Contents:
Introduction
The
new century requires new networking technologies and systems to
cope with the unprecedented surge in data traffic, much of this
traffic due to the increasing popularity of the Internet for
- commerce,
- information
dissemination, and
- communication.
Technologies developed and implemented
in the 20th century did not anticipate that the Internet Protocol
would be the predominant protocol type and Internet traffic the
predominant traffic type as we enter the new century.
The first digital telecommunications
technology, Time Division Multiplexing, or TDM, was eminently suitable
to transform the voice telecommunications network from analog to
digital from the early 1970s. TDM, however, is an inflexible point-to-point
technology, with no sharing of fixed bandwidth allocations by traffic
type. Voice traffic based on fixed 64 kbit/s connection-oriented
circuits, transported between switching systems and to the end-user,
have been well served by TDM. But this technology cannot efficiently
cope with the dynamic, connectionless, and high-bandwidth traffic
synonymous with the Internet.
Telecommunications planners envisioned
that the network would be required to transport and switch more
than just voice, and, subsequently developed the concept of the
broadband integrated services digital network, or BISDN, to facilitate
multi-services communication. The BISDN was designed for implementation
using asynchronous transfer mode (ATM) technology, with the use
of a compromise 48-byte cell payload with a 5-byte header. This
cell size was selected based on the known and projected traffic
types in the early 1980s, well before the Internet and the World
Wide Web. ATM has its own management, addressing scheme, and quality
of service mechanisms, each different to those used by the Internet
Protocol.
Narrowband ISDN at 128 kbit/s was implemented
in the local loop as the precursor to the BISDN, but has met with
limited success for a wide variety of reasons, including that it
no longer able to cope with the data traffic demands of the Internet
and the responsiveness demanded by the Internet generation.
The 21st century will see the introduction
of the All Digital Loop (ADL)-based on packet technology closely
aligned to the Internet and the Internet Protocol. Both TDM and
ATM technologies can transport IP, but have significant drawbacks
both in efficiency and cost-effectiveness in doing so. Integral
Access' PurePacket technology, on the other hand, is optimized
for the transport and interaction with IP, as well as being able
to service 20th century traffic types, which will continue to be
with us well into the next century.
Return To Top
What is a "packet" based access system?
A packet based access system is a system
that packetizes all traffic and transmits all information through
the system in packet format. Packets contain a header or label with
source/destination addresses and other parameters, along with the
payload, and are able to be switched on a packet-by-packet basis
from system ingress to system egress. Forward-looking packet schemes,
such as the one used by Integral Access, also classifies traffic
with quality of service parameters so that the network serves higher
priority traffic before lower priority traffic. The PurePacket
access system operates in the access layer of the network and is
optimized for bringing traffic from customer premise to the central
office in an efficient manner. The system is fully interoperable
with third party equipment using standard protocols over standard
interfaces both in the customer premises and in the central office.
Packet access systems are bandwidth
efficient, and leverage the service-rich features of IP routing,
as well as the resilience of fiber transport systems, and do so
at a fundamentally lower cost but with higher functionality than
traditional access systems.
Return To Top
How does PurePacket enable carriers to transition to packet-based
networks?
PurePacket uses an MPLS-based
architecture to
- route,
transport and switch voice and data transmissions in packet format;
- guarantee
Quality of Service (QoS);
- enable
Class of Service (CoS); and
- maximize
the traffic carrying capacity of access networks.
PurePacket provides a migration
path to the pure IP networks of the future and supports SoftSwitch
Call Control for transporting voice calls across IP core networks.
To allow carriers to leverage their investments in existing TDM-based
infrastructures, PurePacket also supports the GR-303 and V5.2
standards for seamless interoperability with Class 5 voice switches.
Return To Top
What are the elements of the PurePacket System?
The PurePacket platform comprises
the following components:
- PurePacket
Node resides in the carrier's network, at the Central Office,
Collocation or Multi-Dwelling/Multi-Tenant Point of Presence
- PurePacket
OUTburst is a family of Integrated Access Devices (IADs) for the
customer premise. This IAD product family includes solutions for
small-medium business (SMB) and small office/home office (SOHO)
markets
- PurePacket
Operations Management System (OMS) is a powerful management system
for remote network and service provisioning, configuration, monitoring
and management.
