Maximizing content-based billing strategies in next-generation networks

Jun 19, 2002 12:00 PM, Joe Aragona

Ask any carrier in today's market about revenue, and they will all confirm that basic voice, basic cable and basic Internet access are now commodities. The new telecommunications Holy Grail-the nirvana that will provide new revenue streams and lead to future profitability-is content-based billing.

Content-based billing allows carriers to bill for basic access and for the varying types of content and content-based services delivered over the network. It allows carriers to generate more revenue from existing bandwidth. Most importantly, it paves the way for a move from revenue models built on legacy systems and bulk bandwidth to models built on differentiated content-based services for individual market segments.

What do service providers need to achieve these objectives and begin to bill customers based on the content they are receiving?

They need a network that enables intelligent application management, facilitates the delivery of content-based services and drives content-based billing. That network must be able to properly authorize, authenticate and account for customized services by identifying individual packets and classifying them based on individual customer usage and QoS requirements. This can only be achieved in a network equipped with a classification processor that performs Layer 7 content inspection and classification.

Classification enables intelligent packet management

Classification processing enables intelligent packet management. It refers to the identification and classification of individual protocol data units (PDUs) in communication traffic moving through a network. It is the most crucial element in a network processing architecture that usually consists of four primary functional blocks:

• Classification, which establishes a relationship between a PDU and one or many members of a finite set of classes. Class is an abstract that can be expressed as an action or as a number

• Modification, which changes the contents of a PDU. In this case, by modification we mean PDU header editing, not massive payload modifications. Header editing involves insertion, removal and replacing of header fields

• Traffic Management, which shapes PDU traffic. This involves queuing PDUs for transmission with potential discard of PDUs based on a pre-defined discard policy

• Encryption, which encrypts and decrypts the PDU payload.

In most networking equipment such as multiservice switches and IP services gateways, classification of a packet consumes the largest share of the processing bandwidth. The network processor cannot determine how to modify, queue and shape data according to the required services until classification has been completed.

Therefore, the granularity of classification that is performed and the complexity of classification criteria that the network processing solution can handle, will determine how intelligent and powerful the networking device will be.

Three types of classification

There are three types of classification that can be performed by networking devices:

• single-field (SF) classification

• multi-field (MF) classification

• content inspection.

Single-field classification, which looks at a single field in the protocol header, is the most rudimentary and well-understood type of classification. It is typically performed in MAC bridging and IP routing. In these applications one field is being extracted from the protocol header and its value is usually looked up in a table. The result of the look-up determines what happens to the PDU. As long as the field is not too wide existing hardware solutions can handle the processing required and a co-processor approach to classification is not needed.

Multi-field classification is more complicated. With this type of classification, the lookup key is more complex (longer and composite), the set of values is richer and the process of forming the key is more complicated. In these applications, the ability to extract a field's values-to parse the field-is very important.

In a fixed protocol environment, where all protocol headers have a well-known fixed offset from the beginning of a PDU, parsing is relatively easy-IPv4 over 802.3 is a good example. But in multiprotocol environments-where PDUs are often encapsulated in protocols such as L2TP, PPP, IPsec, with MPLS shims-complex parsing operations are required to get to desired fields in the PDU.

Typical applications for multi-field classification are IP services devices and multiservice switches with features such as DiffServ and Access Control List. For these types of applications a classification co-processor is ideal.

Classification processors with extensive parsing capabilities are also necessary to perform multi-field classification in protocol environments such as IPv6, where headers have a variable length.

Given the need for edge equipment to provide increasingly intelligent traffic management capabilities, there is a need for more granular and powerful classification that will enable the delivery of truly differentiated services with varying levels of QoS. That classification capability is provided by content inspection.

Content inspection is the key

Content inspection classification refers to the ability to examine deeply into packet content to make classification decisions based on long character string values (i.e. hundreds of characters).

To perform content inspection, the processor must parse five layers of protocol headers and then perform regular expression matching. This type of classification is best performed by a classification processor, which can look for patterns that involve hundreds of characters anywhere in the packet.

With an integrated classification processor designed to perform content inspection, all networking equipment can be programmed to find, identify and process specific data units of content based on source and destination IP addresses, TCP or UDP port combinations related to a particular user, and locate the uniform resource locator (URL) and the cookie in a hypertext transfer protocol (HTTP) header.

 

Content inspection provides the edge

Without proper classification it is impossible to manage and deliver value-added, customized, differentiated services with true QoS.

Content inspection provides networking equipment with the ability to inspect the content of each packet entering a network processing architecture. The information collected can then be used to enable informed routing and switching decisions for intelligently managed content-based services.

With content inspection, service providers have the foundation for true content-based billing. They can properly authorize, authenticate and account for customized services based on individual customer usage and QoS requirements.

Most importantly content inspection allows carriers to bill for basic access and for the varying types of content and content-based services delivered over the network. It allows carriers to generate more revenue from existing bandwidth and paves the way for a move from revenue models built on legacy systems and bulk bandwidth to models built on differentiated content-based services for individual market segments.

Joe Aragona is Director of Marketing for Solidum. He can be reached at joea@solidum.com.

Visit Solidum online.