1 00:00:00,000 --> 00:00:05,000 So here’s our topology which is very similar to the previous topology 2 00:00:05,000 --> 00:00:08,000 but notice the bridge has been replaced by a switch. 3 00:00:08,000 --> 00:00:13,000 Switches are once again layer 2 devices or data link layer devices 4 00:00:13,000 --> 00:00:17,000 and also have a MAC address table in the similar way to a bridge. 5 00:00:17,000 --> 00:00:23,000 The network topology is also a star topology where devices are cabled directly 6 00:00:23,000 --> 00:00:28,000 to ports on the switch, as an analogy, think of the switch as a bridge 7 00:00:28,000 --> 00:00:32,000 but it’s much more powerful and quicker. 8 00:00:32,000 --> 00:00:35,000 If you had a problem in a bridge environment 9 00:00:35,000 --> 00:00:39,000 and you replace the bridge with a switch, you would still have the same problems 10 00:00:39,000 --> 00:00:41,000 but they would occur a lot quicker. 11 00:00:41,000 --> 00:00:48,000 Bridge problem are not solved by switches. Switches simply increase the performance. 12 00:00:48,000 --> 00:00:51,000 So here’s our sample topology once again, 13 00:00:51,000 --> 00:00:55,000 but in this case we have replaced the bridge with the switch. 14 00:00:55,000 --> 00:00:58,000 How will traffic flow in this example? 15 00:00:58,000 --> 00:01:02,000 So we once again using easy to read MAC addresses ike A, B, C and D 16 00:01:02,000 --> 00:01:09,000 rather than full 48 bit MAC addresses and we're doing that to simplify these examples. 17 00:01:09,000 --> 00:01:14,000 So if A sends a frame to C and the frame arrived at the switch on port 1. 18 00:01:14,000 --> 00:01:16,000 What would the switch do with the frame? 19 00:01:16,000 --> 00:01:20,000 Now in this example let’s assume that the switch is just started up. 20 00:01:20,000 --> 00:01:25,000 So the MAC address table is empty, it hasn’t learnt where devices are on the topology. 21 00:01:25,000 --> 00:01:30,000 Now the switch just like a bridge will flood the frame out of all ports 22 00:01:30,000 --> 00:01:33,000 because it doesn’t know where C is 23 00:01:33,000 --> 00:01:38,000 when the frame arrives at the switch to an unknown destination MAC address 24 00:01:38,000 --> 00:01:44,000 that frame is flooded out of all ports except the port on which the frame arrived. 25 00:01:44,000 --> 00:01:50,000 However just like a bridge the switch doesn’t just flood the frame out of all ports 26 00:01:50,000 --> 00:01:53,000 but it also learns where devices are in the topology. 27 00:01:53,000 --> 00:01:56,000 So because the frames was received on port 1 28 00:01:56,000 --> 00:01:59,000 and the source MAC address in the frame is A 29 00:01:59,000 --> 00:02:03,000 that information is written to the MAC address table of the switch. 30 00:02:03,000 --> 00:02:08,000 The switch now knows that MAC address A can be found on port 1. 31 00:02:08,000 --> 00:02:14,000 When C replies to A, the frame would be received on port 3 of the switch. 32 00:02:14,000 --> 00:02:17,000 And the switch would update its MAC address table with that information 33 00:02:17,000 --> 00:02:23,000 in other words the switch knows that MAC address C can be found on port 3 34 00:02:23,000 --> 00:02:29,000 but in this case because it knows where the destination MAC address is 35 00:02:29,000 --> 00:02:35,000 in other words A, it will only send the traffic out of port 1 and that’s because A 36 00:02:35,000 --> 00:02:40,000 is found in the MAC address table as being available out of port 1. 37 00:02:40,000 --> 00:02:43,000 The switch doesn’t flood the frame out of all ports. 38 00:02:43,000 --> 00:02:49,000 So in the same way as a bridge all subsequent frames between A and C are forwarded 39 00:02:49,000 --> 00:02:53,000 only out of those 2 ports, when A sends another frame to C 40 00:02:53,000 --> 00:02:56,000 the frame is only sent out of port 3 41 00:02:56,000 --> 00:03:00,000 because the switch knows that MAC address C can be found on port 3. 42 00:03:00,000 --> 00:03:04,000 when C replies sending traffic to a destination MAC address of A. 43 00:03:04,000 --> 00:03:08,000 the switch only forwards that traffic out of port 1 because that is learnt 44 00:03:08,000 --> 00:03:11,000 that MAC address A can be found on port 1. 45 00:03:11,000 --> 00:03:17,000 So all traffic between these 2 devices will only flow between port 1 and port 3. 46 00:03:17,000 --> 00:03:24,000 Traffic is not sent out of port 2 or port 4 in a similar way to how a bridge operates. 47 00:03:24,000 --> 00:03:27,000 In the same way as a bridge 48 00:03:27,000 --> 00:03:31,000 each interface on the switch is a separate collision domain. 49 00:03:31,000 --> 00:03:33,000 So if a collision took place on this hub 50 00:03:33,000 --> 00:03:35,000 it would not affect other ports on the switch. 