Question:
Why do LANs have distance limitations?
?
2010-04-20 11:57:48 UTC
Question from a book in my Computer Networking class. Thanks to all who answer.
Five answers:
Andrew S
2010-04-20 12:38:14 UTC
There are several issues. For ethernet the primary limitation is propagation time and the collision detection system used. A transmitter listens to the wire while it is transmitting and if it detects other traffic it sends a jamming signal so that neither packet gets through - that is desirable behaviour since both packets have probably been corrupted.



However, signals do not travel down the wire instantaneously, they take time to arrive. If the line was too long then the time taken for the signal to reach the other end of the cable would increase. If that time gets too long then two machines could simultaneously transmit an entire packet each and they would only collide after the packets had apparently been sent successfully. For this reason it is necessary to engineer things so that the shortest possible ethernet frame must take longer to transmit than the signal to travel the full length of the cable.



This is most notable with gigabit ethernet - the faster speeds meant that a packet could be transmitted faster and break this rule. To counter this the minimum size of an ethernet frame was significantly increased so as not to break the collision detection system.
Masticina Akicta
2010-04-20 12:07:36 UTC
Well think about it this way the materials cables are made of define how fast a signal can travel. Not only that but also metal through what a stream of electrons goes sends out a certain level of electro magnetic signal. It is the way magnetrons work!



But also they can recieve it, yes if you put a metal coil in a magnetic field it will become electronic charged.



Signals require a certain level of signal clarity, if you send a certain signal you desire that the other side gets the same signal. Yes on both sending and recieving point already methods are inbuild to deal with altered signals for instance if you run a simple UTP wire next to a washing machine or a magnetron that isn't shielded well. Trust me the signal/noise ratio will be terrible, the amount of dropped packages will be trouble.



Now this doesn't anwsers it yet but here it comes there is a limitation on wiring based upon the power that is used to send the signal. If a wire is to long there isn't enough power to push all the electrons and voila no signal on the other side.



Most equipment we use has serious limitations in how much power you can put on it. Hence that limits how long the wire can be. UTP wire can be longer then it's limitations only if you put a repeater between it. Yes if you use two UTP cables and use repeaters you can go very far. The signal requires amplification.
Dhiran S
2010-04-20 12:00:36 UTC
Usually the limitation is to do with the wires. For example, the concept of attenutation is a biggie that network builders must at all times be considerate off. To be brief, a signal can only be sent so far down a line before it "dies".



The problem gets worse if your using wireless. Wireless can only be reliably used over short distances. But unlike wired, there is a lot of variables to consider with wireless. For example, if you sit maybe 15 ft away from your wireless and monitor the signal, then get your dog (hopefully a big dog) to stand in front of the router. You will notice a considerable drop in your signal because the dog is made up of so much fluid.



Wired connections don't have these pitfalls, though from what I can remember from college, I think if someone shines a UV light onto the wire - even though its coated in plastic - the UV will still disrupt the signal. This is something you would have to take into consideration if you were putting a LAN in a niteclub. Obviously wireless is out because off all the people (people are made up from water, the same as the dog) and the wires cant be exposed to any UV lights. That would be a complex LAN to design/construct.
2016-04-12 02:31:06 UTC
hello, first of all, since you're saying your NIC si gigabit lan enabled means that should you ever need it or be in a network where everythign communicates with 1000baseT cables your NIC can work effectively. Gigabit is needed when you're transfering large files from computer to computer via network and in home environment a gigabit ethernet is a waste, as the 100Mpbs is more than enough. Gigabit, at this time and in my opinion, really only needs to be used in a corporate environment with heavy network traffic. At home, it's hard to find home users who truly need those kinds of speeds to transfer data between computers. ISP connections are not as fast as your home network, thus the 100mbps connection. ISPs bottleneck the connection speeds to 100. I'm not aware of any home connections that really exceed 100Mbps. Even then, realizing those speeds is very hard. Using a 100Mbps adapter you don't really get the whole 100% performance, however having a gigabit NIC will offer a 100% effectiveness on the 100Mbps adapter (NIC). Gigabit gets a lot of hype for it's speeds but it doesn't help out your internet connection speeds. That solely falls upon your ISP. Gigabit, in my opinion, really isn't needed in the home user market. Stick with 100mbps and you'll see the same speeds as if you switched to 1000mbps unless transferring huge data files between computers. You could go to device manager and open up the properties of your network card. Set under the speed 1000Full duplex as the settings. Even restart if you wish and then go to view the status of your internet connection (n the connections folder). You will see that nomatter the specific speed which you have now set at 1000 the connection is still 100. SO..you connection was never at 1000 unless you worked for EA games, Ubisoft, Microsoft etc etc and companies like that where transfering >1GB files over network is necessary. another gigabit limitation: The problem is how the Gigabit Ethernet chip is connected to the system. If it is connected to the standard PCI bus, it probably won’t achieve its full speed. PCI bus works with a maximum transfer rate of 133 MB/s, while Gigabit Ethernet runs up to 125 MB/s (1.000 Mbps / 8 = 125 MB/s). By just observing these two numbers you could say that Gigabit Ethernet “fits” PCI bus, but the problem is that PCI bus is shared with several other components of your system, thus lowering the available bandwidth. So, even though in theory Gigabit Ethernet can run fine on PCI bus, it is just to close to the bandwidth limit of the bus. PCI Express, on the other hand, has a maximum transfer rate of up to 250 MB/s and is a point-to-point connection, which means that it doesn’t share this 250 MB/s bandwidth with any other device, thus allowing Gigabit Ethernet to achieve its full speed. How can one tell which bus Gigabit Ethernet chip is connected to? There are three basic ways. The easiest way is to see if your motherboard is based on PCI Express bus. If it doesn’t, the Gigabit Ethernet chip can only be connected to the standard PCI bus where will perform as if it was a fast ethernet (100Mbps).
Willy
2010-04-20 12:10:40 UTC
Dhiran is correct about the signal degrading. This can be negated somewhat by having repeaters along the way. This will take the signal while there is still enough to be read, and then digitally amplify it and send it along its way.



As for UV with wired connections, this can be negated with shielded cables and not just using regular plastic coated wires/cables. I think that cat6 cable comes shielded.


This content was originally posted on Y! Answers, a Q&A website that shut down in 2021.
Continue reading on narkive:
Loading...