1.5 Service Architecture

The service architecture depends partly also on the communication paradigm of its information units.

Every networking environment has service architecture, much like the postal delivery system. There are 3 service models associated with IP networks as follows:

BEST-EFFORT SERVICE ARCHITECTURE: Consider an IP network.

		The basic information unit of an IP network is a packet or a datagram which is forwarded from one router to another towards the destination. To do that, the IP network uses a switching concept, referred to as packet switching.
		This means that a router makes decisions by identifying an outgoing link on a packet-by-packet basis instantaneously after the packet arrives.
		At the conceptual level, it is assumed that no two packets are related, even though they might arrive one after another and possibly for the same web-page downloaded.
		Also, recall that at the IP level, the packet forwarding function is provided without worrying about reliable delivery; in a sense, IP makes its best effort to deliver packets. Because of this, the IP service paradigm is referred to as the best-effort service.

1.5 Service Architecture

INTEGRATED SERVICES ARCHITECTURE: This model worked because the data rate provided during a session can be adaptive.

		The concept for integrated services ("int-serv") architecture was developed to allow functionalities for services that are real-time, interactive, and that can tolerate some loss, but require a bound on the delay.
		Furthermore, each session or connection requires a well-defined bandwidth guarantee and a dedicated path.

DIFFERENTIATED SERVICES ARCHITECTURE: The differentiated services ("diff-serv") architecture was developed to provide prioritized service mechanisms without requiring connection-level information to be maintained at routers.

		Specifically, this approach gives priority to services by marking IP packets with diffserv code points located in the IP header.
		Routers along the way then check the diff-serv code point and prioritize packet processing and forwarding for different classes of services.
		Diff-serv code points are identified through a 6-bit field in the IPv4 packet header; in the IPv6 packet header.

1.5 Service Architecture

IPv6: IPv6 (Internet Protocol Version 6) is also called IPng (Internet Protocol next generation) and it is the newest version of the Internet Protocol (IP) reviewed in the IETF standards committees to replace the current version of IPv4 (Internet Protocol Version 4).

IPv6 Packet Header: An IPv6 packet is a block of data that contains a header and a payload. The header is the information necessary to deliver the packet to a destination address; the payload is the data that you want to deliver. IPv6 packets can use a standard or an extended format.

Figure: 1.3: Standard IPv6 header format

1.5 Service Architecture

Standard IPv6 header: The 40-byte IPv6 header consists of the following eight fields:

	Version - Indicates the version of the Internet Protocol.
	Traffic class - Previously the Type-of-Service (ToS) field in IPv4, the traffic class field defines the Class-of-Service (CoS) priority of the packet. However, the semantics for this field (for example, diffserv code points) are identical to IPv4.
	Flow label - The flow label identifies all packets belonging to a specific flow (that is, packet flows requiring a specific Class of Service [CoS]); routers can identify these packets and handle them in a similar fashion.
	Payload length - Previously the total length field in IPv4, the payload length field specifies the length of the IPv6 payload.
	Next header - Previously the protocol field in IPv4, the Next Header field indicates the next extension header to examine.
	Hop limit - Previously the Time-To-Live (TTL) field in IPv4, the hop limit indicates the maximum number of hops allowed.
	Source address - Identifies the address of the source node sending the packet.
	Destination address - Identifies the final destination node address for the packet.