Response Time in Software

Creating Code 3 of 9 in Software Response Time

Response time is normally measured from the time a user requests a service to the time the service is granted Other factors such as capacity and traffic can affect the response time Performance management monitors the average response time and the peak-hour response time Any increase in response time is a very serious condition as it is an indication that the network is working above its capacity

Security management is responsible for controlling access to the network based on the predefined policy We discuss security and in particular network security in s 31 and 32

SIMPLE NETWORK MANAGEMENT PROTOCOL (SNMP)

Accounting management is the control of users' access to network resources through charges Under accounting management, individual users, departments, divisions, or even projects are charged for the services they receive from the network Charging does not necessarily mean cash transfer; it may mean debiting the departments or divisions for budgeting purposes Today, organizations use an accounting management system for the following reasons:

It prevents users from monopolizing limited network resources It prevents users from using the system inefficiently Network managers can do short- and long-term planning based on the demand for network use

SIMPLE NETWORK MANAGEMENT PROTOCOL (SNMP)

The Simple Network Management Protocol (SNMP) is a framework for managing devices in an internet using the TCPIIP protocol suite It provides a set of fundamental operations for monitoring and maintaining an internet

SNMP uses the concept of manager and agent That is, a manager, usually a host, controls and monitors a set of agents, usually routers (see Figure 282) Figure 282 SNMP concept

---,,-- 1-------'1;':'

SNMP is an application-level protocol in which a few manager stations control a set of agents The protocol is designed at the application level so that it can monitor devices made by different manufacturers and installed on different physical networks In other words, SNMP frees management tasks from both the physical characteristics of the managed devices and the underlying networking technology It can be used in a heterogeneous internet made of different LANs and WANs connected by routers made by different manufacturers

A management station, called a manager, is a host that runs the SNMP client program A managed station, called an agent, is a router (or a host) that runs the SNMP server program Management is achieved through simple interaction between a manager and an agent The agent keeps performance information in a database The manager has access to the values in the database For example, a router can store in appropriate variables the number of packets received and forwarded The manager can fetch and compare the values of these two variables to see if the router is congested or not The manager can also make the router perform certain actions For example, a router periodically checks the value of a reboot counter to see when it should reboot itself It reboots itself, for example, if the value of the counter is O The manager can use this feature to reboot the agent remotely at any time It simply sends a packet to force a 0 value in the counter Agents can also contribute to the management process The server program running on the agent can check the environment, and if it notices something unusual, it can send a warning message, called a trap, to the manager In other words, management with SNMP is based on three basic ideas: 1 A manager checks an agent by requesting information that reflects the behavior of the agent 2 A manager forces an agent to perform a task by resetting values in the agent database 3 An agent contributes to the management process by warning the manager of an unusual situation