0:00:02.050,0:00:04.540 The role of a switch[br]is to forward traffic 0:00:04.540,0:00:06.760 based on the[br]destination MAC address 0:00:06.760,0:00:08.960 inside of an ethernet frame. 0:00:08.960,0:00:12.910 This means the switch needs to[br]keep an ongoing and active list 0:00:12.910,0:00:15.730 of all of the devices[br]it happens to know about 0:00:15.730,0:00:18.850 based on the MAC address[br]of those devices. 0:00:18.850,0:00:22.240 The switch builds this list[br]by looking at inbound traffic 0:00:22.240,0:00:24.520 and examining the[br]source MAC address, 0:00:24.520,0:00:26.410 and tying that source MAC address 0:00:26.410,0:00:28.750 to a specific physical interface. 0:00:28.750,0:00:30.520 And for switches[br]that are configured 0:00:30.520,0:00:33.100 with spanning tree[br]protocol, or STP, 0:00:33.100,0:00:36.190 they're also responsible for[br]ensuring that a loop does not 0:00:36.190,0:00:38.880 occur on the switch network. 0:00:38.880,0:00:41.460 The process of sending traffic[br]through a switch network 0:00:41.460,0:00:43.780 is the same for every ethernet frame. 0:00:43.780,0:00:45.870 Let's take this[br]scenario where Sam, 0:00:45.870,0:00:48.210 and you can see the MAC[br]address for Sam's device, 0:00:48.210,0:00:51.210 is sending information[br]to the SGC server, 0:00:51.210,0:00:57.240 and you can see the SGC server's[br]MAC address is 1000:5555:5555. 0:00:57.240,0:00:59.880 We have a switch in the[br]middle, and all of our devices 0:00:59.880,0:01:04.200 are plugged into the switch,[br]including Sam and the SGC server. 0:01:04.200,0:01:06.930 Inside of the switch[br]is a MAC address table. 0:01:06.930,0:01:10.050 It lists all the MAC[br]addresses and the interfaces 0:01:10.050,0:01:12.090 where those addresses are connected. 0:01:12.090,0:01:15.570 When Sam sends traffic to the[br]switch with the destination MAC 0:01:15.570,0:01:20.160 address of[br]1000:5555:5555, the switch 0:01:20.160,0:01:22.410 looks up that address in its table, 0:01:22.410,0:01:26.280 and if one matches one of the[br]entries inside of that table, 0:01:26.280,0:01:29.820 it identifies the output[br]interface for that traffic 0:01:29.820,0:01:32.970 and sends it down that[br]interface to the server 0:01:32.970,0:01:35.130 that has that MAC address. 0:01:35.130,0:01:38.800 If you have multiple switches,[br]it's exactly the same process, 0:01:38.800,0:01:41.610 except it occurs twice,[br]once on the first switch 0:01:41.610,0:01:42.870 and once on the second. 0:01:42.870,0:01:45.300 You can see this is the[br]same configuration where 0:01:45.300,0:01:47.980 Sam is communicating[br]to the SGC server, 0:01:47.980,0:01:50.610 but there is a switch A[br]on one side of the network, 0:01:50.610,0:01:52.530 and a switch B on the other. 0:01:52.530,0:01:55.990 Switch A has a MAC address table[br]specific to the devices plugged 0:01:55.990,0:01:59.340 into Switch A, and Switch B[br]has a completely different 0:01:59.340,0:02:01.710 and unique, MAC address table. 0:02:01.710,0:02:04.950 Sam is going to send traffic[br]again to the SGC server. 0:02:04.950,0:02:10.800 It knows that it's sending this traffic [br]to MAC address 1000:5555:5555. 0:02:10.800,0:02:12.960 As that traffic hits Switch A, Switch A 0:02:12.960,0:02:15.030 refers to its own MAC address table 0:02:15.030,0:02:18.420 and knows that that particular[br]MAC address is located 0:02:18.420,0:02:22.480 on an interface that is[br]a gigabit 0/2 interface, 0:02:22.480,0:02:25.230 and so it sends that[br]traffic out that interface 0:02:25.230,0:02:26.580 to the next switch. 0:02:26.580,0:02:29.040 On that switch, the[br]same lookup process 0:02:29.040,0:02:31.800 occurs, where Switch B will[br]examine the destination 0:02:31.800,0:02:35.220 MAC address, determine that[br]that MAC address is associated 0:02:35.220,0:02:38.280 with the interface fast ethernet 0/5, 0:02:38.280,0:02:42.960 and sends that traffic down that[br]interface to the destination device. 