Filter
Top Products
baseline network characteristics, quickly spotting potential trouble spots and
resolving them before major crises occur.
10.2 RMON Goals
To ensure that RMON can function effectively and efficiently in a distributed
environment, its framework was designed with the following goals:
•
Offline operation
•
Preemptive monitoring
•
Problem detection and reporting
•
Value-added data
•
Multiple managers
10.2.1 Offline Operation
The offline operation prevents the remote network monitoring functions from
being crippled due to network failures or lost communications with the
management station. Fault, performance and configuration information will be
accumulated continuously on the network probe when communication with the
management station is not possible or efficient. When an exception condition
occurs, the probe will attempt to notify the management station.
In wide area network (WAN) environments or when dial-up links are used, this
feature can help reduce communications costs.
10.2.2 Preemptive Monitoring
Preemptive monitoring provides for historical statistical information to be
collected and stored. Historical information captures trends, related events and
information that can allow one to witness the occurrence of the problem. By
playing back the historical information collected, a detailed diagnosis can be
performed to ascertain the cause of a problem.
10.2.3 Problem Detection and Reporting
The RMON agent is fully responsible for problem detection and reporting on the
segment that it manages. It can be configured to detect specific conditions such
as error conditions and continuously monitor their occurrence. When a problem
is detected, the RMON agent is configured to take the necessary action such as
logging the event and raising an alert to the RMON manager.
10.2.4 Value Added Data
Statistics are important for problem determination. Occasionally, statistics don′t
portray the complete scenario for a detailed diagnosis to be performed on the
problem. Since the RMON agents are directly monitoring the LAN segments,
they intimately know what happens in the segments. As such, they are able to
add significant value to the data being collected. For example, we can know
who is causing the network utilization to rise above the threshold level, if the
probe can highlight the hosts that generate the most traffic.
194 8260 Multiprotocol Intelligent Switching Hub