network management

What belongs together and its an abstract thought process.

• Fault management is concerned with monitoring the network to ensure that everything is running smoothly and reacting when this is not the case.
• A large amout of time

is the most important aspect of network monitoring.

• know if switch goes down not all pc
• Send alarms to a pager or send text message

• Event –By device or by severity level or both
• duplicate alarms- maintain history

recognizes patterns of alarms caused by minor faults that point to impending bigger problems

helps track problems

Perform operations that will deliver and modify configuration settings to equipment in the network

Configuring managed resources, whether they are network equipment or services running over the network

Retrieving information from a device already known to exist

Using software to see what is in the network

• The network itself
• Management system’s view of the network
• Views should be the same

your user data is the configuration data of your network

• New feature releases or patches
• (bug fixes)
• You need to be able to keep track of which software images are installed

Collect usage data is based on volume, duration, and/or quality

with tracking down and preventing theft of communication services

Throughput, measured by a number of units of communication performed per unit of time.

• At the application layer for a web service, the number or
• percentage of web requests that could not be serviced. Will be low

• Set up proper processes and procedures to ensure orderly operations
• Intrusion detection and prevention

the day-to-day running of the network

• required to manage the network and that do not involve performing changes (configuring, tuning) to the
• running network itself.

the network andcommunication services operate as they are supposed to. This involves diagnosing, troubleshooting, and repairing

proper setting of configuration parameters on the network so that the network functions as expected

• has the concept of a management lifecycle at its center
• lifecycle stages
• –Fulfillment—Assurance—Billing
• (FAB).

a service order that was received from a customer is carried out properly

all activities required to ensure that a service runs smoothly after it has been fulfilled.

sure that the services provided and resources consumed are accounted for properly and can be billed to the user.

• Congestion
• TCP/IP protocol weakness

permanently increase performance.

• Can enhance the wrong applications
• Can actually reduce throughput

• Cannot detect congestion until after it has occurred
• Bydefinition, packets wait in line, which can result in retransmissions

• Appropriate compression
• Best algorithm is different for different types of traffic

• Best effort
• Packet retransmission can contribute to WAN congestion
• UDP doesn’t manage an End-to-End connection

• 4:1 compression ratio for
• some measurable amount of traffic

• Traffic policing propagates bursts.
• Throw away slower traffic first

retains excess packets in a queue and then schedules the excess for later transmission over increments of time

a packet/traffic shaper

• Partition size = 250kbps; burstable;
• limit=none

• Off-line configuration
• (sometimes called a sandbox)

Theoretical value for TCP window size is the bandwidth delay product

• Nearly all traffic generators follow a client/server model
• client generates data
• D-ITG (Distributed Internet Traffic Generator)

Can be used to determine UDP characteristics

• A More Sophisticated Traffic Generator
• Available for Windows and Linux
• Can generate traffic having different shapes
• and different packet size options