Return To Top
What services does PurePacket support?
Carriers need to boost access bandwidth
and at the same time efficiently provide new-generation IP based
services to end-users. PurePacket addresses these critical
needs and also support customers with a wide range of legacy services
such as TDM voice, digital PABX and IBM SDLC transport, all over
a single access interface to the customer. The PurePacket
system is an IP-optimized system for next generation networks, but
also assists in network migration by providing efficient and effective
interfaces into the embedded switch, transport, customer premises
and services base.
Return To Top
Does the Integral Access PurePacket System comply with open
standards?
PurePacket is MPLS-based and supports
worldwide telecom-industry voice signaling and transport standards.
Standardized interfaces into IP, frame relay, and ATM networks ensure
that the PurePacket system seamlessly integrates into existing
telecom and data communications networks.
Return To Top
How does this compare with ATM, IP, or TDM based technologies?
Access networks are characterized by
the need to support large numbers of customers and service types
as economically and efficiently as possible. ATM systems have the
capability to transmit voice, data and video but have a large overhead
and are relatively expensive and complex to implement for low customer
and service densities and over low capacity links. IP networks are
an important class of packet network, but are optimized for the
transfer of IP (Internet Protocol) traffic alone and are ineffective
in the transport of other traffic types, such as TDM voice, frame
relay, SNA/SDLC, etc. TDM based systems can transport multiple traffic
types, but are inflexible and not optimized for IP and other dynamic
traffic types. A pure packet based system is native to most popular
traffic types such as IP (including Voice over IP) and frame relay
and can also transport circuit switched voice (TDM) traffic efficiently,
when quality of service measures are applied. Packet based systems
also transport voice over xDSL.
Return To Top
What is MPLS?
MPLS
or Multi-Protocol Label Switching is an IETF initiative that integrates
Layer 2 information about network links (bandwidth, latency, utilization)
into Layer 3 (IP) in order to simplify and improve IP-packet exchange.
MPLS gives network operators a great deal of flexibility to divert
and route traffic around link failures, congestion, and bottlenecks.
From a QoS standpoint, network operators will better be able to
manage different kinds of data streams based on priority and service
plan. For instance, those who subscribe to a premium service plan,
or those who receive a lot of streaming media or high-bandwidth
content, see minimal latency and packet loss.
What is the Role of MPLS in Integrated Multi-Service Access Networks?
The
single biggest challenge for service providers is migrating their
existing TDM-based access network to one that is optimized for delivering
packet-based IP services while still delivering TDM-class voice
services. MPLS provides the tools for network operators to manage
traffic, implement QoS and provision secure data Virtual Private
Networks (VPNs). Until now the majority of work around MPLS has
concentrated on solving the backbone traffic engineering problems
of large ISPs. The next step is to extend the reach of MPLS to the
customer premises in order to enable enhanced QoS and VPN capabilities
for delivering integrated voice and data services across packet-based
access networks.
Return To Top
With what other systems and networks can the Integral Access PurePacket
system interwork?
The PurePacket system can interwork
with a wide range of customer premises and core network equipment
such as
- analog
POTS, ISDN,
- T1/E1,
- SONET/SDH,
and
- ATM.
Voice switch interfaces are terminated
using standard TR-08, GR-303 and V5.2 protocols which are supported
by the major circuit-switched voice network suppliers such as Lucent,
Nortel, Siemens, Alcatel and Ericsson. The system also operates
as a Voice Access Gateway operating under 'SoftSwitch' control,
passing voice traffic to the IP Backbone. A flexible interworking
function within the PurePacket system allows the access network
to communicate with standard routers and frame relay switches using
MPLS and other protocols such as those supported by Cisco, Juniper
and Lucent.
Return To Top
How does Integral Access support third party CPE devices and equipment?
With standard physical interfaces for
POTS, ISDN, xDSL and T1/E1 CPE, equipment using these types of interfaces
can easily be interfaced to a PurePacket system. Inherent
support for TDM, frame relay and other data link layer protocols,
ensures that the majority of network protocols can be natively transported
and concentrated with the added benefit of quality of service being
applied.
