XXXII Simpósio Brasileiro de

Redes de Computadores e Sistemas Distribuídos

Florianópolis, 5 a 9 de Maio de 2014

Evoluções da Tecnologia NFV, sua

Sinergia com SDN e Impactos e
Oportunidades na Rede FIBRE
Cesar Marcondes (UFSCar)
 Agenda
• Motivation;
– Problem Statement, Trends in IT & Telecom challenges
• Network Functions Virtualization
– Vision; Approach; Benefits & Promises
– The ETSI NFV ISG; WG; Architecture
• NFV Requirements and Challenges
• Use Cases, Proof-of-Concepts
• Enabling Technologies
• DEMO: Vyatta
• Discussion on FIBRE Future
Motivation
Problem Statement
• Complex carrier networks
– with a large variety of proprietary nodes and hardware appliances.
• Launching new services is difficult and takes too long
– Space and power to accommodate
– requires just another variety of box, which needs to be integrated.
• Operation is expensive
– Rapidly reach end of life Traditional Network model
– due to existing procure-design,-
integrate-deploy cycle.

 Network functionalities are based on specific HW&SW

 One physical node per role
 Sisyphus on Different Hills
Telco Cycle Service Providers Cycle
Idea !! Idea !!
AVAILABLE AVAILABLE

Telco Operators Deploy Develop Deploy Publish

Demand Service Providers

2-6 Months

Equipment Sell
Vendors
Drive

Standardise
Implement
SDOs Critical mass of
supporters
2-6 Years

2-6 years 2-6 months

Source: Adapted from D. Lopez Telefonica I+D, NFV
 Enter the Software-Defined Era
Traditional telcos Internet players

• Very intensive •

•
in hardware
Software not at
x •
Very intensive
in software
Hardware is a
the core necessary base

HARDWARE+ SOFTWARE
+
-

AT&T, Telefonica, Google, Facebook

Telebras

Adapt to survive: Telco evolution focus shifting from hardware to software

Source: Adapted from D. Lopez Telefonica I+D, NFV
Trends Challenges
• High performance industry • Huge capital investment to deal with
standard servers shipped in very current trends
high volume
• Network operators face an increasing
• Convergence of computing,
storage and networks disparity between costs and revenues
• New virtualization technologies • Complexity: large and increasing
that abstract underlying hardware variety of proprietary hardware
yielding elasticity, scalability and appliances in operator’s network
automation • Reduced hardware lifecycles
• Software-defined networking • Lack of flexibility and agility: cannot
• Cloud services move network resources where &
• Mobility, explosion of devices and when needed
traffic
• Launching new services is difficult and
takes too long. Often requires yet
another proprietary box which needs to
be integrated

Source: Adapted from D. Lopez Telefonica I+D, NFV

 The NFV Concept
A means to make the network more flexible and simple by
minimising dependence on HW constraints

Traditional Network Model: Virtualised Network Model:

APPLIANCE APPROACH v v
VIRTUAL APPLIANCE APPROACH

DPI CG-NAT GGSN/ VIRTUAL

BRAS SGSN APPLIANCES
DPI
Firewall PE Router
BRAS
GGSN/SGSN
ORCHESTRATION, AUTOMATION
& REMOTE INSTALL
PE Router STANDARD
Session Border
Firewall CG-NAT Controller HIGH VOLUME
SERVERS
 Network Functions are based on specific HW&SW  Network Functions are SW-based over well-known HW
 One physical node per role  Multiple roles over same HW

Source: Adapted from D. Lopez Telefonica I+D, NFV

 Target
Independent Software Vendors
Classical Network Appliance Approach

Message CDN Session Border WAN

Acceleration Orchestrated,
Router Controller
automatic &
remote install.

