Introduction
Lets first see what IT is. IT (information technology) is a term that encompasses all forms of technology used to create, store, exchange, and use information in its various forms (business data, voice conversations, still images, motion pictures, etc). It's a convenient term for including both telephony and computer technology in the same word. It is the technology that is driving what has often been called "the information revolution”. In this paper we are dealing with the combination of telephony and computer technology that is INTERNET.The explosive growth of the Internet in the last decade challenged service providers to expand the capacity of their Internet backbones at an extraordinary pace. While this rapid build-out delivered the required connectivity, the resulting network lacks the features necessary to control and manage bandwidth. This significant infrastructure limitation has prevented the Internet from becoming the ubiquitous, revenue generating packet service network many had predicted. Without the ability to guarantee bandwidth and control capacity, Internet backbone networks fall short of creating a suitable alternative to higher-priced but less adaptable data services such as Frame Relay or Private Line.
Yet despite its lack of bandwidth control, the Internet introduced a compelling new packet service feature -connectivity to any point, anywhere. All previous packet services provided only point-to-point bandwidth, limiting the scope of the service. The Internet provides global any-to-any connectivity allowing businesses to easily connect any and all corporate sites worldwide. This functional gap between public and private networks opens up an opportunity for the service provider with a packet service network that supports any-to-any connectivity as well as bandwidth guarantees.
This is an opportune time for new packet network architecture. The next phase of service provider network expansion is just beginning, driven by the broadband modernization of local exchange carrier (LEC) infrastructure and enabled by new low-cost optical transport solutions. Unlike the infrastructure build-out that supported the rapid expansion of the Internet and focused only on capacity and connectivity, this new build-out requires attention to overall infrastructure cost and support for the evolution of high-value services.
A new type of network router is the conduit to the new packet network. Next-generation service edge routers are high-capacity, high-density intelligent network routers designed and architected for the service edge. Service edge routers represent a cost-effective solution for service providers to go beyond capacity and connectivity, supporting the evolution of scalable IP services. By expanding packet network capacity using next-generation service edge routers, service providers.
· Introduce Virtual Private Network (VPN) services that combine connectivity with guaranteed bandwidth to create a new highly profitable revenue stream.
· Easily and quickly create profitable Internet services, such as tiered destination-sensitive packet delivery, that combine connectivity with bandwidth guarantees.
· Reduce the cost and complexity of the Internet backbone infrastructure.
· Reduce data service network infrastructure cost by consolidating onto a common MPLS packet switched infrastructure.
· Introduce new high-capacity packet switched services to continue meeting bandwidth demand and attract new customers.
· Prepare for expansion of network and service capacity in a cost-effective manner.
Basic Definition
RouterA router is used to connect broadcast domains. This helps in better use of the bandwidth .due to reduction in broadcast the network congestion is avoided.
Router is a layer-3 device. The main function of the router is to logically connect two different networks.
Edge Router
A term used in asynchronous transfer mode (ATM) networks; an edge router is a device that route data packets between one or more local area networks (LANs) and an ATM backbone network, whether a campus network or wide area networks (WAN). An edge router is an example of an edge device and is sometimes referred to as a boundary router.
Core Router
A core router is a router that forwards packets to computer hosts within a network (but not between networks). A core router is sometimes contrasted with an edge router, which routes packets between a self-contained network and other outside networks along a network backbone.
Switching
Switch is a network device that selects a path or circuit for sending a unit of data to its next destination. A switch is a simpler and faster mechanism than a router, which requires knowledge about the network and how to determine the route. There are two types of switching Circuit Switching and Packet Switching.
Internet Backbone Case Study: The Route to Complexit
Few Internet Service Providers (ISP) were prepared to meet the massive bandwidth demand created by the explosion of the Internet in the late 1990s. In an effort to gain market share and keep up with demand, providers doubled the capacity of their networks several times a year. Network architectures unfolded that were made up of a variety of products, incrementally adding specialized routers and switches from multiple vendors to grow capacity and address interoperability.
