WEBVTT 00:00:02.080 --> 00:00:03.919 When you're designing, building, or 00:00:03.919 --> 00:00:05.839 troubleshooting a network, you'll be 00:00:05.839 --> 00:00:08.720 using a network topology type that can 00:00:08.720 --> 00:00:10.960 vary depending on the technology you're 00:00:10.960 --> 00:00:13.120 using. If you're looking to understand 00:00:13.120 --> 00:00:15.280 the way the data is flowing, or you want 00:00:15.280 --> 00:00:18.000 to plan to create a new type of network, 00:00:18.000 --> 00:00:19.600 then it's useful to understand what 00:00:19.600 --> 00:00:22.080 these topologies might be. This is also 00:00:22.080 --> 00:00:23.439 useful during the troubleshooting 00:00:23.439 --> 00:00:25.680 process because you'll know exactly what 00:00:25.680 --> 00:00:27.920 flow the traffic takes to get between 00:00:27.920 --> 00:00:30.720 point A and point B. One of the most 00:00:30.720 --> 00:00:33.440 popular types of topologies is the star 00:00:33.440 --> 00:00:35.200 topology. You might also hear this 00:00:35.200 --> 00:00:37.440 referred to as a hub and spoke, where the 00:00:37.440 --> 00:00:39.520 hub is in the middle and the spokes are 00:00:39.520 --> 00:00:42.320 along the outside. This is a topology 00:00:42.320 --> 00:00:44.480 that you'll find on almost any network, 00:00:44.480 --> 00:00:47.120 regardless of the size of the network. 00:00:47.120 --> 00:00:49.360 And you'll also find that most devices 00:00:49.360 --> 00:00:51.840 are connecting back to this central hub 00:00:51.840 --> 00:00:54.480 of the star. For example, a switched 00:00:54.480 --> 00:00:57.039 Ethernet network has the Ethernet switch 00:00:57.039 --> 00:00:59.680 in the middle of this star. And then all 00:00:59.680 --> 00:01:02.960 of the devices run directly back to this 00:01:02.960 --> 00:01:05.040 particular switch. They're not connecting 00:01:05.040 --> 00:01:07.119 to each other. They're instead connecting 00:01:07.119 --> 00:01:09.360 back to the central switch in this star 00:01:09.360 --> 00:01:11.119 topology. 00:01:11.119 --> 00:01:12.960 Although we don't often see a ring 00:01:12.960 --> 00:01:16.000 topology used on our local area networks, 00:01:16.000 --> 00:01:17.920 it's still a topology type that's used 00:01:17.920 --> 00:01:20.720 quite often for wide area networks. So 00:01:20.720 --> 00:01:22.560 although many people will remember the 00:01:22.560 --> 00:01:24.799 older Token Ring technologies that we 00:01:24.799 --> 00:01:26.880 used to run inside of our local area 00:01:26.880 --> 00:01:29.200 networks, we don't generally find those 00:01:29.200 --> 00:01:31.439 ring networks any longer. But if you're 00:01:31.439 --> 00:01:33.439 connecting over a metropolitan area 00:01:33.439 --> 00:01:36.320 network or a wide area network, we use 00:01:36.320 --> 00:01:38.960 ring networks extensively. It's not 00:01:38.960 --> 00:01:41.439 because a ring technology somehow lends 00:01:41.439 --> 00:01:43.600 itself to work better over a wide area 00:01:43.600 --> 00:01:45.680 network. It's because we can create 00:01:45.680 --> 00:01:48.159 additional redundancy using the ring 00:01:48.159 --> 00:01:49.520 topology. 00:01:49.520 --> 00:01:51.759 For example, a very common way to send 00:01:51.759 --> 00:01:54.000 traffic over a ring network is to have 00:01:54.000 --> 00:01:56.640 the traffic simply go in a circle. Now, if 00:01:56.640 --> 00:01:58.479 we're on this wide area network and 00:01:58.479 --> 00:02:00.320 there's construction going on and 00:02:00.320 --> 00:02:02.320 someone happens to sever a fiber 00:02:02.320 --> 00:02:04.159 connection that's being used for this 00:02:04.159 --> 00:02:06.640 metropolitan area network, then we're not 00:02:06.640 --> 00:02:08.640 going to be able to send that traffic 00:02:08.640 --> 00:02:10.640 through the rest of that ring. But the 00:02:10.640 --> 00:02:13.200 devices that are on either side of that 00:02:13.200 --> 00:02:15.680 severed link