DPI Firewall Carrier Tester/QoE

Grade NAT monitor Standard High Volume Servers

Standard High Volume Storage

SGSN/GGSN PE Router BRAS Radio Access

Network Nodes
Standard High Volume
• Fragmented non-commodity hardware. Ethernet Switches
• Physical install per appliance per site.
• Hardware development large barrier to entry for new
vendors, constraining innovation & competition. Network Virtualisation Approach
Source: NFV
 Network Functions Virtualization
• Network Functions Virtualization is about implementing network
functions in software - that today run on proprietary hardware -
leveraging (high volume) standard servers and IT virtualization
• Supports multi-versioning and multi-tenancy of network functions, which
allows use of a single physical platform for different applications, users
and tenants
• Enables new ways to implement resilience, service assurance, test and
diagnostics and security surveillance
• Provides opportunities for pure software players
• Facilitates innovation towards new network functions and services that
are only practical in a pure software network environment
• Applicable to any data plane packet processing and control plane
functions, in fixed or mobile networks
• NFV will only scale if management and configuration of functions can be
automated
• NFV aims to ultimately transform the way network operators architect and
operate their networks, but change can be incremental

Source: Adapted from D. Lopez Telefonica I+D, NFV

 Benefits & Promises of NFV
• Reduced equipment costs (CAPEX)
– through consolidating equipment and economies of scale of IT industry.
• Increased speed of time to market
– by minimising the typical network operator cycle of innovation.
• Availability of network appliance multi-version and multi-tenancy,
– allows a single platform for different applications, users and tenants.
• Enables a variety of eco-systems and encourages openness.
• Encouraging innovation to bring new services and generate new
revenue streams.

Source: NFV
 Benefits & Promises of NFV
• Flexibility to easily, rapidly, dynamically provision and
instantiate new services in various locations
• Improved operational efficiency
• by taking advantage of the higher uniformity of the physical network
platform and its homogeneity to other support platforms.
• Software-oriented innovation to rapidly prototype and test
new services and generate new revenue streams
• More service differentiation & customization
• Reduced (OPEX) operational costs: reduced power, reduced
space, improved network monitoring
• IT-oriented skillset and talent

Source: Adapted from D. Lopez Telefonica I+D, NFV

 So, why we need/want NFV(/SDN)?
1. Virtualization: Use network resource without worrying about where it is
physically located, how much it is, how it is organized, etc.
2. Orchestration: Manage thousands of devices
3. Programmable: Should be able to change behavior on the fly.
4. Dynamic Scaling: Should be able to change size, quantity
5. Automation
6. Visibility: Monitor resources, connectivity
7. Performance: Optimize network device utilization
8. Multi-tenancy
9. Service Integration
10. Openness: Full choice of modular plug-ins
Note: These are exactly the same reasons why we need/want SDN.

Source: Adapted from Raj Jain

 NFV and SDN
• NFV and SDN are highly complementary
• Both topics are mutually beneficial but not dependent on each other

Creates competitive Software

Creates network
supply of innovative Open abstractions to
Innovation Defined
applications by third Networking enable faster
parties innovation

Network
Functions
Virtualization Reduces CAPEX, OPEX,
Space & Power
Consumption

Source: NFV
 NFV vs SDN
• NFV: re-definition of network equipment architecture
• NFV was born to meet Service Provider (SP) needs:
– Lower CAPEX by reducing/eliminating proprietary hardware
– Consolidate multiple network functions onto industry standard
platforms
• SDN: re-definition of network architecture
• SDN comes from the IT world:
– Separate the data and control layers,
while centralizing the control
– Deliver the ability to program network behavior using well-
defined interfaces
 Software Defined Networking

SDN

Network equipment as Open interfaces (OpenFlow) for

Black boxes instructing the boxes what to do

FEATURE FEATURE
OPERATING SYSTEM

SPECIALIZED PACKET
FEATURE FEATURE FORWARDING HARDWARE FEATURE FEATURE

OPERATING SYSTEM OPERATING SYSTEM

SPECIALIZED PACKET SPECIALIZED PACKET

FORWARDING HARDWARE FEATURE FEATURE FORWARDING HARDWARE

OPERATING SYSTEM

SPECIALIZED PACKET
FORWARDING HARDWARE
SDN
Boxes with autonomous
behaviour Decisions are taken out of the box

