Ethernet Demarcation Devices for managing end to end Ethernet service delivery

Carrier Ethernet Services

White Paper

Ethernet Demarcation Devices
for managing
End-to-End Ethernet Service Delivery

July, 2012

Metrodata Ltd Fortune House, TW20 8RY U.K.

Tel: +44 (0)1784 744700 www.metrodata.co.uk E-mail: sales@metrodata.co.uk


Introduction
The use of Ethernet by Telecommunications Service Providers, for Corporate connectivity services, is
growing rapidly and is set to dominate within just a few years.

'Layer 2' (i.e. of the OSI connection hierarchy model) Ethernet offers some attractive advantages to both
Infrastructure Carriers and Telecoms Service Providers in terms of service provisioning, the inherent set-
up complexities of Managed Router networks being largely eliminated. Nevertheless, challenges still exist
in both the provisioning and ongoing management of Ethernet services, particularly since not only the
quantity, but also the sophistication of services demanded by customers is rising. For example, until very
recently even in the more advanced economies, the norm for Voice and Data services has been for two
separate infrastructures. Now, co-incident with the rise of Ethernet as a transport vehicle, the emergence
of SIP trunking for Voice over IP (VoIP) communications, and its associated need for tighter constraints in
service quality than is required for most Data (e.g. simple Internet access) applications, is creating a need
for greater management visibility to Ethernet service delivery on the part of the Service Provider.

Much effort, by both Carriers and Telecoms Equipment Manufacturers, principally under the auspices of
the 'Metro Ethernet Forum' (MEF), has been put into the development of management tools for both
Connectivity and, latterly, Performance Assurance for Ethernet services. Nevertheless, a reality today is
that relatively few Infrastructure Carriers (by which we mean those with either Long-haul or local transport
infrastructure, i.e. cable and/or fibre) make use of the capabilities now potentially available to them for
management control and performance validation. Moreover, those which do generally choose to limit the
extent of their 'manageable domain' to network segments bounded by their own Ethernet Demarcation
Devices (EDDs). As National Telecommunications markets become increasingly de-regulated, there has
been a corresponding growth in the number and diversity of Telecoms (in this case Ethernet) Service
Providers, offering both a range of Value-Added Services (i.e. Data and/or Voice 'cloud' services) and
Connectivity, some or all of which they typically secure under wholesale agreements from the
Infrastructure Carriers.

Network
Operations Value-Added
Services
Centre

Service Provider
Core Network

Infrastructure Carrier
Infrastructure Carrier
Ethernet Circuit
Ethernet Circuit
(Typ. via wholesale)
Customer (Typ. via wholesale) Customer
Site A Site B

Customer's
L2 Network

Fig. 1: Typical Ethernet Service Provision model, within which Service Providers may exist in
separation from Infrastructure Carriers, from whom they acquire wholesale links

Managing End-to-End Ethernet Service Delivery
Page 1 of 9

It is worth making a distinction between Ethernet Service Providers and Infrastructure Carriers, albeit that
particularly in the case of the large National Providers, they may well be the same entity. The point here is
that the Service Provider more often than not acts in the role of 'Aggregator', piecing together an end-to-
end connectivity solution for their customers from elements of their own Core network, on which their
value-adding services are based, and long or short-haul (i.e. 'tail') circuits from other Infrastructure
Carriers.

This paper examines the challenges of Management of end-to-end network connections experienced by
Ethernet Service Providers and looks at how the use of cost-effective, manageable Ethernet Demarcation
Devices can help them to meet those challenges.

Backgrounder
It will not be a surprise to hear that much has changed over the past quarter century in the range and
nature of the Telecommunications Services offered by Carriers to Corporate customers. In reality though,
much of the change in Wide Area Networking (WAN) services has occurred in a rather compressed
timescale relative to the evolution of the Local Area Networking (LAN) services used within customer
premises, at least in terms of speed if not so much of the underlying technology.

With the advent of Ethernet in the mid 1980s, Corporates have (with a few distractions along the way)
kept their faith with Ethernet in the LAN through a continual evolution process, which has seen Ethernet
progress rapidly from a 'shared' 10Mbps bandwidth environment to the relatively common 'per-port'
Switched 1Gbps and 10Gbps Ethernet services of today.

In contrast, Wide-Area Network (WAN) bandwidths and technologies remained relatively static for long
periods over the past 25 years. Even as the relatively much faster Ethernet Corporate LAN environment
progressed, WAN services, barriered by Routers, evolved independently and at a slower pace.