Return To Top
How are optical rings created using PurePacket?
Integral Access has a patent pending,
redundant path data communications technique for creating optical
rings and other multiple path topologies, including point-to-point
and mesh. This approach includes the capability of sending packets
over multiple paths and for the receiver to determine which sent
packet to accept. By sending packets over more than one path to
the receiver, hitless levels of redundancy can be achieved very
economically.
Return To Top
What are the advantages in using optical rings based on PurePackettechnology?
One of the significant advantages of
the Integral Access packet optical ring is the ability to protect
traffic on a services and customer basis, rather than on a physical
or logical facility basis which is the case with SONET/SDH and other
traditional technologies. Customers that pay for full protection
can have their traffic protected using dual paths, while unprotected
traffic can use the balance of the dual path connection bandwidth.
In the event of a path failure, the protection traffic is guaranteed
to continue on a hitless basis, with other traffic transported on
the remaining bandwidth based on traffic priority levels.
Return To Top
How does this compare with SONET/SDH rings, ATM rings defined
in Telecordia GR-2837, rings based on a "thin layer of ATM",
or on Cisco's Dynamic Packet Transport technology?
SONET/SDH rings have dedicated protection
capacity and typically reserve as much network capacity for the
protection circuit as it does for the revenue-generating payload.
The SONET/SDH rings are also based on TDM technology and are provisioned
point-to-point with no sharing possible of this provisioned capacity.
While providing a high level of resilience with sub 50ms switchover
in the event of a fiber ring cut, the rings are bandwidth inefficient.
ATM and derivative rings typically utilize
some SONET/SDH framing and place ATM cells in SONET containers leading
to multiple layers of overhead and management. ATM routing protocols
alone are unable to provide the level of network protection and
automatic protection switching that is provided in standard SONET/SDH
facilities.
Cisco's DPT technology is suitable for
packet level transport at the OC-3 level and above, but does not
scale down to bandwidths below that which are the most prevalent
in the access network. The Integral Access PurePacket system
provides:
- The
network level protection of standard SONET/SDH,
- Bandwidth
efficiency synonymous with packet oriented systems through statistical
multiplexing gains, and
- Effective
operation on low speed links, which are predominant in the access
network.
Return To Top
How does PurePacket work with the Internet Protocol and
with routers?
PurePacket provides both layer
two and layer three transport services for IP. Layer two transport
through the access network is especially useful in cases where the
administration of the IP network is not the responsibility of the
access network provider. A layer two access network is fully transparent
to IP and the need to coordinate configuration between the two administrations
is minimized. Layer three transport requires coordination at the
IP level but provides more advanced tools for traffic management
and troubleshooting. Both IP transport modes are suitable for public
Internet Access and for the provisioning of secure corporate LAN-IC
services. The connection towards the IP backbone router can be based
either on Ethernet, PPP over SONET, MPLS or ATM.
Return To Top
How will PurePacket interwork with emerging MPLS devices
in the core network?
Most of the effort around MPLS has concentrated
on solving the traffic management problems of the IP backbones of
large ISPs. While MPLS is a very good tool for that purpose, MPLS
also enables the IP networks to provide CoS and security comparable
to that provided by Frame Relay and ATM, which are currently used
for provisioning corporate LAN-IC services. Integral Access PurePacket
provides the access solution required between the MPLS enabled IP
backbone and the sites of small and medium size enterprises or the
regional/branch offices of large corporations. Integral Access PurePacket
can be used both in cases where the customer has MPLS enabled equipment
and in cases where the customer's equipment only supports IP and
the labeling of traffic is carried out by the PurePacket access
network.
Return To Top
How does the PurePacket system support Virtual Private Networks?
Secure, flexible and economical data
VPN services present a huge revenue opportunity for public network
operators. MPLS is a protocol that allows provisioning of VPNs over
the same infrastructure that is used for providing public Internet
services. The use of MPLS makes VPNs secure and allows provisioning
of guaranteed bandwidth thus enabling the use of these VPNs in mission
critical applications. PurePacket is an optimized tool for
extending MPLS VPNs to the customer premises while simultaneously
being able to support the customer's voice access needs over the
same packet access infrastructure.
Return To Top
|