In their operations, network control and monitoring systems can use a lot of bandwidth. More controlled devices and is the latest monitoring, but also require much more bandwidth than your system of control than usual traffic on the network.
Configuring a network monitoring tool: basic principles
The tool table is the basic command of a control system network. Usually you type ping host1 to obtain a series of results. But what are the transactions with this command?
By default, Windows, four ping packets to send 32 bytes to the host you specified and waits for a response for each package. Finally, calculate the percentage that was lost, the return period of maximum, minimum and average and as the following example:
Z:\> ping corerouter
Pinging corerouter [10.253.100.1] c on 32 bytes of data:
Reply from 10.253.100.1: bytes = 32 time = 1ms TTL = 255
Reply from 10.253.100.1: bytes = 32time <1ms TTL = 255
Reply from 10.253.100.1: bytes = 32time <1ms TTL = 255
Reply from 10.253.100.1: bytes = 32time <1ms TTL = 255
Ping statistics for 10.253.100.1:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milliseconds:
Minimum = 0ms, Maximum = 1ms, Average = 0ms
Z: \>
To improve and customize the table, you can use some switches. For example:
- Ping-n (count) tells the table how many times to send a ping packet. The default is four.
- Table W (milliseconds) tells ping how long to wait after each package, opened a response.
- Ping-l (bytes) tells ping how many bytes each ping packet should be.
Then, you can send a huge amount of traffic to a host by typing: Ping - L65500 - t killtherouter
This command transmits packets of 65,500 bytes to the host until the user says to stop pinging.
What are the control systems of the network?
Although the table is the most basic tool for the existing monitoring network will be like everyone else in the sense that they have few options to get the information you need without slowing down the network.
But there are other tools for network monitoring. Here there are some of the most popular:
- PRTG
- Cacti
- WhatsUp Gold
- Nagios (formerly known as netsaint)
Most network monitoring systems now control not only the status of node UP/DOWN but also the performance of node or network. All data are reported back to a database of some type to be analyzed, reported and graphed.
There are many facets to the network monitoring. Some network administrators to incorporate many types of network monitoring in this category, including the status of devices, device inventory, monitoring the performance, alerting, trending, and intrusion detection.
All the tools mentioned above not only include the list of network monitoring tools, but probably also in the tools for monitoring network performance.
What can improve the system of controlling the Internet?
With today’s network monitoring systems, virtually everything can be improved. But how to optimize the time for best performance? Most partAe of network monitoring systems today use SNMP to collect more information than just “you are there.” With SNMP, these systems can collect performance data and be notified when the network interface goes down or comes back on.
Generally, with each network monitoring application, you should be able to answer the following questions:
- What network node is controlled (for example, your core router or application server)?
- Which interface on that network node is monitored (for example, a router, you may interface GigE0 / 0; on a server, could be the CPU 1)?
- How often on that device polling takes place (for example, every 10 seconds or every 60 minutes)?
- How much information is collected each time the network node is “probed” by the system host?
- How are calculated and stored the data collected (or stored)? For example, your network monitoring system will calculate the average response time over the last five minutes and only store that value, or store all the values calculated in the last five minutes? This is a significant improvement.
How do I configure my network monitoring system for optional performance?
Among the applications of network monitoring available, it is worthy to advise of PRTG Paessler, which is free if you use it to check only a single network node (a router or just one server) and a single interface on that node (one interface on a single router). Besides this, PRTG is pretty inexpensive and easy to use. There are applications for network monitoring of network performance is completely free (for example, Nagios and Cacti) but none of them is simple, easy and comparatively the economic PRTG.
The amount of bandwidth sceglieteA use to check your network can be determined by the bandwidth you have available and what updates are important to you in real-time network. Configure an application like PRTG to explore your core router every 10 seconds or every 43,200 seconds? Whether you are managing only a single router on the LAN, a poll every 10 seconds does not create problems either connectivity or performance. On the other hand, if you are controlling devices 10,000 and you choose not to poll a network connection already busy, going to generate a flood of unnecessary traffic, possibly causing performance problems on the network.
In this sense, the followings are suggested:
- Find a balance in the frequency of polling of your network devices in order to reduce bandwidth requirements.
- Use a network protocol analyzer like Wireshark (which reflects the legacy of Etheral) to really understand how much traffic is generating your network monitoring, and then adjust the poll accordingly.
- Usually, the devices on the LAN can be used much more frequently than devices on the WAN.
- The more “long” is the average of the data, the lower the chances of tracing the peaks.