SDN
FEATURE FEATURE

OPERATING SYSTEM

SPECIALIZED PACKET
FEATURE FEATURE FORWARDING HARDWARE FEATURE FEATURE
OPERATING SYSTEM OPERATING SYSTEM

SPECIALIZED PACKET SPECIALIZED PACKET

FORWARDING HARDWARE FEATURE FEATURE FORWARDING HARDWARE

OPERATING SYSTEM

SPECIALIZED PACKET

Simpler OSS to manage the SDN

FORWARDING HARDWARE

Adapting OSS to manage black boxes controller

Source: Adapted from D. Lopez Telefonica I+D, NFV
Scope of NFV and OpenFlow/SDN

Source: NEC
Networking with SDN & NFV

Source: NEC
(Network Virtualization)2 = SDN + NFV

SDN: Software Defined

Networking

NFV: Network Functions

Virtualisation

Source: Adapted from D. Lopez Telefonica I+D, NFV

 Some Use Case Examples
…not in any particular order

• Switching elements: BNG, CG-NAT, routers.

• Mobile network nodes: HLR/HSS, MME, SGSN, GGSN/PDN-GW.
• Home networks: Functions contained in home routers and set top boxes to
create virtualised home environments.
• Tunnelling gateway elements: IPSec/SSL VPN gateways.
• Traffic analysis: DPI, QoE measurement.
• Service Assurance: SLA monitoring, Test and Diagnostics.
• NGN signalling: SBCs, IMS.
• Converged and network-wide functions: AAA servers, policy control and
charging platforms.
• Application-level optimisation: CDNs, Cache Servers, Load Balancers,
Application Accelerators.
• Security functions: Firewalls, virus scanners, intrusion detection systems,
spam protection.

Source: NFV
 The ETSI NFV ISG
• Global operators-led Industry • Currently, four WGs and two EGs
Specification Group (ISG) under the – Infrastructure
auspices of ETSI – Software Architecture
– ~150 member organisations – Management & Orchestration
• Open membership – Reliability & Availability
– ETSI members sign the “Member – Performance & Portability
Agreement” – Security
– Non-ETSI members sign the
“Participant Agreement”
– Opening up to academia
• Operates by consensus
– Formal voting only when required
• Deliverables: White papers
addressing challenges and operator
requirements, as input to SDOs
– Not a standardisation body by itself
Source: Adapted from D. Lopez Telefonica I+D, NFV
High-level Architecture
 NFV Layers
E2E Network Service
End End
Point Network Service Point
Logical Abstractions
VNF VNF

VNF VNF VNF

Logical Links

VNF Instances
SW Instances
VNF VNF VNF VNF
VNF : Virtualized Network Function

NFV Infrastructure

Virtual Resources Virtual Virtual Virtual

Compute Storage Network
Virtualization SW Virtualization Layer
HW Resources Compute Storage Network

Source: Adapted from D. Lopez Telefonica I+D, NFV

Rethinking relayering
Requirements and Challenges
NFV
 First: A Few Challenges
• Achieving high performance virtualised network
appliances
– portable between different HW vendors, and with different
hypervisors.
• Co-existence with bespoke HW based network platforms
– enabling efficient migration paths to fully virtualised network
platforms.
• Management and orchestration of virtual network appliances
– ensuring security from attack and misconfiguration.
• NFV will only scale if all of the functions can be automated.
• Appropriate level of resilience to HW and SW failures.
• Integrating multiple virtual appliances from different vendors.
– Network operators need to be able to “mix & match” HW,
– hypervisors and virtual appliances from different vendors,
– without incurring significant integration costs and avoiding
lock-in.
• NFV and SDN But... Based on what?
Use Cases
Then... More challenges!
NFV Performance Challenges

Source: Ivan Pepelnjak

Use Cases
NFV
 Use Cases Matrix

Use Case Matrix – 4 big horizontal themes, and 9 use cases

ETSI NFV POC
 NFV Infrastructure as a Service
(NFVIaaS)
NFV Infrastructure :
• provide the capability or
functionality of providing an
environment in which Virtualized
network functions (VNF) can
execute

• NFVIaaS provides compute

capabilities comparable to an IaaS
cloud computing service as a run
time execution environment as
well as support the dynamic
network connectivity services
that may be considered as
comparable to NaaS
VNF Forwarding
Graph
VNF FG Logical View

VNF FG Physical View

 Mobile Core Network and IMS
• Mobile networks are populated with a large
variety of proprietary hardware appliances

