CompTIA Network+ N10-008 – Wide Area Networks (WANs)

  1. Wide Area Networks (WAN)

wide-area networks, or wands. We finally made it to the point where we could get outside our local area networks. So we’ve covered routing and switching and wireless networks, but all of that was really inside our network. Now we want to move outside our network. Now, when we look at computer networking over time, back in the early 1990s, we had this thing that we called the Pareto principle, which is actually a business principle and that we also know as the 80/20 rule. Now, as it applied to networks, that meant that 80% of your traffic stayed inside your local area network and only 20% went out to the WAN, or wide area network. And for that reason, we could have slow bandwidth pipes for our external connections.

Well, today that’s really been flipped on its head. Instead, it’s more like 80% of your traffic goes outside of your network and 20% stays inside. Let’s consider our businesses. Are you using things like the cloud? Are you using Google Drive and Dropbox? and Google Docs and Office 365 All of these cloud-based applications exist outside your network, and you probably spent a large portion of your day doing things like watching online video, going to YouTube, going to Facebook, and other things like that.

All of that resides on the WLAN, not the LAN, and not on your local area network. So let’s take a look at WANs and how important they are to our networks. How are we going to use wands? Well, we have to be able to connect to the outside world, and that’s what a WAN is for. So there are three main types of Wan connections. There are dedicated leased lines, circuit-switched connections, and packet-switched connections. When we talk about a dedicated lease line, this is a logical connection that’s going to connect two of your sites together through your service provider’s facility or a telephone company’s central office. This is more expensive than other WLAN technologies because the customer doesn’t share bandwidth with other customers.

If you have a direct copper cable, like a T1 connection to your office, that’s a dedicated lease line. Now, we’ll talk about all the different types of dedicated lease lines in the next lesson. For right now, I just want you to see the concept. The next type we have is called a “circuit switched connection,” and this is a connection that’s only brought up when it’s needed. It’s very similar to making a phone call. So back in the old days, we had dial-up networks.

Well, those were good, but they were pretty slow. And so there was a digital variety of that called ISDN, or integrated service digital network. We’ll talk more about those in the next lecture as well. But the idea was that it was like making a phone call. When you wanted the connection, it turned on and made the phone call. You could then go out to the Internet, and when you were done, it would hang up. Now this on-demand bandwidth can provide cost savings because you’re not getting 24/7 access. Instead, you’re getting it when you need it, and the other time that you’re not using it, the telephone company or your ISP can sell that service to somebody else.

Thirdly, we have packet switch connections, and these are like an always-on dedicated lease line, but multiple customers share the bandwidth, and your SLAs, or service level agreements, are used to guarantee a certain quality, like in your contract you might say that you’re going to get five megabits per second at least 80% of the time. These are virtual circuits, so they’re not dedicated connections like a T-1 or an E One.Instead, they are virtual, and they are represented as dashes when you look at them on a network diagram—things like frame relay or packet switch connections. Now, what are the types of connectivity that you’re going to use when you’re using a WLAN? Well, you have to have a way to get the ones and zeros out of your network. And that’s where physical media is going to come in. One of the most common ones is UTP, or unshielded twisted pair. We can use UTP to connect outside our network in the same way that we did with Ethernet. This will support analogue or digital connections, whether those are dial-up ISDN, E1, or T1 connections. And these are great, inexpensive, and easy to work with. Right now, the second type of cable you can have is coaxial cable, and this is usually RG-6 cabling.

A cable modem is an excellent example of this. So if you’re using Comcast or Time Warner as your Internet service provider, you probably are using a coaxial cable. Now, the next one you can use is fibre optic. These are high bandwidth, long distance andlow EMI, but they’re usually more expensive. If you’re going to have a direct fibre connection to your residence or your office, that can actually cost you a significant amount of money. Thankfully, prices have come down. And here in the United States, for instance, at my home, we have Verizon Fios, which is fiberoptic to your house and can support speeds of up to one gigabit per second. As the price of fibre continues to fall, more and more people will continue to use it.