Over the past ten years, and in particular the past five, this picture has started to change dramatically.
During the ten-year timeframe, there has been an inexorable evolution from SDH-based to so-called 'Next
Generation' packet-based core switching in Carrier networks. Whilst SDH is still very much at the heart of
many Carrier networks most, if not all, no longer invest in any development of their SDH infrastructure
and look to their emerging Packet-based networks for development. In general, Next Generation packet
networks have used IP/MPLS switching at the core and the choice of many Service Providers has been to
offer 'Managed IP' network services across these platforms.

Over the past five years, the ubiquity of Ethernet in the LAN, driven not only by its connection speed but
also by its relative simplicity, has led the Telecoms Carrier community to start to offer WAN services
based on Layer-2 Ethernet connection, enabling Corporates to use largely the same familiar technologies
in both LAN and WAN. Whilst 'Ethernet over SDH' services have emerged, extending the return of
investment by the traditional infrastructure Carriers in their SDH core switched networks, for the majority
of newer entrants, leveraging their MPLS switched core platforms to offer layer-2 point-to-point VPN and
VPLS multi-point connectivity services has become an increasingly significant part of their service
portfolio.

Coming right up to date, the diverse manner in which Ethernet WAN services are delivered today is
something of a minefield of complexity, with several core network architectures being used, often
championed by competing Network Equipment Manufacturers. A debate about the different WAN
technologies underpinning the transport of Ethernet services today might invoke a deeply technical foray
into the relative merits of;

 Ethernet over SDH / SONET, typically today via GFP encapsulation with VCAT

 Ethernet over MPLS, via Point-to-Point L2 VPN or VPLS Pseudowire tunnelling

 Ethernet over Carrier-Ethernet Transport (CET)

Page 2 of 9

The last of these, and the most recent, might itself lead to a sub-debate over the merits of alternative
approaches such as 'MPLS-TP' and 'PBB-TE' which is beyond the focus of this paper.

If this may sound a little bewildering, there is a similarly varied approach to the so-called 'last-mile'
delivery of Ethernet services, i.e. between 'local' Carrier exchanges and customer premises.

In this area, amongst the transport mechanisms in regular deployment are;

 Ethernet over both Passive (i.e. PON) and Active Fibre

 Ethernet over TDM services, including single or multiple 'bonded' SDH/PDH circuits

 Ethernet over multiple bonded 'telephony' grade copper wiring

 Ethernet over cable-modem (Co-ax) infrastructure

Increasingly, these Ethernet services are being used not only for Data, but also for Voice, i.e. for the
delivery of VoIP 'SIP' Trunks directly to the modern generation of Corporate IP-PBX Servers.

One of the main challenges to Ethernet as a delivery vehicle for Corporate Voice and Data services is
that, unlike its SDH predecessor, Ethernet is not intrinsically deterministic, nor does it carry, at least by
default, clock synchronisation signalling for Voice and other services, so it is not sufficient simply to
provision an end-to-end Ethernet network connection safe in the knowledge that any underlying
architecture will assure the performance of the link for a given customer application. Verification of the
integrity of the circuit is set to become an important aspect of Carrier Ethernet implementation.

Ethernet Demarcation Devices, otherwise known as Network interface Devices (NIDs) are being deployed
by Carriers as a manageable interface between the edge of their network and their customers' switched
Ethernet LANs.

NOC
Management Management
Access Access
Customer Customer
Site A Site B

Carrier Ethernet
Network
Ethernet Circuit Ethernet Circuit
demarcation demarcation

Fig. 2: Deployment of Ethernet Demarcation Devices (EDDs) for end-point management

Increasingly, Corporate clients are asking of their Ethernet Service Provider not just that they provide a
certain bandwidth of connection, but that they demonstrate, both at the time of network provisioning and
on an ongoing basis during in-service use, that they are meeting specific 'Service Level Agreement' (SLA)
conditions. To add further complexity, the customer SLA may not relate to the singular 'pipe' of the
Ethernet connection, but may indeed relate to a number of distinct service 'flows' within the pipe, such as
VoIP, Video or Internet Data traffic, for which markedly different service parameters may apply.

Page 3 of 9


Management Challenges for the Ethernet Service Provider
For the Ethernet Service Provider, the management of customer end-points may be a challenge. As
National Telecommunications Services worldwide undergo deregulation, opening up to an ever-
increasing number of Service Providers, more and more often a Customer's Service Provider is not itself,
either wholly or partially, an Infrastructure Carrier, but acts as an 'Aggregator', deploying Infrastructure
Networks acquired on a wholesale basis for either for long-haul circuits, local 'last mile' access, or both.

