R. E. Newman-Wolfe
, University of Florida
Last modified 1/30/95.
Protocols
An earlier page described how to
transfer frames
of data error-free from one node to another, another described
multiplexing
on a shared line, while others
described
circuit switching
and
packet switching.
The
previous page
described LANs and MANs.
All these allow machines that are directly connected to communicate,
but other procotocols are needed for machines that are only
indirectly connected.
This page describes protocols in general, and prepares the way for
network and internetwork layer protocols described in the next page.
- Services v.s. Implementation
- i. Virtual Circuit
- ii. Datagrams
- iii. Examples
- a. VC/VC - SNA
- b. VC/DG - Arpanet
- c. DG/DG - DNA
- Routing
- i. Objectives
- ii. Classification
- a. Adaptive
- b. Non-adaptive
- c. Centralized
- d. Distributed
- e. Isolated
- iii. Non-adaptive Algorithms
- a. Static - Directory Routing
- b. Random
- c. Probabilistic
- d. Flooding
- e. Selective Flooding
- iv. Adaptive Algorithms
- a. Centralized
- 1. Status-based
- 2. Delta Routing
- b. Isolated
- 1. Hot potato
- 2. Combined with static routing
- 3. Backward Learning
- c. Distributed
- 1. Optomistic
- 2. Hold-down
- 3. Chu's Algorithm
- 4. Dynamic Sink Trees
- v. Hierarchical Routing
- a. Need
- b. Technique
- c. Optimal number of levels, nodes per level
- d. Costs
- vi. Broadcast Routing
- a. Need
- b. Techniques
- 1. Flooding
- 2. Multi-casting
- 3. Spanning Tree
- 4. Reverse Path Forwarding (Approx. ST)
- Traffic Control
- i. Flow Control
- ii. Congestion Control
- a. Deadlock
- 1. Direct
- 2. Indirect
- 3. Reassembly Lockup
- b. By Circuit
- 1. Flow Control
- 2. Pre-allocation
- 3. Choke Packets
- 4. Traffic Shaping - Leaky Bucket
- c. Global
The next page describes protocols for
networking and internetworking.
This document is
copyright 1995
by Richard E. Newman-Wolfe.
Send comments to nemo@cis.ufl.edu