Next, we have electric power lines. That’s right. If you have the right provider, those power lines outside your house can actually cover and provide Internet service. This is called broadband over power lines, or BPL. Now, this supports 2.7 megabits per second, which is pretty slow nowadays and pretty equivalent to slower DSL connections. The good news is that it was a good stopgap measure while we were trying to roll out these fibre and coaxial networks to everyone’s homes. All of us already had power lines. Now, this electrical power line is not very popular here in the United States, but in some countries, it’s a way for them to get Internet quickly because the power lines already exist and the infrastructure is already in place. Next, we have some wireless options. For instance, if you’re on the go, you might pull out your iPhone and use cellular. That can be a phone, a hotspot, a smartphone, or a tablet. They can have a modem that uses cellular signals. These can be things like 2G, 3G, 4G, or even the newest version, LTE, or Long Term Evolution, which is the fastest out there right now.

Now, when you’re dealing with cellular, you’re going to have to buy what is used in your area. So here in the United States, we have two different methods that are used. We have GSM, which is mostly used by providers like T-Mobile and AT&T. And we have CDMA, which is used by Verizon, Sprint, and all of the other providers. If you’re travelling outside of the United States, you’ll be using CDMA if you’re in South Korea or Japan, and GSM for the rest of the world. Now, which is better? Which should you use? It really doesn’t matter. Most of them are pretty equivalent at this point. The only thing you have to realise is that whichever one you’re going to use, you need to make sure the place you live supports it. So, ask your cell phone provider about hotspot options in your area. Now, your cell phone can act as a hotspot using tethering or Internet connection sharing, or you can get a dedicated device like the T-Mobile 4GLTE one shown on the slide. The next technology that we have for wireless is what’s called HSPA Plus, or Evolved High Speed Packet Access. Now, this has some advancements over the LTE networks, and it can provide wireless broadband speeds up to 84 megabits per second, which is really fast for networks on the go.

This is still evolving and getting rolled out across the country and across the world. And right now, LTE is still pretty much the standard. The last one we have here on the screen is YMAX, or Worldwide Interoperability for Microwave Access. Now, this is a good alternative to cellular or DSL service because it does provide faster speeds, but the radios for them are kind of large. As a result, unlike cellular, which can be used in your car or on the go, this is really intended for a home or business network solution to be used as a wide area connection. Now, this is a wireless fixed location service because you’re going to have to have a larger antenna and a larger radio to pick up that strong signal, but it is at a very good speed.

So if you see the chart there on the bottom, y max is faster than GSM. It’s faster than HSPA, but it’s not quite WiFi speeds, though it’s getting close. Now, there’s two other wireless media types we have to COVID but they’re only for very specific uses. The first is the satellite. Now, if you happen to live out in the country or in a remote area where you can’t get cable, cellular, fiber, or DSL, then you might want to look at satellite. The great thing about satellite connectivity is that you can get it anywhere. At this point, satellite Internet is available to almost everyone on the planet. Now, that even includes the ocean. So if you’re in a remote area of the country, you can use a commercial service like HughesNet Generation Five, which puts a small satellite dish on your roof, much like DirecTV or Dish Network would for TV. And you can then use that to get some decently high-speed Internet. But it is going to cost you more than the equivalent fibre, cable, or DSL service. The place where satellite really makes its mark for mobile users on the go So if you’re somebody who’s working out of a recreational vehicle like an RV or a truck, or if you’re on a boat and you’re in the middle of the ocean, satellite is useful for that. For example, about a month ago I took a cruise, and while I was on the cruise ship, I was able to get online and answer student questions because they had satellite Internet.

Now, it was very expensive to use, but it was easy to use, and the speed was exceptional. On the newer satellites, I was able to stream Netflix and Hulu, watch streaming video, and answer student questions with no problems at all, much like sitting at a cable or DSL modem. The final option is radio, which will be determined by your country in terms of which frequencies you can use and how they will operate. For example, there are the old high-frequency networks that we used to use for ham radio operators and probably still use today. You can actually run dial-up Internet over those services, but they only operate at about 9.6 kbps. So they’re very slow, but they can give you at least chat capability and some other Internet capabilities like that if you need to use them. Like I said, it’s a very specialised use case for most of us. We’re not going to be using radio to run our WANs.

