Model Features: IP

This section provides a list of the main features available in the Internet Protocol model:

  • The IP model suite captures the following protocol behavior:

Table 97: IP Model Features

Feature Description
IP Addressing IP addresses are specified using dotted decimal notation (also called dotted quad notation). The model incorporates addresses from class A, B, C, and D types of addresses. Class D addresses are used for IP Multicasting, and are modeled in the specialized IP Multicasting model.
Routing and Forwarding IP routers are modeled using a finite buffer with two available processing modes: Slot-based and Central.
Routing in the IP model is provided by
• dynamic routing tables—created by dynamic routing protocols
• static routing tables—user-configured
Fragmentation and Reassembly The IP model uses the same principles and methods as actual IP implementations to break up datagrams before transmission and to reassemble them at the destination IP module. This is essential if the model is to produce the same traffic characteristics that occur in the modeled network.
Fragmentation occurs when the length of a datagram forwarded by an IP module exceeds the maximum transfer unit (MTU). Reassembly functions are performed on datagrams when they reach their final destination.
Processing Delay and Queuing Capacity The performance of IP-based networks depends on the characteristics and capabilities of the network’s switching elements. To support accurate studies involving these parameters, the IP model allows you to specify queuing capacities and packet processing speeds for each IP module.
Queuing Algorithms The model offers several queuing management options:
• a standard FIFO queue
• a weighted fair queue (WFQ) — Implemented as a bitwise round-robin fair queuing algorithm, it simulates the WFQ found in most routers.
• priority queuing (PQ)
• custom queuing (CQ)
• random early detection (RED)
• weighted random early detection (WRED)
IP Cloud Models The standard model library includes IP cloud models that allow you to abstract parts of a network’s infrastructure. A cloud behaves like a simplified IP-based router with many ports: it supports connectivity between attached devices, but does not include complex operations like WFQ. Clouds allow you to model only packet-loss and latency characteristics of a subnetwork. For more information, see “LAN and Cloud Models”.