Typically, although Infrastructure Carriers may have their own manageable circuit end-points, they do not
typically grant management access to these devices to wholesale customers. Unless the Service Provider
has their own manageable entities at the customer premises they may not be able, without sending an
Engineer to site, to be certain either that an effective connection is made to the customer network, or that
the Infrastructure Carrier's circuit is fully configured and operational. In short, they have little or no
customer premise or 'end-to-end' circuit visibility. This scenario is illustrated in fig. 3 below.

Value-Added
NOC Services

INTERNET
PSTN
(via SIP G/W)
Service Provider
Core Network

Typically
VPLS over MPLS

Infrastructure Carrier Infrastructure Carrier
Ethernet Circuit Ethernet Circuit
(Typ. via wholesale) (Typ. via wholesale)
Problem!
Lack of Management Customer
Customer Visibility here for the SP Site B
Site A

Customer demarcation
(Ethernet connection)

Customer's L2 Switches

Fig. 3: The Ethernet Service Provider may only have visibility to customer connections and the end-
to-end network path by deployment of their own managed Ethernet Demarcation Devices

By introducing their own manageable Ethernet Demarcation Devices, the Service Provider gains
management visibility to the end-points of the network, between Carrier 'tail' circuits and the Customer
LAN.

One requirement for the Service Provider is to keep their own management traffic separated from that of
their Customer(s). This is normally achieved via the use of a dedicated VLAN for Network Management,
within which typically resides the management tools and platforms associated with the Service Provider's
Operations Centre (NOC). It should be possible to terminate a given management VLAN within the
Demarcation Device at the customer premises, in such a manner that management traffic does not
propagate through the customer facing network port(s) of the device. The management VLAN itself may
be designated either via a dedicated 'carrier-class' Q-in-Q S-Tag or alternatively, for less convoluted
multi-Carrier networks, via a more conventional customer C-Tag designation, which requires the Service
Provider to ensure that the customer itself does not use the same VLAN designation for their own traffic
separation and/or prioritisation purposes.

Page 4 of 9


Management visibility Vs. Performance assurance
In order to provide a degree of Management visibility at the customer premise, only a limited capability is
required for Ethernet Demarcation Device functionality. The Service Provider requires to 'see' beyond any
demarcation device deployed by their 'tail circuit' infrastructure carrier, with the ability both to report on
connectivity both towards the network and similarly towards the customer's switched LAN environment. In
general, switch-based demarcation devices, supporting VLAN/prioritisation and software-based
management functionality, accessible via SNMP and/or a command-line interface via Telnet, can be
sufficient. This scenario is illustrated in Fig. 4 below:

Value-Added
NOC Services

Management Ethernet Management
Access Service Provider Access
Core Network

Customer Infrastructure Carrier Infrastructure Carrier Customer
Site A Ethernet Circuit 1 Ethernet Circuit 2 Site B

Carrier Circuit 1 Carrier Circuit 2

Service Provider Service Provider
demarcation Basic Ethernet Demarcation Devices provide End- demarcation
Point Manageability, typically via dedicated
Management 'Carrier VLAN' (Q-in-Q S-Tag), or via
protected 'Customer VLAN' (Reserved C-Tag)

Fig. 4: Basic end-point management via cost-effective Ethernet Demarcation Devices such as the
Metrodata FCM8000

In recent times, support for various degrees of Carrier Ethernet 'Operations, Administration and
Management' (OAM) protocols has been promoted, mostly via the auspices of the 'Metro Ethernet Forum'
(MEF), which is active in the promotion of Carrier Ethernet Services.

Relatively simple visibility and connectivity checking of single segment Ethernet connections is supported
by the 'Link OAM', or 'Ethernet First Mile' (EFM) protocol, formalised initially as IEEE 802.3ah, by which it
is still generally best known, albeit that this functionality has now been fully incorporated to the core of the
802.3 standard itself.

An additional level of connectivity assurance is offered by those Demarcation Devices adopting the
'Connectivity Fault Management' (CFM) protocol, sometimes referred to as a member of the 'Service
OAM' (S-OAM) suite, and formalised under IEEE 802.1ag. CFM offers the ability for a number of end-
point devices to establish and monitor a 'community' of reachable end-points corresponding to a
customer's network, which can offer some degree of pro-activity to the Service Provider with regard to
connectivity fault detection.

Above and beyond connectivity management though, customers are increasingly asking of their Service
Providers that they provision multiple traffic streams across their Ethernet 'pipe' connections, to which

Page 5 of 9

potentially different criteria may apply for key network performance parameter, including latency and
delay variation (a.k.a. 'jitter').