With this rapid expansion, Internet Service Providers were successful in keeping up with demand, yet their resulting architectures were overly complex and limited in service delivery by the narrow capabilities of routers and switches. What started as a network with two layers -- one core and one access -- quickly became a network with five or more layers between the customer access link and long-haul capacity.
In the core, routers lacked the capacity to keep up with demand. As a result, high-capacity switches were introduced to compensate for router limitations. However, these switches were unable to replace core routers because they used ATM control protocols, which were incompatible with the IP control protocols available on routers. This forced an IP overlay network. Core routers could not be removed because the remaining access routers would have been forced to communicate in a full mesh, stressing the IP control plane and limiting network growth.
In the access layer, access routers lacked the capacity and port density to meet demand for new higher-speed customer connections. To meet the demand for high-speed service, ISPs added more access routers and additional interfaces on core routers for interconnection. The ports on core routers were quickly used up and an additional layer of switches was introduced in the POP to interconnect access and core routers. The access layer was further complicated by the lack of traffic management features on access routers. As service providers introduced tiered bandwidth services to cover the bandwidth gap between interface speeds (e.g. there is a gap between T1 at 1.5 Mbps and DS-3 at 45 Mbps), they needed bandwidth management and service aggregation. To address this need, yet another layer of switches were introduced between customer links and access routers. While each addition of equipment was critical to meet demand and add customers, the end result is a network architecture that is costly, complex and difficult to scale, with excessive layering of routing and switching technology. The new IP service network dictates a new approach with fewer layers and an end-to-end infrastructure capable of delivering scalable IP services.
In the core, routers lacked the capacity to keep up with demand. As a result, high-capacity switches were introduced to compensate for router limitations. However, these switches were unable to replace core routers because they used ATM control protocols, which were incompatible with the IP control protocols available on routers. This forced an IP overlay network. Core routers could not be removed because the remaining access routers would have been forced to communicate in a full mesh, stressing the IP control plane and limiting network growth.
In the access layer, access routers lacked the capacity and port density to meet demand for new higher-speed customer connections. To meet the demand for high-speed service, ISPs added more access routers and additional interfaces on core routers for interconnection. The ports on core routers were quickly used up and an additional layer of switches was introduced in the POP to interconnect access and core routers. The access layer was further complicated by the lack of traffic management features on access routers. As service providers introduced tiered bandwidth services to cover the bandwidth gap between interface speeds (e.g. there is a gap between T1 at 1.5 Mbps and DS-3 at 45 Mbps), they needed bandwidth management and service aggregation. To address this need, yet another layer of switches were introduced between customer links and access routers. While each addition of equipment was critical to meet demand and add customers, the end result is a network architecture that is costly, complex and difficult to scale, with excessive layering of routing and switching technology. The new IP service network dictates a new approach with fewer layers and an end-to-end infrastructure capable of delivering scalable IP services.
Packet Network Architecture in Only Two Layers
Using service edge routers, service providers can implement a new lower-cost, higher-performance network architecture that reduces the number of layers between the core and access optical networks. A reduction to two layers (from the five or six found in many of today's service provider architectures) decreases complexity and operating costs. At the same time, service edge routers increase performance and add the functionality required to support new scalable services. The new service edge router-based architecture is shown 
The streamlined two-level design deploys service edge routers at the network edge and high-capacity MPLS switches at the core. The advanced functionality of next-generation edge routers is key to reducing the number of devices required, including the number of intra-POP links.
This new two-level POP architecture provides the cost-effective scalability and reduced complexity required to support new, revenue-creating services.Cost Effective Scalability,Reduced Complexity.
This new two-level POP architecture provides the cost-effective scalability and reduced complexity required to support new, revenue-creating services.Cost Effective Scalability,Reduced Complexity.
Evolution, Not Revolution
Although there are multiple migration strategies to successfully introduce service edge routers into networks, the recommended approach is to deploy service edge routers along with existing equipment to scale network capacity and introduce new services.refer above fig
This approach provides the following benefits:
· Offers an incremental upgrade strategy that follows planned network expansion. Service edge routers are installed in locations as higher speeds or new services are required.
· Protects existing investment in data service networks (i.e. Frame Relay, ATM) by redeploying switches and routers in the new infrastructure.