recognize that traffic is 00:02:15.680 --> 00:02:17.599 no longer able to traverse that 00:02:17.599 --> 00:02:19.840 connection, and instead will loop back 00:02:19.840 --> 00:02:22.239 the connection on those individual 00:02:22.239 --> 00:02:24.560 endpoints. So instead of having data go 00:02:24.560 --> 00:02:26.879 around a ring, the data will instead go 00:02:26.879 --> 00:02:29.440 as far as it can around the ring and 00:02:29.440 --> 00:02:31.840 then loop back to get to the other side 00:02:31.840 --> 00:02:34.000 of the ring, maintaining uptime and 00:02:34.000 --> 00:02:36.640 availability even in the case where part 00:02:36.640 --> 00:02:39.519 of that ring may be severed. 00:02:39.519 --> 00:02:41.360 Early types of Ethernet networks were 00:02:41.360 --> 00:02:43.360 not switched Ethernet, but were instead 00:02:43.360 --> 00:02:46.480 run over coax. And this coax was quite 00:02:46.480 --> 00:02:48.879 simply a cable that was run down the 00:02:48.879 --> 00:02:51.440 middle of the room, very similar to this 00:02:51.440 --> 00:02:54.239 cable. This is a bus network. And although 00:02:54.239 --> 00:02:56.080 it was commonly used on those early 00:02:56.080 --> 00:02:58.319 Ethernet networks, we can still find 00:02:58.319 --> 00:03:01.920 modern networks that use the same bus topology. 00:03:01.920 --> 00:03:04.400 One problem with bus networks is that it 00:03:04.400 --> 00:03:06.800 is a single cable that is running either 00:03:06.800 --> 00:03:08.879 through the walls or down the center of 00:03:08.879 --> 00:03:11.120 the room. And if we happen to have a 00:03:11.120 --> 00:03:13.440 break in this cable, you can see 00:03:13.440 --> 00:03:15.599 immediately that it would suddenly 00:03:15.599 --> 00:03:17.519 segment the network into different 00:03:17.519 --> 00:03:20.400 pieces, or in some cases, cause no data to 00:03:20.400 --> 00:03:22.879 be transferred across the network. That's 00:03:22.879 --> 00:03:24.959 one of the reasons we moved away from 00:03:24.959 --> 00:03:27.680 bus networks for our local area networks-- 00:03:27.680 --> 00:03:29.760 because one single disconnect could 00:03:29.760 --> 00:03:33.280 cause an outage for everybody else on the network. 00:03:33.280 --> 00:03:35.680 In our modern automobiles, we have bus 00:03:35.680 --> 00:03:37.680 networks that we use extensively. These 00:03:37.680 --> 00:03:41.280 are Controller Area Network buses, or CAN 00:03:41.280 --> 00:03:43.200 bus connections, and they're used to 00:03:43.200 --> 00:03:45.200 connect all of the different sensors and 00:03:45.200 --> 00:03:47.760 controllers inside of our automobiles to 00:03:47.760 --> 00:03:49.840 be able to make all of our cars much 00:03:49.840 --> 00:03:52.480 safer to drive on the roads. 00:03:52.480 --> 00:03:54.640 Another popular topology, especially in 00:03:54.640 --> 00:03:57.040 larger networks, is to create a mesh 00:03:57.040 --> 00:04:00.319 between devices or a mesh between sites. 00:04:00.319 --> 00:04:02.480 We may have devices that are connected 00:04:02.480 --> 00:04:04.720 in different locations, and we might want 00:04:04.720 --> 00:04:06.480 to connect them all together. But instead 00:04:06.480 --> 00:04:08.480 of having a single connection to a 00:04:08.480 --> 00:04:10.720 particular site, we may want to create 00:04:10.720 --> 00:04:12.799 multiple connections to mesh these 00:04:12.799 --> 00:04:15.760 together. That way, if we do lose any one 00:04:15.760 --> 00:04:17.840 of these network links, we're able to 00:04:17.840 --> 00:04:20.320 work around that problem by simply using 00:04:20.320 --> 00:04:22.800 one of the redundant connections. 