In a more advanced deployment, a Service Provider may need to provision multiple services per physical
Ethernet connection. They may then be faced with the challenge to demonstrate to their customer, at the
time of provisioning, that specific performance parameters are complied with for each individual Service
data stream within a given end-to-end Ethernet connection. Such parameters may be detailed within a
tightly defined 'Service Level Agreement' (SLA), to which compliance should be verified.

Furthermore, Service Providers may not only need to demonstrate SLA compliance at the time of
provisioning, but they may be required to subsequently monitor 'in service' traffic and take a pro-active
position with regard to any potential breach of SLA.

Ethernet Demarcation Devices equipped with more advanced packet processing capabilities can offer a
very effective tool to Service Providers in this regard. For example if a Service Provider, from a Network
Management console, can interact with an EDD in such a manner as to configure this device to issue one
or more test traffic streams across the network to a corresponding remote end-point, at which traffic may
be 'looped' and returned, this can be highly beneficial. Such test stream(s) can enable accurate reporting
of throughput, packet loss, latency and delay variation, for the end-to-end network link. Demarcation
Devices with such capabilities are now emerging. Necessarily, such devices contain more than simple
switch and management processing functionality. Dedicated packet processing hardware is required in
order to ensure accurate time-stamping, test collation and reporting in real-time for line rates up to 1Gbps
and beyond.

Another of the Service-OAM protocols, this time the ITU-T's Y.1731 suite, relates to the ability to provide
in-service testing and reporting of SLA compliance, which a number of vendors refer to as 'Performance
Assured Ethernet' (PAE). Fig. 5 below, indicates the use of advanced Ethernet Demarcation Devices,
such as the Metrodata FCM9004, for SLA verification:

Value-Added
NOC Services

Management Ethernet Management
Access Service Provider Access
Core Network

Customer Infrastructure Carrier Infrastructure Carrier Customer
Site A Ethernet Circuit 1 Ethernet Circuit 2 Site B

Carrier 1 Carrier 2
Circuit 1 SLA SLA Circuit 2

Service Provider Service Provider
demarcation Service Provider's
End-to-End SLA demarcation

Infrastructure Carriers may offer a clear SLA for their short or long-haul circuits, but this does
not provide full end-to-end SLA assurance. Advanced EDD equipment offers this functionality

Fig. 5: Performance Assured Ethernet with Customer SLA verification via Advanced Ethernet
Demarcation Devices, such as the Metrodata FCM9004.

Page 6 of 9


Metrodata Ltd Ethernet Demarcation Devices
For basic managed Demarcation applications, Metrodata offers the FCM8000 & FCM9000 products,
members of the popular family of 'MetroCONNECT' Ethernet Service Access and LAN Extension
solutions.

The FCM8000 and FCM9000 provide the following common features:

 10/100Mbps or 1Gbps Service Delivery:
 Copper or Fibre network port, Copper or Fibre user port
 VLAN tag-switching (802.1q) & prioritisation (802.1p) with Q-in-Q support
 IEEE 802.3ad with LACP support and RSTP for link resilience
 Per port traffic Policing/Shaping
 Jumbo Frame Support (10k byte)
 Management via Console port, Telnet, SNMP & Carrier E'net OAM (802.3ah, 802.1ag CFM)
 Service Provider's Management traffic isolation via designated VLAN
 Link loss forwarding
 Zero-touch provisioning in association with other MetroCONNECT family members
 TACACS+ & SSH for secure authentication & access

Fig. 6: Metrodata FCM8000 (front) and FCM9000 (rear) Managed Ethernet Demarcation Devices

For advanced applications, enabling the Service provider to provision and validate end-to-end Ethernet
Performance, Metrodata offers the FCM9004 product. This product offers Copper (RJ45) or Fibre (SFP)
Network Connection up to 1Gbps and up to 4x RJ45 & 1x SFP/Fibre LAN Connections up to 1Gbps.

Fig. 7: Metrodata FCM9004 Advanced Managed Ethernet Demarcation Device

In addition to the features of the more basic managed devices, the FCM9004 offers:

 Advanced VLAN Traffic Mapping & Q-in-Q handling
 ITU-T Y.1731 for in-service performance monitoring
 Core-Edge and End-End Network SLA verification capabilities, including:
 Embedded wirespeed test traffic generator with packet time-stamping
 Layer 2/3 SA/DA Loopback for assurance measurement over extended networks
 Embedded wirespeed SLA calculation & reporting tools

Page 7 of 9

 Dedicated hardware Service Assurance Module, 'MetroSAM', ensuring effective SLA
measurements for wire-speed applications to 1Gbps, including:
 Sophisticated provisioning and in-service measurement tools
 Throughput, Frame Loss, Frame Delay and Frame Delay Variation analysis
 Multiple traffic flow characterisation