• Flexible allocation of Network Functions on such

hardware resource pool could highly improve
network usage efficiency

• Accommodate increased demand for particular

services (e.g. voice) without fully relying on the
call restriction control mechanisms in a large-
scale natural disaster scenario such as the Great
East Japan Earthquake
V-EPC

• Examples of Network
Functions include MME,
S/P-GW, etc
• This use case aims at
applying virtualization to
the EPC, the IMS, and these
other Network Functions
mentioned above
 Virtualization of Mobile Base Station
• Mobile network traffic is significantly increasing by the
demand generated by application of mobile devices, while
the ARPU (revenue) is difficult to increase

• LTE is also considered as radio access part of EPS (Evolved

Packet System) which is required to fullfil the requirements
of high spectral efficiency, high peak data rates, short
round trip time and frequency flexibility in radio access
network (RAN)

• Virtualisation of mobile base station leverages IT

virtualisation technology to realize at least a part of RAN
nodes onto standard IT servers, storages and switches
Virtualization of Mobile Base Station

Functional blocks in C-RAN

LTE RAN architecture evolution by centralized BBU pool

(Telecom Baseband Unit)
Proof-of-Concepts
NFV
 Ongoing Proof of Concepts
• CloudNFV Open NFV Framework Project • C-RAN virtualisation with dedicated
– Telefonica - Sprint - 6WIND - Dell - EnterpriseWeb – hardware accelerator
Mellanox - Metaswitch - Overture Networks - Qosmos - – China Mobile - Alcatel-Lucent - Wind River
Huawei - Shenick Systems - Intel
• Service Chaining for NW Function Selection in Carrier • Automated Network Orchestration
Networks – Deutsche Telekom - Ericsson - x-ion GmbH -
– NTT - Cisco - HP - Juniper Networks Deutsche Telekom Innovation Laboratories
• Virtual Function State Migration and Interoperability • VNF Router Performance with DDoS
– AT&T - BT - Broadcom Corporation - Tieto Corporation Functionality
• Multi-vendor Distributed NFV – AT&T - Telefonica - Brocade - Intel - Spirent
– CenturyLink - Certes - Cyan - Fortinet - RAD • NFV Ecosystem
• E2E vEPC Orchestration in a multi-vendor open NFVI – Telecom Italia - DigitalWave - SunTec - Svarog
environment Technology Group - Telchemy - EANTC
– Telefonica - Sprint - Intel - Cyan - Red Hat - Dell - • Multi-Vendor on-boarding of vIMS on a
Connectem cloud management framework
• Virtualised Mobile Network with Integrated DPI – Deutsche Telekom - Huawei Technologies -
Alcatel-Lucent
– Telefonica - Intel - Tieto - Qosmos - Wind River Systems -
Hewlett Packard • Demonstration of multi-location, scalable,
stateful Virtual Network Function
– NTT - Fujitsu - Alcatel-Lucent
 CloudNFV
Dell Lab infrastructure for CloudNFV

Source: ETSI Ongoing PoC

http://nfvwiki.etsi.org/index.php?title=On-going_PoCs
 Service Chaining for NW Function
Selection in Carrier Networks

vDPI: CSR 1000v (Cisco Systems)

vCPE: VSR1000 (Hewlett-Packard)
vFW: FireFly (Juniper Networks)
VIM (NW Controller): Service Chaining
Function (prototype) + Ryu (NTT) Source: ETSI Ongoing PoC
 Multi Vendor on-boarding of vIMS on
Cloud Management Frame
Scenario 1 – One-click service deployment.
IMS service is provided by several 3GPP Network
Functions, such as CSC, HSS, MMTel, etc. These functions, all
from Huawei, are virtualized. With the pre-defined
templates and scripts, all functions can be deployed
automatically, onto the cloud platform provided by DT and
ALU.
Scenario 2 – Auto-scaling of VNF
Traffic load generator by a simulator increases and pushes
up the workload of the VNF. When the workload exceeds
the pre-defined threshold, additional resources (VM) are
automatically allocated. In situations of reducing VNF
capacity due to decreasing traffic load, similar in reverse
direction
Scenario 3 – Automated healing of VNF
When a VM containing a component of a VNF (VNFC) fails, a
new VM will be automatically allocated and created with
appropriate component instantiated on it. This process heals
the VNF with no service interruption.

