Date of Issue: 05-01-2023 Rate this Study Guide

Managing Performance and QoS Part 1: General Principles and Ingress Handling

by Howard Berkowitz

What Problem Are You Trying to Solve?
  Problems Differ at the Edge and Core
    Essence of Edge and Core Issues
  End-to-End Management
    A Note of Reality
"Round Up the Usual Suspects": Performance Is More than QoS Mechanisms
    Some Things Aren't the Way We Would Like Them to Be
    The Router Queues Only When Necessary
  Classification and Marking
  Enforcement Part 1: Packet-Oriented
  Enforcement Part 2: Connection- and Flow-Oriented
  Output Queuing
General Concepts
  Sources of Delay
    Internal Router Processing Delay
    Serialization Delay
    Propagation Delay
    An Ideal Model
    Wait a Minute...Why Does a Host Have an "R" Router Designation?
  Buffering and Queuing
    Queuing Delay
    Tail Drop
    Shaping and Policing
  Statelessness, Soft State, and Hard State
  End-to-End Relationships
    Bucket Models of Traffic
    Flow Control
  DiffServ Architecture
    A Special Case: Expedited Forwarding
  Classification, Marking, and Metering
    Essential versus Desirable
    ToS, IP Precedence, and DiffServ
    Looking Forward to IPv6
    DiffServ Changes in the Header
IOS Language Features Used in Performance Management
Handling Incoming Traffic
  Configuring Classification
    Configuring Class Maps for Classification
    Policy Maps
    QoS Policy Propagation via BGP
  Traffic Policing
    Priority Queuing (PQ)
    Committed Access Rate (CAR)
    Random Early Detection (RED) and Weighted RED
  More about Queuing
    First-in-First-Out (FIFO)
    Configuring WFQ
    Class-Based Queuing, Custom Queuing, and Other Relatives
    Special Cases for High-Speed Routers
  Traffic Shaping
    Policy Maps
    Generic Traffic Shaping (GTS)
    Distributed Traffic Shaping (DTS)
    Frame Relay Traffic Shaping (FRTS)
    Low-Latency Queuing (LLQ)
Conclusion and Looking Ahead

Copyright © 2002 Genium Publishing Corporation