A typical (simplified) scenario for deployment of the FCM9004, illustrating test-traffic generation and loop-
back during initial circuit provisioning, is represented in Fig. 8 below:

FCM9004 A Core Network FCM9004 B

Customer Site A Customer Site B
NMS

Fig. 8: FCM9004 SLA Verification during initial circuit provisioning

In this scenario, under control from the Service Provider's Network Management console, FCM9004-A is
instructed to instigate loop-back at corresponding end-point FCM9004-B and then to generate and launch
a test traffic stream (or streams) towards FCM9004-B. Using accurate time-stamping and wire-speed
computation, FCM9004-A can provide a detailed analysis of traffic compliance Vs. the customer's
required SLA with respect to throughput, packet loss, circuit latency and latency variation (jitter) per traffic
flow for the A-B connection, as illustrated in fig. 9 below. Thereafter, during in-service use, 'background'
test-packets are exchanged (using mechanisms of the Y.1731 S-OAM protocol) to continually monitor
SLA adherence, alerting the Management Console to any potential breach of required performance
levels.

SLA Verification Testing / Circuit Provisioning

A-end: "Site A" B-end: "Site B" Test Traffic Type: L2, MAC Addr Loop-back

Traffic Parameters Stream 1 Stream 2 Stream 3 Stream 4
Traffic Category: Real-Time High Priority Best-Effort Best-Effort
EVC Number (S-Tag): 100 100 100 100
VLAN Number (C-Tag): 101 102 103 104
Frame Size (Byte): 250 250 1000 10000

Committed SLA
CIR (Mbps): 100 10 0 0
CBS (Bytes):
EIR (Mbps):
Site B
Site A EBS (Bytes):
Core Colour Mode: Blind Blind Blind Blind
Frame loss (%): <0.05 <0.001 N/A N/A
Frame delay (µs): <300 <5000 N/A N/A
NMS
Frame delay variation (µs): <100 <2000 N/A N/A

Test Results
Throughput (Mbps): 100 10 10 80
Lost frames (%): 0.0250 0.0005 0.0100 0.0250
Typical Management Average frame delay (µs): 100 1000 5000 10000
Screen Presentation Frame delay variation (µs): 10 100 250 1000

Pass/Fail Pass Pass Pass Pass

Fig. 9: FCM9004 SLA verification report for end-to-end circuit Site A - Site B

Page 8 of 9


Summary
Infrastructure Carriers and Telecoms Service Providers alike see advantages in the provision of end-to-
end Layer-2 Ethernet connectivity as a solution to their clients' connectivity needs. Some of the
advantages of Ethernet, Vs. the current generation of largely 'Managed Router' services are:

 Lower provisioning complexity, no need for IP address schema

 Lower costs: Fewer Routers, lower cost CPE & lower complexity

 Lower network latency

 Protocol-agnostic network

 More rigorous QoS with precise end-to-end service levels in terms of throughput, loss, latency
and jitter

 Scalable bandwidth

 Evolution flexibility: Core network may transition from MPLS/VLPS to pure Ethernet Transport
(e.g. PBB-TE, MPLS-TP) with little or no impact on customer connection configuration

Whilst Infrastructure Carriers utilise their own manageable end-points for the link-services which they
provide, Service Providers generally require their own manageable Demarcation devices if they are to be
able to ensure even the most basic of management visibility, both towards their customers' LAN
connection point and towards their Wide-Area network infrastructure provider's termination point.

By using advanced, and increasingly cost-effective Demarcation Devices, such as the FCM9004 from
Metrodata, the Service Provider is potentially able to gain a high degree of Performance Assurance in
relation to their customers' end-to-end Ethernet networks. By deployment of Demarcation Devices based
on architectures incorporating custom packet-processing hardware for the generation of test-traffic, loop-
back and data collation, the Service Provider can verify, both at the time of circuit provisioning and
subsequently in-service, that their service meets stringent SLAs increasingly being demanded by their
customers. Such SLAs may relate not only to the Ethernet service 'pipe' between sites, but to multiple
traffic streams, of different types and potentially demanding different levels of performance in relation to
the key parameters of throughput, packet loss, latency and jitter, contained within the service pipe.

Page 9 of 9

Ethernet Demarcation Devices for managing end to end Ethernet service delivery

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Ethernet Demarcation Devices for managing end to end Ethernet service delivery (20)

Recently uploaded (20)

Ethernet Demarcation Devices for managing end to end Ethernet service delivery