  1. WAN Technologies (Part 1)

Wide-area network technologies We’re going to break this lecture into two parts because there are just so many technologies that we have to talk about. The first is a dedicated lease line. Now this is a point-to-point connection between two sites, and you get all of the bandwidth all the time, which is great. This will include T1, E1, T3, and other dedicated circuits.

Now, when you get this digital circuit, it’s going to be measured in 64Kbps channels called digital signal zero. and based on which connection you buy. For example, if you buy a T1, you’re going to get 24 of those 64-kilobit channels at your location. You’re going to get a device that’s like a modem, and it’s called a channel service unit, data service unit, or CSU DSU. This is what’s used to terminate the digital signal at your location and then tie it into your router to be able to connect it to your network. So for exam day, I want you to remember that ones are dedicated lease lines, and they use CSU DSU to connect to your network. Now, what are some examples of digital signal levels? Depending on your carrier signal (T 1, T 3, E 1, and E 3) and digital signal level (DS 1, DS 3, for E ones and E three S), Those are the European standards, so they don’t fall into this digital signalling measurement. And then you’ll see the number of channels that you’re going to get and how many voice channels or 64 kilobyte channels you’re going to get.

And your speed can be found on the right. What from this chart should you really memorize? You should be familiar with T1 and T3 speeds. And E1 speeds: if you remember that 1.54 megabits per second is a T 144.736, 45 megabits per second is a T 3, and 2.0 is an E1, you’ll be fine on test day. Next. We have Metro Ethernet. Now, service providers are beginning to get away from those T1s, E1s, and T3 connections because those CSUs and DSUs have just been kind of cumbersome to work with. Instead, they’re starting to migrate toward metro Ethernet. This is where the service provider instals an Ethernet jack in your building, and you simply plug it into your router. They’re less expensive and more common than using specialised serial ports in a CSU DSU. And the technology used by the service provider really doesn’t matter to me as a customer. I don’t care what’s behind that Ethernet jack. I just want to be able to connect to my network. And so by giving me something as simple as an RJ-45 that I can plug into my router and get online, that’s a great win for the customer. And it enables the search provider to change the backbone at any time. And that’s the benefit of metro Ethernet. Next we have the point-to-point protocol, or PPP. This is a commonly used layer-two protocol that we use on top of these dedicated lease lines, whether we’re using Metro Ethernet, a T-1, an E-1, a T-3, or an E three.

And this is going to allow us to use multiple layer-3 protocols simultaneously, like IPX and IP. Most of the time, you’re probably only using IP in your networks. Each layer three control protocol runs an instance of PPP’s link control protocol, which manages that link and does some basic error checking for you. This will perform the multilink interface for you, allowing multiple physical connections to be bound together to form a single logical interface, similar to how link aggregation works in switches. You could do the same thing where I can buy two or three T batteries, bind them together, and get more speed. It will also do looped link detection to find any kind of error for you. It’ll also do error detection by checking your frames, and it will perform basic authentication over the link to make sure you’re authorised to use that point-to-point connection. And the way it does that is with three different mechanisms, depending on which one you choose. You could be using PAP, which is a password authentication protocol.

You could be using the Challenge handshake authentication protocol, Chap, or Microsoft because they can’t seem to follow standards like everybody else, who made up their own called Ms. Chap, which is the Microsoft Challenge handshake authentication protocol. Now I say that a little tongue-in-cheek, but Microsoft actually made a better implementation of the old chat protocol when they made Microsoft Chat. Now, PAP, how does it work? Well, it performs one-way authentication between the client and the server. So in my basic diagram here, you’ll see I have a client on one side and the server’s router on the other. The credentials are sent in clear text from the client to the server. Basically, here’s my username and password, and the server comes back and says, “I acknowledge it, and I let you in.” Right, the bad thing about this is that PAP is in clear text, which means anyone can read your authentication and steal it from you. So we had to come up with a better way. And that’s where Chap comes in.

