Transcript Network Management - Eastern Illinois University

School of Business
Eastern Illinois University
Network Management 1
(Week 15, Friday 4/21/2006)
(Week 16, Monday 4/24/2006)
© Abdou Illia, Spring 2006
Learning Objectives

Generating Useful Statistics
–
–

Availability
Reliability
Centralized Network Management
2
Generating Useful Statistics

Statistics: Data about network operation or network
devices operation

Example: Availability of a modem, Reliability of a Hub,
transmission speed, etc.

Statistics are very helpful for network management
–
–
Could help identifying problems in Network operation
Could be used to demonstrate the need to invest in technology
Q: What kind of tools, already introduced in class, can be used to generate useful statistics?
3
Availability
4

Availability: probability that a particular component
or system will be available during a fixed time period

Availability is function of:
–
–
Mean time between failures (Given by manufacturer or
generated based on past performance)
Mean time to repair (Found in studies or in our archives)

Mean time between failures (MTBF) is the average
time a device or system will operate before it fails.

Mean time to repair (MTTR) is the average time
necessary to repair a failure
Availability
 Standard equation:
A(t) = a/(a+b) + b/(a+b) x e-(a+b)t
in which:
a = 1/MTTR
b = 1/MTBF
e = natural log function
t = the time interval
 Approximation equation:
Availability% = (Total available time – Downtime)/Total available time
5
Availability
A(t) = a/(a+b) + b/(a+b) x e-(a+b)t
6
Suppose we want to calculate the availability of a modem that has a
MTBF of 3000 hours and a MTTR of 1 hour. The availability of this
modem for an 8-hour period is:
a = 1/1
b = 1/3000 = 0.00033
A(8 hours) =1/(1 + 0.00033) + 0.00033/(1 + 0.00033) x e-(1 + 0.00033)8
= 0.9997 + 0.00033 x 0.000335
= 0.9997
Q: What will be the availability of the modem if the Approximation equation is used?
Availability

7
A component has been operating continuously for
three months. During that time, it has failed twice,
resulting in downtime of 4.5 hours. Calculate the
availability of the component during that threemonth period using the Approximation method.
Availability

To calculate the availability of a system of
components:
–
–

Calculate the availability of each component
Find the product of all availabilities
Example: If a network has tree devices with
availabilities of 0.992, 0.894, and 0.999, the
availability of the network is:
0.992 x 0.894 x 0.999 = 0.886
8
Reliability

Reliability: probability that a component or system
will be operational for the duration of a transaction
time t.

Reliability is function of:
–
–
9
Mean time between failures
Transaction time

Mean time between failures (MTBF) is the average
time a device or system will operate before it fails.

Transaction time is the time interval of operation to
complete a given transaction.
10
Reliability
Reliability is defined by the equation:
R(t) = e -bt
in which:
b = 1/MTBF
t = the time interval of the operation
Reliability
11
What is the reliability of a modem if the MTBF is 3000 hours
and a transaction takes 20 minutes, or 1/3 of an hour (0.333
hours):
R(t) = e -bt
b = 1/MTBF = 1/3000
t = 0.333
R(0.333 hours) = e -(1/3000)(0.333) = e -0.000111 = 0.99989
Q: If a component has a MTBF of 500 hours and a transaction
takes 4 seconds, calculate the reliability of the component
12
Summary Questions
See slides # 6, 7, 11
School of Business
Eastern Illinois University
Network Management 2
© Abdou Illia, Spring 2006
Learning Objectives

List main elements in Centralized network
management

Describe Centralized Network Management
principles

Describe Manager-Agent communications
14
Centralized network management ?

Remotely managing a network from a central point (for
example a PC) :
–
–
–
–
–
–

15
Gathering data about network performance
Fine-tuning network performance
Diagnosing problems
Fixing problems
Controlling security
Etc.
Centralized network management is possible with:
–
–
Managed networking components, and
Communication systems based on Network management protocols
16
Managed networking component ?
Managed Switch
Software components
inside

Port 2
Status
On
Port 3
Status
Off
Managed Networking components
–
–

Object Property Value
On
Port 1 Status
Have objects that can be managed remotely
Include software components that gather data about their
operation.
Managed networking components can be printers,
hubs, switches, routers, application programs, etc.
Centralized network management
Network Management
Software (Manager)
 The Manager (an application
program) is used to communicate with
managed components.
 The RMON is a stand-alone device or a
software running in a switch or a router
 The RMON collects data on network
traffic
Switch
RMON device
* RMON = Remote Monitoring
17
Centralized network management
Network Management
Software (Manager)
18
Network
Management
Agent (Agent)
Network
Management
Agent (Agent)
 The Agents (software programs) acts
on behalf of managed device
 The Manager doesn’t communicate
with the managed device, but rather with
its Agent
RMON device
Centralized network management
Network Management
Software (Manager)
Management
Information
Base (MIB)
Management
Information
Base (MIB)
 The MIB is a data base on objects and
their characteristics (properties, etc.)
 There is a small MIB on managed device
There is a complete MIB on the manager’s
computer.
Management
Information
Base (MIB)
RMON device
19
Centralized network management
Network Management
Software (Manager)
Management
Information
Base (MIB)
Simple Network
Management Protocol (SNMP)
Management
Information
Base (MIB)
Command (Get, Set)
Response
Management
Information
Base (MIB)
Trap
RMON device
20
Summary Questions (1)
21
1) List the main elements in centralized network
management
2) Does the Manager communicate directly with the
managed devices? Explain.
3) Explain the difference between a managed device
and objects.
4) Where is the MIB (database) stored?
Manager-Agent communications
Network Management
Software (Manager)
22
Network
Management
Agent (Agent)
Simple Network
Management Protocol (SNMP)

Communication between the Manager and Agents
works through command-response cycles

The Manager sends a command

The Agent sends back a response
Manager-Agent communications
Network Management
Software (Manager)
23
Network
Management
Agent (Agent)
Command (Get, Set)

Get commands tell the Agent to retrieve certain
information and return this information to the Manager.
Example: Get status of port 1 and 2 on the switch

Set commands tell the Agent to set a parameter (a
property of an object) on the managed device. Example: Set
the status of Port 2 on the switch to “off”.
Manager-Agent communications
24
Network
Management
Agent (Agent)
Network Management
Software (Manager)
Trap

Sometimes Agents do not wait for commands to send
information

If the Agent detects a condition the Manager should know
about, it can send a Trap message to the Manager
Summary Questions (2)
1) In Manager-Agent communications, which
device creates commands? Responses? Traps?
2) Explain the two types of commands.
3) What is a trap?
25