00:04:22.800 --> 00:04:25.040 You'll commonly use this type of mesh 00:04:25.040 --> 00:04:26.960 design if you're creating redundancy or 00:04:26.960 --> 00:04:28.800 fault tolerance, or perhaps you're 00:04:28.800 --> 00:04:30.800 designing a load-balanced network and you 00:04:30.800 --> 00:04:32.479 can use different parts of the network 00:04:32.479 --> 00:04:34.000 to share that load. 00:04:34.000 --> 00:04:36.479 Probably the most common place to find a 00:04:36.479 --> 00:04:38.960 wired mesh network is over a wide area 00:04:38.960 --> 00:04:41.120 network, where you can create multiple 00:04:41.120 --> 00:04:43.040 links to other sites so that you can 00:04:43.040 --> 00:04:45.120 have a primary connection from one site 00:04:45.120 --> 00:04:47.040 to the other, and then a backup or 00:04:47.040 --> 00:04:49.120 secondary connection that you can use if 00:04:49.120 --> 00:04:51.440 you run into problems. 00:04:51.440 --> 00:04:53.040 When you start combining these different 00:04:53.040 --> 00:04:55.440 topologies together, you create a hybrid 00:04:55.440 --> 00:04:58.000 network. A hybrid network is more than 00:04:58.000 --> 00:05:00.320 one of these topology types all working 00:05:00.320 --> 00:05:02.320 together. For example, you might have 00:05:02.320 --> 00:05:04.240 three remote sites all connecting 00:05:04.240 --> 00:05:06.320 devices together using a switched 00:05:06.320 --> 00:05:08.720 Ethernet or star network. And then you 00:05:08.720 --> 00:05:10.639 may be connecting those together over a 00:05:10.639 --> 00:05:14.880 wide area network that uses a ring topology. 00:05:14.880 --> 00:05:16.560 If you're using a wireless network, you 00:05:16.560 --> 00:05:18.560 may be communicating in a number of 00:05:18.560 --> 00:05:20.880 different ways. If you're using an access 00:05:20.880 --> 00:05:22.880 point, you're probably communicating over 00:05:22.880 --> 00:05:25.280 an infrastructure connection. This means 00:05:25.280 --> 00:05:27.039 that all of the devices on your network 00:05:27.039 --> 00:05:29.039 are communicating through an access 00:05:29.039 --> 00:05:31.199 point. This is probably the most common 00:05:31.199 --> 00:05:33.600 way to use wireless connections, but it's 00:05:33.600 --> 00:05:35.840 not the only way to communicate over a 00:05:35.840 --> 00:05:37.759 wireless link. If you just have two 00:05:37.759 --> 00:05:40.000 devices and there's no access point that 00:05:40.000 --> 00:05:42.240 you can use, you can connect directly 00:05:42.240 --> 00:05:45.280 from one device to another using ad hoc 00:05:45.280 --> 00:05:47.520 networking. You don't need an access 00:05:47.520 --> 00:05:49.520 point or any other type of wireless 00:05:49.520 --> 00:05:51.440 infrastructure. You simply have 00:05:51.440 --> 00:05:53.759 one device communicate directly to another 00:05:53.759 --> 00:05:56.639 device over this wireless connection. 00:05:56.639 --> 00:05:58.800 And if you've added Internet of Things 00:05:58.800 --> 00:06:01.120 devices, which are commonly wireless 00:06:01.120 --> 00:06:03.680 devices that control our lights, our door 00:06:03.680 --> 00:06:06.080 locks, or the air conditioning systems, 00:06:06.080 --> 00:06:08.000 then you're probably using a mesh 00:06:08.000 --> 00:06:10.400 network where all of these devices can 00:06:10.400 --> 00:06:12.639 communicate to all the other devices 00:06:12.639 --> 00:06:14.960 simultaneously to create an 00:06:14.960 --> 00:06:17.680 interconnected mesh of communication 00:06:17.680 --> 00:06:21.120 between all of these IoT devices. 00:06:21.120 --> 00:06:23.039 One of the advantages of these mesh 00:06:23.039 --> 00:06:25.840 wireless networks is that it allows many 00:06:25.840 --> 00:06:28.080 devices to communicate to each other, 00:06:28.080 --> 00:06:30.560 even if those devices are very far apart 00:06:30.560 --> 00:06:32.720 from each other. This also allows the 00:06:32.720 --> 00:06:35.440 mesh network to self-heal. So if you turn 00:06:35.440 --> 00:06:37.680 off one of those IoT devices, the 00:06:37.680 --> 00:06:40.000 remaining devices will self-heal and 00:06:40.000 --> 00:06:42.880 redesign themselves into a mesh network 00:06:42.880 --> 00:06:46.313 that will allow them to continue the communication.