Chap is the Challenge Handshake authentication protocol, and it performs one-way authentication using a three-way handshake. So, when you want to connect to a router, in this case the server’s router, it begins by performing a challenge. Essentially, it says, “Hey, who are you?” Then the client will go back and say, “I am this person; here’s my username and password.” The router then checks the hashed credentials you sent of username and password, and if they match its stored version, it will send a success or failure message. Now, Microsoft’s version does the same type of thing, except there’s a two-way authentication there where the client verifies the server and the server verifies the client. For the exam, I want you to remember that PAP is in the clear, and that is a huge security risk. Chap and Microsoft Chap actually hash the credentials, making it a more secure way of doing PAP. Next, we have point-to-point protocol over Ethernet, or PPPoE. This was commonly used with DSL modems, and it actually took your PPP protocol that we would use over AT1, and it would encapsulate those frames within Ethernet. If you’re using Metro Ethernet, you’re likely using this as well. This allows for authentication to occur over Ethernet using something like PAP Chap or Microsoft Chap. Next, we have DSL, or digital subscriber line.

Now, when I first started building networks back in the late 1990s and early 2000s, DSL was all the rage. It was everywhere because it was a very inexpensive way to get high-speed data to our end users and our small office and home office environments. Instead of having to pay for a T1 connection, which may have cost us several hundred dollars a month, we could buy a DSL connection for $50 or maybe $100 a month. Now, there were three different types of DSL. ADSL, SDSL, and VDSL Now, ADSL is asymmetric DSL, and what that means is there’s a different speed on the download versus the upload. The maximum download speed was about eight megabits per second, according to current textbooks, although some were going a little bit faster than that. The upload speed was equivalent to a T1 with 1.54 megabits per second. Now, why would it be OK to have different upload and download speeds? Well, this goes back to the 80/20 rule. When you’re online, do you upload more or download more? For most of us, we download more. For example, when you’re watching this video right now, you selected the video with your mouse and told the server what you wanted to watch. That may have been one or two KB in size. The video I sent you back is 100 megabytes or more in size, and so it’s a very, very large file, and you’re downloading all of that, but your upload was very, very minuscule. And for the majority of users, they upload very little, but they download a lot. So DSL maximised the downloads and minimised your uploads. SDSL, or symmetric DSL, worked just like T One. They would have equal upload and download speeds.

Now, the speeds are much slower here for Symmetric, but they’re dedicated access, so you’re pretty much guaranteed that speed. With ADSL, if there were a lot of people on the line, it could actually slow down your speeds. Because ADSL was so popular with home users, it started getting a lot more funding, and the ADSL speeds increased quickly and got to those one and a half megabits per second, four megabits per second, and eight megs per second. When they first came out, they were all very slow, at about 256K or 512K. As technology progressed, they shifted more towards ADSL and abandoned SDSL. Now, the last one we have is very high-bitrate DSL. This is where you could get very, very high speeds. In fact, download speeds can reach 52 megabits per second and upload speeds can reach 12 megabits per second. Now, the big limitation here was your distance from the DSLAM. Now. What’s a DSLAM? That is the point of presence that is actually owned by the telephone company. So you could only be 4000 feet from them, which is less than a mile. So not all home users or offices could get VDSL, but most of them were within 18,000 feet and could get ADSL again. DSL has been declining in popularity in recent years as people have started moving towards cable and fiber, which we’re going to talk about.

  1. WAN Technologies (Part 2)

Area network technologies So we just finished up talking about DSL, and we talked about how there was a decline in popularity because cable modems and fibre modems started taking over. So let’s get started with cable modems. Cable modems use the cable TV network, which is made up of a hybrid fibre-coax, or HFC, distribution network. Cable TV can ride on top of it, but so can our Internet signals. And this is a mixture of coax and fibre optic cabling. There are specific frequency ranges that are used for upstream and downstream transmissions, as determined by the Data Over Cable Service Interface Specification, which is known as DOCSIS. Anytime on the exam you see HFC or DOCSIS, I want you to think of the answer as cable modems. Now, upstream, they use between five and 42. Downstream, they use 50 to 860 MHz. Don’t worry about those frequencies. But remember that the terms HFC and DOCSIS are associated with cable modems. They can transmit and receive over cable TV’s infrastructure, which has already been rolled out to most of America and many other places worldwide. As a result, getting cable modems into the market was a very quick sell.

