network is not impacted when managing
and orchestrating a large number of virtual appliances between different
hardware vendors and hypervisors. This is particularly important when, for
example, virtual functions are relocated, or during re-configuration events (e.g.
due to hardware and software failures) or due to cyber-attack. This challenge is not unique to Network
Functions Virtualization. Potential instability might also occur in
current networks, depending on
unwanted combinations of diverse control
and optimization mechanisms, for example acting on either the underlying
transport network or
on the higher
layers’ components (e.g.
flow admission control,
congestion control, dynamic
routing and allocations,
etc.). It should
be noted that occurrence of network
instability may have primary
effects, such as jeopardizing,
even dramatically, performance
parameters or compromising
an optimized use of resources. Mechanisms capable of ensuring network
stability will add further benefits to Network Functions Virtualization.
Ø Simplicity. Ensuring that virtualized network platforms will be
simpler to operate than those that exist today. A significant and topical focus
for network operators is simplification of the plethora of complex network
platforms and support systems which have evolved over decades of network
technology evolution, while maintaining continuity to support important revenue
generating services. It is important to avoid trading one set of operational headaches
for a different but equally intractable set of operational headaches.
Ø Integration. Seamless integration of multiple virtual appliances onto
existing industry standard high volume servers and hypervisors is a key
challenge for Network Functions Virtualization. Network operators need to be
able to “mix & match” servers from different vendors, hypervisors from
different vendors and virtual appliances from different vendors without
incurring significant integration costs and avoiding lock-in. The ecosystem
must offer integration services and maintenance and third-party support; it
must be possible to resolve integration issues between several parties. The ecosystem will require mechanisms to validate
new Network Functions Virtualization products. Tools must be identified and/or
created to address these issues.
Recommendations/Call for Action
Network Functions Virtualization is already occurring. In
a few years, we can expect the communications industry to look and feel similar
to the IT industry. There will be a wider range of business models more suited
to a software industry. Operations complexity will be abstracted away by more
automation and self-provisioning will be more common. As detailed in this white
paper, Network Functions Virtualization will deliver many benefits for network
operators and their partners and customers whilst offering the opportunity to
create new types of eco-systems (alongside traditional supply models based on
preferred strategic partners) which will encourage and support rapid innovation
with reduced cost and reduced risk. To reap these benefits the technical
challenges, as described above, must be addressed by the industry.
The authors of this white paper believe that solutions to
these technical challenges are available (or could be made available) and
recommend that the IT and Network industries combine their complementary expertise
and resources in a joint collaborative effort to reach broad agreement on standardized approaches and common
architectures which address these technical challenges, and which are
interoperable and have economies of scale.
To accelerate progress, a new network operator-led Industry
Specification Group (ISG) with open membership is being setup under the
auspices of ETSI to work through the technical challenges for Network Functions
Virtualization as outlined in this white paper.