Continuous Monitoring of Hosts on the Web

Monitoring hosts in the local network is designed to control the number of devices connected to it, the serviceability of services and the collection of information on the average daily and peak load. These data are necessary for the system administrator for timely and rapid response to host failures, for optimizing the performance of the equipment based on the values of the measured peaks, and also for planning and carrying out the modernization of the local network in connection with the increase in the load.

There are server and client-server solutions for monitoring hosts . Typically, server scans are performed by periodically polling the entire range of IP addresses on the network, which shows all network devices connected to it. Such a scan is usually repeated every 5-10 minutes, which allows you to collect the current information and, at the same time, not heavily load the network with spam from insignificant requests. The second type of server monitoring of hosts is reading the cache of arp-tables from network routers. This method is faster than the previous one and does not require a flurry of parasitic requests that can acquire significant volumes with a large network size. However, its main drawback is the inability to detect connected hosts, which do not show any network activity at the same time.

Unlike server-based monitoring methods for hosts that provide extremely limited amounts of information, client-server solutions can constantly read a large number of computer parameters and send them for analysis to the server. The main advantage of such monitoring of hosts is the ability to build a fault-tolerant system with a repeated polling of the client in the absence of data from it for a certain period of time and automatic generation of an alarm signal. In addition to information about the actual presence of the device on the network, the client-server solution for monitoring hosts controls the services running on the remote computer, the correct operation of protocols, open ports, and much more. In this case, such network equipment as switches and printers is polled through all the same parasitic arp-requests or simple ping. A common drawback of all client-server methods of monitoring hosts is the additional load on the PC, especially - on the RAM, which can be felt on weak machines, especially in conjunction with several other applications to control other computer parameters.

Regardless of the way in which hosts are monitored , mechanisms for visualizing and building statistical reports are needed to work with the information received . As visual means of data display, graphs, diagrams, graphs of networks and subnets, as well as any other visual methods of data representation can be used. Especially convenient are the graphs for determining the number of working hosts in time. In addition, as a rule, all programs for monitoring hosts have the ability to quickly view information in the form of tables without using the graphic component. Reports can be built either on a preliminary request, or automatically for a certain period of time.

Based on the foregoing, when choosing the optimal solution for monitoring hosts on a local network , several important factors need to be considered:

1. Ability to organize a fault-tolerant system;
2. Presence of the graphic interface and means of construction of reports;
3. The principle of the system;
4. Functionality.

Only with the help of the analysis of the work of hosts can you create an efficient system with minimal load of the client PC and the network itself.