And the increased speeds they offer over DSL have made them extremely popular. Now, the next one we have is satellite modems. Now, I’ve already touched on this briefly, but I’d like to go a little deeper now that we’re discussing technology. Now, satellite modems are used in remote,  rural, and disconnected locations, like out at sea, when other connections are not available. Satellite modems provide you with relatively fastspeeds like DSL modems, but not nearly the speeds of cable or fiber. And the other bad thing about satellite is that it provides low usage caps. So if you’re going to use a lot of streaming video services, satellite may not be the right answer for you because they may only let you use five or 10 GB per month. and that can go away pretty quickly if you watch a lot of Netflix. Now, there are some other issues with satellite that you have to be concerned with, and one of them is delay. You have to realise that when you’re going by satellite, you’re not just going across the country; you’re going out into space. And so you have to go from your satellite terminal all the way up to space and then all the way down to the ground station. Every time it goes up and down from the satellite, that is one quarter of a second right off the bat. So if you’re trying to do VoIP service over a satellite, for instance, it kind of has this echoing sound because there’s a delay in your voice. “How are you?” it’s like. You? and it’s just really annoying. But again, when you have that connectivity versus no connectivity, it’s better than nothing. There’s also an issue with weather connectivity because anytime you have a big thunderstorm or a snowstorm, that’s actually putting interference between your satellite dish and the satellite in space, which can cause an issue and a loss of service.

The next one we have is the plain old telephone service, or POTs. POT runs on what’s called the PSTN, or public switched telephone network. And this consists of all of the telephone carriers from around the world. So I can make a phone call from my house to my neighbour’s house or from my house to the other side of the world, and it’s still running on the PSTN using the POT service. This is analog. This is based on beeps and boops. These analogue connections can be voice when I’m talking, or data, using ones and zeros as a modulation of the voice, or beeps and boops using the PSTN. And these are called dial-up or POD connections. Dial-up modems had a maximum bandwidth of 53.3 kbps because they could only access one of the channels at a time. Because pots essentially used the same technology as T-1 connections, Now, why did I say “used to”? Well, because if you’re still running dialup, you’re not watching this video. Nobody really uses dial-up modems anymore. They’re just too darn slow for anything we really want to do on the real Internet nowadays.

And while you may find them in some very specialized, unique situations or legacy systems in your networks, they are simply not available to home and office users. The next one we have is the Integrated Service Digital Network, or ISDN. Now, I mentioned this briefly when we talked about circuit switch technologies. Because this is a circuit switch technology, it supports multiple 64K bearer channels. So it worked a lot like dial-up, but you could actually bind those channels together to get speeds similar to a T-1 connection. It’s an older technology designed to carry voice,  video, or data over these B channels. They also had D channels, which were data or delta channels that existed for signalling to be able to do the control measures. You don’t need to go into too much detail about ISDN because, honestly, you’re unlikely to come across one because most people now use cable, fiber, or a T1 digital connection. But they were very popular in the late 1990s.

They had two different types of circuits: BRI and PRY. basic rate interface, or primary rate interface. The Basic Rate Interface could basically give you 128 kbps by tying two B channels together. Prie could give you almost T1 speeds by putting 23 B channels together and using one as a data channel. Next, we have Frame Relay, and this is losing market share as well. Due to the cheapness of cable and DSL frame relay, we’re able to connect to virtual circuits. So, if I had two branch offices and a headquarters, I could connect them using point-to-point or point-to-multipoint connections, as shown in the diagram. This is a point-to-point connection. They were low-cost and widely available. They always provided on demand, based on how much you wanted to pay. They were also Layer II technologies. Frame Relay was really great for a long time, but again, because of the lower cost of fibre cable and DSL, that’s really been taking over in most small business environments. Next, we have SONET, which is the synchronous optical network. It is a layer-one technology that uses fibre as its medium.