51 00:03:35,000 --> 00:03:39,000 Each port on a switch is a separate collision domain. 52 00:03:39,000 --> 00:03:44,000 They are therefore 4 collision domains in this topology. 53 00:03:44,000 --> 00:03:48,000 A hub once again is a single collision domain. 54 00:03:48,000 --> 00:03:50,000 So this interface is a single collision domain 55 00:03:50,000 --> 00:03:54,000 separated from the other interfaces on the switch. 56 00:03:54,000 --> 00:03:59,000 A switch however, will flood broadcast and multicast traffic by default. 57 00:03:59,000 --> 00:04:02,000 So this is a single broadcast domain. 58 00:04:02,000 --> 00:04:07,000 If A sends a broadcast that broadcast will be flooded out of all ports 59 00:04:07,000 --> 00:04:10,000 and will be received by all devices in the topology. 60 00:04:10,000 --> 00:04:13,000 This is very similar to the way bridges operated. 61 00:04:13,000 --> 00:04:18,000 You once again have the same issues in a switch environment 62 00:04:18,000 --> 00:04:20,000 as you would have in a bridge environment. 63 00:04:20,000 --> 00:04:26,000 But switches operate at much higher speeds and support a greater number of ports. 64 00:04:26,000 --> 00:04:30,000 So typically you wouldn’t have only 4 ports on the switch. 65 00:04:30,000 --> 00:04:36,000 But in this example we have 4 collision domains and the single broadcast domain. 66 00:04:36,000 --> 00:04:41,000 Now the reason why a broadcast is flooded out of all ports except 67 00:04:41,000 --> 00:04:50,000 the ingress port is not a broadcast address consist of 8 hexadecimal Fs at layer 2. 68 00:04:50,000 --> 00:04:57,000 So when a switch receives the frame with a destination address of 8 Fs it will flood 69 00:04:57,000 --> 00:05:04,000 that frame out of all ports because this address of 8 Fs at layer 2 indicates everyone. 70 00:05:04,000 --> 00:05:07,000 In other words the switch will flood this out of all ports 71 00:05:07,000 --> 00:05:10,000 except the port on which it received. 72 00:05:10,000 --> 00:05:16,000 So in this example, it was received from A and that frame is then flooded everywhere 73 00:05:16,000 --> 00:05:21,000 because of broadcast is supposed to go to everyone at layer 2. 74 00:05:21,000 --> 00:05:24,000 That’s what a broadcast is designed to do. 75 00:05:24,000 --> 00:05:29,000 Broadcast addresses also indicate all devices rather than 76 00:05:29,000 --> 00:05:34,000 a single device so the MAC address table is never populated with the broadcast address. 77 00:05:34,000 --> 00:05:38,000 This information is not written to the MAC address table 78 00:05:38,000 --> 00:05:41,000 as the Unicast MAC address would have been. 79 00:05:41,000 --> 00:05:44,000 Broadcast addresses are not associated with specific 80 00:05:44,000 --> 00:05:48,000 or an individual ports on which the broadcast is received 81 00:05:48,000 --> 00:05:53,000 it is always flooded out of all ports except the port on which it's received. 82 00:05:53,000 --> 00:05:58,000 Now as always there are exceptions and we'll talk more about those exceptions later. 83 00:05:58,000 --> 00:06:04,000 There are some major advantages to using switches over hubs and bridges. 84 00:06:04,000 --> 00:06:08,000 The first advantage is that, switches can support many ports 85 00:06:08,000 --> 00:06:11,000 and some switches can support 100 of ports. 86 00:06:11,000 --> 00:06:15,000 The second advantage is that switches can operate at wire speed. 87 00:06:15,000 --> 00:06:20,000 So as I've mentioned previously the switch will not slow frames down. 88 00:06:20,000 --> 00:06:25,000 The switch can physically move a frame from one port to another port 89 00:06:25,000 --> 00:06:27,000 without slowing the frame down. 90 00:06:27,000 --> 00:06:32,000 Some switches have backplanes that operate a terabits per second. 91 00:06:32,000 --> 00:06:37,000 In other words a very, very fast backplanes in comparison to interface speeds. 92 00:06:37,000 --> 00:06:41,000 So the back plain of the switch is operating at the much greater speed 93 00:06:41,000 --> 00:06:44,000 than the physical ports. So what does that mean? 94 00:06:44,000 --> 00:06:48,000 The switch can move traffic from 1 port to another port 95 00:06:48,000 --> 00:06:53,000 faster or quicker than it can received traffic on an interface or port. 96 00:06:53,000 --> 00:06:57,000 So traffic from A to D is not slowdown by the switch. 97 00:06:57,000 --> 00:07:01,000 Another major advantage of switches over hubs is that 98 00:07:01,000 --> 00:07:05,000 each device is directly connected to a switch port. 99 00:07:05,000 --> 00:07:11,000 So A is connected to port 1, B to port 2, C to port 3, D to port 4. 100 00:07:11,000 --> 00:07:16,000 Each device is individually cabled to port on the switch.