· Focuses new infrastructure investments on the converged MPLS packet architecture, which eliminates the cost associated with upgrading multiple service-specific packet networks.
· Enables new service offerings to customers over their existing connections
· Offers an incremental upgrade strategy that follows planned network expansion. Service edge routers are installed in locations as higher speeds or new services are required.
· Protects existing investment in data service networks (i.e. Frame Relay, ATM) by redeploying switches and routers in the new infrastructure.
· Focuses new infrastructure investments on the converged MPLS packet architecture, which eliminates the cost associated with upgrading multiple service-specific packet networks.
· Enables new service offerings to customers over their existing connections
The Service Edge Router: Combine and Conquer
The emergence of MPLS (Multi protocol Label Switching) technology enables the creation of a new class of product that will dramatically change the way the Internet and packet service networks are created: the service edge router (sometimes referred to as an optical edge router or next-generation edge router). The service edge router is deployed in the service provider POP (Point-of-Presence), providing the link between the long-haul MPLS backbone and the metropolitan or access networks. It has the capacity to match that of an optical core network, and the aggregation capabilities to handle diverse traffic coming in from the access network. By enabling service creation at a centralized POP, the more device-intensive access network is subject to fewer equipment and technology upgrades, simplifying network management and setting the stage for more scalable IP services.The long-haul network remains focused on capacity. At the network edge, the service edge router relies on MPLS switching that is compatible with the IP routing protocol control plane used to create connectivity in the Internet, eliminating a major complexity of today's infrastructure. Using MPLS, the high capacity, high-density service edge router combines the routing protocol control plane of a modern Internet router with the underlying switching technology of a modern packet switch. The service edge router uses the routing protocol control plane to discover the location of each address in the network. In addition, unlike an Internet router, the service edge router switches traffic to its destination using a pre-established circuit (known as MPLS Label Switched Path or LSP) with associated bandwidth.
By mapping packets onto LSP, the service edge router enables a new packet network architecture that is capable of delivering services with the any-to-any connectivity associated with traditional routers and the guaranteed bandwidth associated with switches. With this new packet network architecture, service providers simplify their infrastructure and create a network purpose-built for scalable IP services.
Opportunities that Scale
With a new architecture in place, service providers can turn their attention to delivering high-capacity, scalable services. From a business perspective, the more easily a service provider can introduce and scale a particular service, the greater the return. Successful services will be those that:· Easily and cost-effectively scale to meet customer demand.
· Provide the flexibility of the any-to-any connectivity found in the Internet backbone and corporate IP networks.
· Are compatible with existing network infrastructure and protocols to enable a smooth transition and reduce the cost associated with new service introductions.
· Transition existing customers to a new service. Delivering IP services with the same level of bandwidth control as ATM and Frame Relay represents a compelling transition plan for current customers.
Opportunity 1: New Virtual Private Network Services
Virtual Private Networks (VPNs) provide connectivity and guaranteed bandwidth like that found on private lines, but across a common IP/MPLS backbone. For service providers, VPNs are a compelling service that can be offered to customers for applications such as private content distribution networks or as the basis of a managed network service. With a service edge router, layer 2 and layer 3 VPNs can be implemented that use the routing control plane to discover the location of addresses in the VPN, then map these addresses to MPLS LSPs in order to provide guaranteed bandwidth across the core.
Opportunity 2: Premium Internet Services
Premium Internet services are a way for service providers to offer differentiated service levels for Internet traffic, and set up the accounting mechanisms to charge customers at different billing rates. Using an MPLS backbone, the service edge router allocates bandwidth for different service levels and then maps customer traffic onto LSPs. The service edge router based on the classification of the IP packet controls the service level applied to the traffic and the billing counter incremented. This can be set up as a very fine-grained classification, often based on the destination of the traffic.