So it is very, very fast. When we talked about T1 connections, we were talking about 1.5 megabits per second. When we talked about a T3, we were talking about 45 megabits per second. Now, when we’re talking about SONET, I’m talking about speeds of 155 megabits per second to start, and they go up to ten gigabits per second or more. Sonet uses transport layer two encryption like ATM or asynchronous transfer mode to do the shaping of those frames and really send this stuff around very fast. And because it’s fiber, it can go long distances. The physical topology for this can be either a bus or a ring. But most of the time I’ve seen this implemented as an FiDi ring, which has two counter-rotating rings for redundancy. Now, ATM rides on top of Sonit, and this is the layer-two way to shape those frames and send them over the network, much like we use Ethernet over copper cabling. Now, this is going to use two different things: permanent virtual circuits, or PVCs, and switched virtual circuits, or SVCs. It is similar to frame relay, except all of the frames have a fixed length, which we call cells, and we can move them around as its protocol data units because they’re all fixed length. We can have a shortened header that is only five bytes in size, which saves us a lot of time and actually makes this much, much faster.

Now, as you can see here on the diagram, I have broken out what an ATM header looks like for you. You don’t need to memorise all of those pieces, but do realise that it is a fixed header of five bytes and a payload of 48 bytes, giving us a 53-byte cell for ATM. And because it’s the same, it’s very easy to transfer around. Now, when we look at ATM virtual circuits, they are designated in two different ways. The user network interface (UNI) is what connects the ATM switches to the endpoints. And then we have these network node interfaces that connect the ATM switches together. Again, you don’t need to know this in depth except to know that if you see these two terms, I want you to think about ATM when it comes to test day. Now, the next one we have is what’s called MPLS, or multiprotocol label switching. Multiprotocol label switching is going to support multiple protocols on the same network. And MPLS is not something you’re going to use in your networks; it is something that your service providers use on their backbone networks. This can support frame relay and ATM on the same MPLS backbone.

And because everything is transferred using switching and labeling, it is extremely fast and works extremely well for backbone networks. This allows traffic to be dynamically routed based on load conditions and pass availability. And using this label switching, it is much more efficient than standard logical layer 3-based IP addressing. And so it makes quick work of all of this routing and switching for us. Again, this is used by service providers. So if you’re working for a large telecom company, you may run into MPLS, but in your own small office or home office network, you’re never going to be able to see MPLS. The customer doesn’t even know it exists. Most of the time, it just happens for the service provider. And lastly, we have the DMVPN, or Dynamic Multipoint Virtual Private Network. This allows the Internet itself to be used as a Wang connection between two sites and for secure communication. You create a VPN tunnel with authentication and encryption so that the users on the unsecured network cannot see what you’re doing. So if I have a couple of regional offices, I can connect them together over a VPN.

Now I can connect remote users with low-cost services without dedicated or leased line access. We’re going to talk a lot more about VPNs in their own lecture when we get into the network security section. Lastly, I wanted to talk about data rates. And when we talk about data rates, there are a couple that you just have to memorize. And so this chart is one you should add to your notes. Bandwidth can be measured in kilobits per second, megabits per second, or gigabits per second. ATM and Sonet are measured by optical carriers, or OC connections. They start out with OC-1, which is 51.84 megabits per second, and everything else becomes a multiple.

So if you’re using an OC 3, it’s three times 51. If you’re using an OC 12, it’s twelve times 51. But here on the chart, I’m going to show you just a couple that you need to memorize. Frame Relay. It goes from 56 kbps, which is a dial-up connection, all the way up to a T1 connection of 1.5 megabits per second. A T-1 connection is 1.5 megabits per second. A T3 is 45 or 44.7 megabits per second. E one and E three are shown here on the board as well. And these are both in Europe. It’s a two-megabit connection and a 34-megabit connection. And the last two are both available on fibre networks. ATM, which will be anywhere from an OC-3 connection up to an OC-12 connection, and Sonet, which uses anything from an OC-1 connection up to an OC-3072, which is 160 gigabits per second, which is quite fast. Now, do you have to memorise all the decimal points on this? Like 1.544 for this one? Well, no, because, if you remember, this is a multiple-choice exam. If you see the one that looks close, that’s the right answer. So put this in your notes, look it over a couple of times before test day, and you’re going to do great.