Please enable JavaScript.

Coggle requires JavaScript to display documents.

7.5.5 (4) (Network Management Tools (Response time reports (reports…

- - - - reports should be reviewed by IS management and system support personnel to track potential problems
        
        If response time is slow, all possible causes, such as
        
        input/output
        (I/O) channel bottlenecks,
        
        and central processing unit
        (CPU) capacity,
        
        an appropriate and cost-justified corrective action should be taken
        
        various solutions should be analyzed;
        
        bandwidth utilization
        
        should be investigated
  - - - power/line failure #
      - operator error
        
        identified in a downtime report.
      - traffic overload #
      - or other anomalous conditions
    - - • Add or replace telecommunications lines.
      - • Switch to a more dependable transmission link (such as dedicated lines versus shared lines).
      - • Install backup power supplies.
      - • Improve access controls.
      - • Closely monitor line utilization to better forecast user needs, both in the near and long term.
  - - - perform all of these tasks, each device
        (routers, switches, hubs, PCs, servers) needs to have a SNMP agent running.
      - actual SNMP communications occur between all the agents and the console.
    - - Based on the operator instructions,SNMP software sends specific commands to an SNMP-enabled device and retrieves the required information.
  - - - help desk provide a history of the problems and their resolution.
    - - since they provide end users with an easy means of identifying and resolving problems quickly,
        
        before they have a major impact on IS performance and end-user resource utilization.
- - - - network OSs provide many functions aimed at shaping the network as a unified, controlled and uniform computing environment, i
        
        INCLUDING
        
        • Support local and remote terminal access to hosts and servers.
        
        • Support sharing of common network resources, such as file and print services.
        
        • Establish links to hosts and servers.
        
        Network OSs have the following user-oriented features:
        
        • Allow transparent access to the various resources of the network hosts.
        
        • Check the user authorization to particular resources.
        
        • Mediate and simplify the access to remote resources as easily as local resources.
        
        • Establish uniform logon and logging procedures throughout the network.
        
        • Make available up-to-the-minute online network documentation.
        
        • Permit more reliable operation than possible on a single host or server, particularly when groups of equivalent hosts are used.
      - software is the network OSs software associated with specific network devices, principally switches and routers
- - - - to a lesser extent, because signals must travel some distance (propagation delay).
  - - - In telecommunications, it is the number of bytes per second that are passing through a channel
- - - - Detects the devices that present some kind of technical fault
    - - Allows users to know, define and change, remotely, the configuration of any device
    - - Holds the records of the resource usage in the WAN (who uses what)
    - - Monitors usage levels and sets alarms when a threshold has been surpassed
    - - Detects suspicious traffic or users, and generates alarms accordingly
- - - - a server (a source of services and data)
        
        the server provides data distribution and security functions to other computers that are independently running various applications
      - or a client (a user of these services and data that relies on servers to obtain them). I
        
        Clients may be thick or thin.
        
        THIN #
        
        thin client (sometimes called a lean client) is a client computer or client software that depends primarily on the central server for processing activities and
        
        mainly focuses on conveying input and output between the user and the remote server
        
        FAT/THICK #
        
        does as much processing as possible and passes only data
        for communications and storage to the server.
    - - distributing the work among servers and performing as much computational work as possible on the client workstation to save bandwidth and server computing power.
      - manipulating and changing data, may be performed locally and without the need for controlling resources on the main processing unit. In this way, the applications may run more efficiently
    - - Each processing component is mutually dependent on the others.
        
        That tasks are performed on both client and server is the main difference between client-server processing and the traditional mainframe/distributed processing.
    - - Two tiered
        
        • A thick client, focused on GUI tasks and running the application logic
        
        • A group (one or more) of database servers
        
        main disadvantages of this model are the requirement to keep the programs on the clients synchronized (ensuring that they are running the same logic) and its scalability.
      - Three tiered
        
        • A thin client, focused on GUI tasks (most often but not always webbrowsers)
        
        • A group (one or more) of application servers, focused on running the application logic
        
        • A group (one or more) of database servers
        
        Advantages
        
        Thin clients, which are less complex and less costly to buy and maintain
        
        More scalability (up to several thousands of concurrent users) because the load is balanced among different servers.
        
        improves overall system performance and reliability since more of the processing load can be accommodated simultaneously
        
        Can be implemented in applications for internal usage only or in ebusiness applications
        
        All of the program logic is separated from the rest of the code (via application servers
      - N-tirered
        
        each instance of the client software can send data
        requests to one or more connected servers. I
        
        servers can accept these requests, process them and return the requested information to the client
        
        applied to many different kinds of applications the
        architecture remaining fundamentally the same.
        
        interaction between client and server is often described using SEQUENCE DIAGRAMS. .
        
        Sequence diagrams are standardized in the Unified Modeling Language