Opportunity 3: Scaling an Internet Backbone
Scaling Internet capacity to keep pace with demand remains a top priority for service providers. The current multi-layered network infrastructure used by many ISPs is difficult to scale because of the number of devices and complexity of the architecture. The routing functionality of the service edge router improves scalability of the Internet backbone by collapsing many layers into just two. At the core, the edge router eliminates the need for core routers by using MPLS switching to interconnect from edge to edge, resulting in a single core layer. At the access layer, the service edge router connects directly to optical access networks without the need for external grooming -- any port, any service.
Opportunity 4: Scaling an Existing Data Service Backbone
Current data service networks based on an ATM backbone are not scaling well due to the lack of traffic aggregation features in ATM. MPLS provides multiple levels of traffic aggregation (instead of two as with ATM). Additional levels of aggregation are critical to adding more customers and higher capacity links to the network as well as introducing new meshed layer 2 VPN services.
Scaling is also limited by the capacity of the links. ATM switch architectures are designed for link speeds of 622 Mbps (OC-12 / STM-4) or less, which is less than the capacity of many customer access interfaces. The service edge router can scale the capacity of the ATM backbone by extending it across an MPLS core. The service edge router maps ATM connections onto LSPs to transport ATM traffic across high-capacity MPLS backbone links and switches. By moving traffic onto the MPLS core, the service provider can reduce the level of investment in the legacy ATM network and concentrate new investments on a higher-capacity common packet network based on IP/MPLS.
Scaling Internet capacity to keep pace with demand remains a top priority for service providers. The current multi-layered network infrastructure used by many ISPs is difficult to scale because of the number of devices and complexity of the architecture. The routing functionality of the service edge router improves scalability of the Internet backbone by collapsing many layers into just two. At the core, the edge router eliminates the need for core routers by using MPLS switching to interconnect from edge to edge, resulting in a single core layer. At the access layer, the service edge router connects directly to optical access networks without the need for external grooming -- any port, any service.
Opportunity 4: Scaling an Existing Data Service Backbone
Current data service networks based on an ATM backbone are not scaling well due to the lack of traffic aggregation features in ATM. MPLS provides multiple levels of traffic aggregation (instead of two as with ATM). Additional levels of aggregation are critical to adding more customers and higher capacity links to the network as well as introducing new meshed layer 2 VPN services.
Scaling is also limited by the capacity of the links. ATM switch architectures are designed for link speeds of 622 Mbps (OC-12 / STM-4) or less, which is less than the capacity of many customer access interfaces. The service edge router can scale the capacity of the ATM backbone by extending it across an MPLS core. The service edge router maps ATM connections onto LSPs to transport ATM traffic across high-capacity MPLS backbone links and switches. By moving traffic onto the MPLS core, the service provider can reduce the level of investment in the legacy ATM network and concentrate new investments on a higher-capacity common packet network based on IP/MPLS.
RESEARCH EFFORTS:
OXC : - fastest optical cross connects promise to fastest switching technology between high speed fibers, but it is clear how well they integrate into IP over optical networkMPLS : - multiprotocol label switching is an effort to traffic engineer high volume flows and may ease the integration of ATM, VCs and IP. It may be more efficient to utilize load sensitive IP streams if they could be implemented without instabilities for example by optimising OSPF weights or e.g. using an ensemble Riccati state controller.BBN Labs : - 50 Gbps research effort based on the Butterfly switching matrix, ARPA funded, Craig Partridge involved, forwards 6 to 42 million packets per second depending on configuration. This was one of the earliest multigigabit routers built.
British Telecom : - 100 Gbps research effort, pure optical routing, read header in optical IC and marked packets are optically switched into another fiber, 100 Gbps expected for the first light box.
Broadcom : - 9 Gbps world's first single chip switch with routing (L3) and quality of service. The Strataswitch BCM5600 (CMOS) can filter up to 6.6 Mpps and supports 24 10/100 interfaces and two 100/1000 interfaces which can all run at wire speed. The chip can run data, voice and video simultaneously.
IPSILON :- invented IP switching, a method to efficiently put TCP/IP over any network fabric, routing calculation on AMD/Intel, can be used e.g. over ATM, alliance with DEC, developped IGRP, competitor to Cisco, but seems to have flamed out recently and was purchased by Nokia.
0 comments:
Post a Comment
Thanks for your Valuable comment