Resilient Packet Rings--Foundation For A New MAN.

To accommodate an exponential increase in data traffic, communication networks are going through a massive overhaul. Significant capacity increases have already been achieved in long-haul networks through the deployment of DWDM (dense wavelength division multiplexing) equipment, which enables data transmission on multiple wavelengths of light multiplexed onto a single fiber strand. At the same time, enterprise networks continue to migrate to higher levels of bandwidth. Many enterprise backbone networks are based on gigabit Ethernet solutions.

Metropolitan area networks (MANs)--the link between enterprise networks and long-haul networks, however, remain a bottleneck. Enterprises access public data networks using data services offering T1 (1.5Mbps), fractional T1, or multiple T1 data rates. Enterprise islands with gigabits of capacity access terabits of capacity in the core via megabit-capacity "sipping straws".

Ethernet--The Future Of Data Services

Most business networks today are built from a mix of network services that include private lines, ATM, and Frame Relay. While adequate for the data needs of the past, these services do not offer the scalability or achieve the cost points required to meet the market demands of the networked economy.

With growing customer demand for affordable network access at speeds of 10Mbps, 100Mbps, and more, there is a clear need for a new class of data service that is more scalable and cost-effective than what is available today. Ethernet services fill this void. Defined simply, an Ethernet service is any data service delivered via an Ethernet interface (a 10Mbps, 100Mbps, or 1Gbps Ethernet port). One of the first Ethernet services available in the market is the traditional point-to-point circuit, much like a T1 circuit or Frame Relay PVC but offered via an Ethernet interface. Over time, however, service providers will come up with more sophisticated data services such as multi-point bridged Ethernet services, application-specific Ethernet services, and multicast services, tailored to meet the demands of complex multimedia applications.

An Ethernet service offers several advantages over its legacy counterparts. Ethernet services tend to be far less expensive than traditional data services. This is because they can be deployed using packet switched networks, which offer significant gains in bandwidth efficiency and require lower capital expenditure. The savings can be passed on to the service provider's customers.

Ethernet services are also more cost-effective for the enterprise, because they eliminate the need for adapter modules or service-specific DSUs as required for ATM and Frame Relay. The customer's access router can connect directly to the network using one of its native Ethernet ports.

Another advantage of Ethernet service is its scalability. A service provider can drop a Fast Ethernet (100Mbps capacity) or Gigabit Ethernet (1000Mbps capacity) port to a subscriber once and upgrade many times, without additional truck rolls beyond the initial installation. Bandwidth and other service changes can be administered remotely, simplifying and facilitating service provisioning.

Resilient Packet Rings--Foundation For Advanced Data Services

Ethernet services, and data services in general, can be delivered over a number of different networking technologies. Incumbent networking technologies, however, were not designed to meet the requirements of carrying data through public networks. SONET's circuit-oriented nature is a great fit for TDM (time division multiplexed) traffic such as voice or nailed up circuits. The same circuit-oriented nature is also its handicap when it comes to carrying bursty data traffic.

Ethernet switching and routing products, on the other hand, though data-friendly, lack the robustness, scalability, and deterministic service required of public network infrastructure.

There is a compelling need for a new technology for metropolitan packet networks. The new architecture must be packet switched like Ethernet switches and routers, but it must also be reliable, scalable, and deterministic like SONET. In addition, it must incorporate features such as QoS, multicast, and application-awareness, which are essential to the successful deployment of data services.

Resilient Packet Ring (RPR) is an emerging network architecture and technology designed to meet the requirements of a packet-based metropolitan area network. An RPR network consists of a set of RPR switches connected together by two counter-rotating optical fiber rings. Data service subscribers connect to the RPR network via service interfaces (mostly Ethernet interfaces) on the RPR switches. The ring functions as a shared transport medium. Bandwidth on the packet ring is shared by all the nodes on the ring.

The RPR MAC (media access control) protocol ensures fairness and guarantees service quality (i.e. delay and jitter) for each subscriber's data service. It controls access to the ring and arbitrates competing requests for use of ring bandwidth. In addition, the RPR MAC implements a service protection mechanism to protect against ring failures (in the event of ring failures service can be restored within 50 milliseconds), and a congestion avoidance scheme that enables the system to operate near full-capacity while ensuring quality of service to all configured services.

The RPR MAC is being standardized in the IEEE 802.17 Resilient Packet Ring (RPR) Working Group. This is the first time the industry is defining a metro network architecture optimized for packet-based services.

Resilient Packet Ring networks will be used as feeder rings in the metropolitan area. These feeder rings, also known as collector rings, will be used in much the same way as SONET rings today, but for packet traffic. Packet access networks such as cable modem systems and fixed wireless networks will feed RPR networks as shown in the Figure. The RPR network aggregates the packet traffic from around the ring onto one or more highly concentrated links connecting to the network edge. At the network's edge, these links are the hand-offs to the long-haul DWDM and router networks that transport the traffic to its final destination.

Some RPR switches are application-aware and can offer deterministic service guarantees for individual application flows. RPR networks built using such switches enable service providers to differentiate their service offerings with application-specific services and service level agreements. Instead of selling bandwidth, which is quickly becoming a commodity, they can sell specialized data services that are far more profitable.

An RPR network is also ideal for the delivery of multicast services such as streaming media and video broadcasts. Competing networking technologies use hundreds of point-to-point connections to send a media stream to individual subscribers. An RPR network is far more efficient with multicast traffic distribution because one traffic stream can reach every node on the ring.

Resilient Packet rings will play a vital role in the architecture of the new PSDN (public switched data network). Designed from the ground up to enable efficient delivery of data in a public network environment, RPR is positioned to become the technology of choice for the creation of the new, high-speed optical MANs. Offering carrier-class reliability, optimized packet transport, point-to-point and multipoint services, RPR networks will provide network operators a reliable, cost-effective, and flexible edge infrastructure on which to deliver their advanced service offerings.

Kanaiya Vasani is marketing director at Lantern Communications (San Jose, CA).