0:02:42.960,0:02:45.350 You can see that building [br]that MAC address table 0:02:45.350,0:02:46.870 is extremely important. 0:02:46.870,0:02:48.900 If we didn't have the MAC [br]address table, the switch 0:02:48.900,0:02:50.970 would not know where to send that traffic. 0:02:50.970,0:02:53.040 In order to build[br]that table, the switch 0:02:53.040,0:02:55.770 is going to examine[br]all incoming traffic 0:02:55.770,0:02:58.590 and make a note of the[br]source MAC address. 0:02:58.590,0:03:00.750 It will then associate[br]that source MAC address 0:03:00.750,0:03:03.510 to a specific interface[br]on the switch. 0:03:03.510,0:03:06.720 So let's take a scenario where[br]we've just powered up a switch, 0:03:06.720,0:03:09.180 it has nothing in the[br]MAC address table, 0:03:09.180,0:03:11.730 and we're going to send[br]information from Sam's computer 0:03:11.730,0:03:13.350 to the SGC server. 0:03:13.350,0:03:15.660 Sam's going to send that[br]traffic to the switch, 0:03:15.660,0:03:18.930 the switch is going to examine[br]the source MAC address, 0:03:18.930,0:03:24.150 and in the case of Sam's[br]device, that's 1000:1111:1111. 0:03:24.150,0:03:26.970 It will then put that MAC[br]address into the MAC address 0:03:26.970,0:03:29.550 table, and it will[br]identify the interface 0:03:29.550,0:03:34.440 where that information was received.[br]In this case, interface F0/1. 0:03:34.440,0:03:38.020 That information is then[br]sent on to the SGC server, 0:03:38.020,0:03:42.000 and then when the SGC server[br]responds to that communication, 0:03:42.000,0:03:44.490 it has a different[br]source MAC address, 0:03:44.490,0:03:47.350 and the process is repeated.[br]Except in this case, 0:03:47.350,0:03:49.650 the switch identifies[br]that MAC address is 0:03:49.650,0:03:53.490 coming from fast ethernet 0/5. 0:03:53.490,0:03:56.580 In that previous example, we [br]were sending information to the SGC 0:03:56.580,0:03:59.610 server, but the SGC [br]server's MAC address 0:03:59.610,0:04:01.680 was not yet in the switch. 0:04:01.680,0:04:04.590 If the switch does not have[br]an entry for that MAC address 0:04:04.590,0:04:07.260 in the table, then it[br]will send that information 0:04:07.260,0:04:09.420 to everyone on the network. 0:04:09.420,0:04:12.000 For example, we'll take Sam[br]sending this information 0:04:12.000,0:04:13.380 to the SGC server. 0:04:13.380,0:04:15.540 You can see in this case,[br]the MAC address table 0:04:15.540,0:04:17.970 has nothing inside[br]of it at the moment. 0:04:17.970,0:04:20.730 The MAC address table will be[br]updated with the source MAC 0:04:20.730,0:04:23.790 address because Sam did send[br]that information to the switch, 0:04:23.790,0:04:27.810 and it did associate that[br]with fast ethernet 0/1, 0:04:27.810,0:04:29.550 But we're sending this information 0:04:29.550,0:04:32.370 to a destination MAC[br]address that's not currently 0:04:32.370,0:04:34.470 listed in the switch's table. 0:04:34.470,0:04:38.460 In that case, it's going to now[br]send that traffic to everybody 0:04:38.460,0:04:40.650 on the network, and[br]effectively flood 0:04:40.650,0:04:44.520 that traffic to all of the[br]other interfaces on that switch. 0:04:44.520,0:04:47.070 If you're familiar with[br]the operation of a hub, 0:04:47.070,0:04:49.830 then you'll notice that this[br]is very similar to the way 0:04:49.830,0:04:51.330 a hub works normally. 0:04:51.330,0:04:53.790 But this traffic being[br]sent to every device 0:04:53.790,0:04:56.520 ensures that at least[br]the destination will 0:04:56.520,0:04:58.470 receive this particular frame. 