CloudBand is the Alcatel-

Source: ETSI Ongoing PoC Lucent Cloud Platform
ENABLING TECHNOLOGIES
 Remarkable Enabling Technologies
• Minimalistic OS
– ClickOS
• Improving Linux i/O
– Netmap, VALE, Linux NAPI
• Programmable virtual switches / bridges
– Open vSwitch
• Exploiting x86 for packet processing
– Intel DPDK
• Some example start-ups
– LineRate Systems, 6WIND, Midonet, Vyatta (bought by BCD)

Image source: NEC

 ClickOS Archtecture
Martins, J. et al. Enabling Fast, Dynamic
Network Processing with ClickOS. HotSDN
2013.
 Intel DPDK

Fonte: Network
Intel Data Plane Development Kit (Intel DPDK) Overview – Packet Processing on Intel Architecture
Function
Virtualisation - NFV
 Intel DPDK
 Buffer and Memory Manager
– Manage the allocation of objects non-NUMA using
hugepages through rings, reducing TLB access,
also, perform a pre-allocation of fixed buffer space
for each core
 Queue Manager
– Implements lockless queues, allow packets to be
processed by different software components with no
contention
 Flow Classification
– Implements hash functions from information tuples,
allow packets to be positioned rapidly in their flow
paths. Improves throughput
 Pool Mode Driver
– Temporary hold times thus avoiding raise NIC
interruptions
Network Function
Virtualisation - NFV
 Vyatta vRouter (5400 e 5600)

 Vrouter 5600
Licensing bare metal, VM and Amazon
Features:
–Network Conectivity
–Firewall
–IPv6
–CLI, GUI and Brocade Vyatta Remote Access API
–Authentication (TACACS+, RADIUS)
–Monitoring and log
–IPSec VPN
–QoS
–High-Availability
–vPlane

Network Function
Virtualisation - NFV
 Openstack

OpenStack is a global collaboration of developers and cloud computing

technologists producing the ubiquitous open source cloud computing platform for
public and private clouds. The project aims to deliver solutions for all types of
clouds by being simple to implement, massively scalable, and feature rich. The
technology consists of a series of interrelated projects delivering various
components for a cloud infrastructure solution.

Source: Openstack.org
Network Function
Virtualisation - NFV
Network Function
Virtualisation - NFV
http://www.fibre-ict.eu/
IMPACT AND OPPORTUNITIES
(FIBRE)
 Resources
OCF – OpenFlow focused + Xen
OMF – wireless focused
 NFV PoC (vRouter) on FIBRE
• We have developed a demo • How to do the same on FIBRE?
for the SBRC 2014 Tutorial 1 – • Vyatta vRouter 5600
“Network Function • DPDK – how to do it? Without
Virtualization: Perspectivas, impacting shareness?
Realidades e Desafios”
• Vyatta vRouter 5600 • OpenFlow Stitching – change
Linux Bridges to OVS?
• KVM/RedHat • Generating Traffic using
• DPDK (hardware and 4-1G NIC netFPGA (ongoing work
support it) UFSCar)
• OVS • Orchestration? How to do
• Iperf that? (OpenStack on top of
• OpenStack OCF?) - OpenStack versus OCF
 NFV PoC (vIMS) on FIBRE
• Using OMF resources to
operate as advanced
Signal Processing
antennas
• Virtual machines to
support the IMS
components (Xen +
OpenSource IMS)
• Stitching using OpenFlow
• Orchestration? How to do
that? (OpenStack on top
of OCF?)
 Architectural Changes and Evolution
• OCF & OMF vs SFA vs • Evolution
OpenStack – More hardware
• OpenStack with • 1 server is not enough!
OpenFlow (challenge) • At least 3 IBM-similar
servers
• Orchestration Module – • upgrade memory for more
Heat (OpenStack) VM
• OF1.0 vs OF1.3+ – Include OpenStack
• As another direct CMF
– Virtualization, OpenStack
and OpenFlow
Technologies Courses
(during all RNP events)
– More Use as it becomes
open to the community