0:04:58.470,0:05:02.160 And in this example, you can see[br]that the SGC server did indeed 0:05:02.160,0:05:05.280 receive that frame, and[br]when the SGC server responds 0:05:05.280,0:05:08.280 back to Sam with a response,[br]the source MAC address 0:05:08.280,0:05:10.320 will be identified by the switch. 0:05:10.320,0:05:13.380 That information will be added[br]to the MAC address table, 0:05:13.380,0:05:15.830 and the switch will no longer [br]need to flood the traffic 0:05:15.830,0:05:20.010 across all interfaces if communication[br]is occurring between Sam 0:05:20.010,0:05:23.270 and the SGC server again. 0:05:23.270,0:05:26.750 On an IPv4 network,[br]devices are able to obtain 0:05:26.750,0:05:31.190 the MAC address of a remote[br]device using the ARP protocol. 0:05:31.190,0:05:34.430 ARP stands for [br]Address Resolution Protocol. 0:05:34.430,0:05:37.730 ARP will query the network[br]for a specific IP address, 0:05:37.730,0:05:41.660 and that IP address will respond[br]back with its MAC address. 0:05:41.660,0:05:44.450 Your local computer keeps[br]a cache of all of the MAC 0:05:44.450,0:05:46.340 addresses that it currently knows. 0:05:46.340,0:05:48.720 If you wanted to look at [br]the ARP address table on 0:05:48.720,0:05:52.850 your local machine, you [br]can use the command arp-a. 0:05:52.850,0:05:55.800 Let's run the arp-a[br]command on my machine. 0:05:55.800,0:06:00.930 You can see that I have a number[br]of local devices on the 10.1.10 network. 0:06:00.930,0:06:02.640 You can see them all listed here. 0:06:02.640,0:06:06.120 There's also some other[br]devices on my local network, 0:06:06.120,0:06:10.880 including some APIPA addresses[br]and some multicast addresses. 0:06:10.880,0:06:13.610 Let's say that I want to[br]communicate to a switch 0:06:13.610,0:06:15.410 that I have on my network. 0:06:15.410,0:06:19.460 That switch's IP[br]address is 10.1.10.210, 0:06:19.460,0:06:21.440 and you can see in[br]my ARP address table, 0:06:21.440,0:06:24.450 I don't currently have[br]that address in the list. 0:06:24.450,0:06:29.420 So I'm going to perform a ping,[br]and I'm gonna ping 10.1.10.210, 0:06:29.420,0:06:32.990 and I'll get some responses back[br]from that particular device. 0:06:32.990,0:06:37.025 If I now look at my ARP[br]address table with an arp-a, 0:06:37.025,0:06:41.990 you will see that I have a[br]new entry for 10.1.10.210, 0:06:41.990,0:06:46.670 and you'll see that I have a MAC[br]address associated with that IP address. 0:06:46.670,0:06:49.640 When I performed that ping,[br]the first thing that occurred 0:06:49.640,0:06:51.890 was an ARP request[br]made to the network 0:06:51.890,0:06:54.080 to try to find that[br]particular device, 0:06:54.080,0:06:56.330 and I received an[br]ARP response, which 0:06:56.330,0:07:00.050 then allowed me to send traffic[br]to that device directly. 0:07:00.050,0:07:03.170 I captured the ARP[br]communication using Wireshark, 0:07:03.170,0:07:06.680 which is a packet analyzer, and[br]you can download and install 0:07:06.680,0:07:09.920 Wireshark on your own[br]machine to see not only ARPs, 0:07:09.920,0:07:12.770 but all of the network[br]traffic on your system. 0:07:12.770,0:07:15.650 The first frame that I'm[br]sending is from my device, 0:07:15.650,0:07:18.170 and it's sending it[br]out as a broadcast, 0:07:18.170,0:07:22.280 and the ARP itself is[br]requesting the MAC address 0:07:22.280,0:07:25.880 for who has 10.1.10.210. 0:07:25.880,0:07:28.010 You can see the[br]details of the ARP 0:07:28.010,0:07:30.800 that are located further[br]down in the detail. 0:07:30.800,0:07:32.700 You can see the[br]sender MAC address, 0:07:32.700,0:07:34.160 which is my Apple computer. 0:07:34.160,0:07:38.870 You can see my local IP[br]address, which is 10.1.10.249. 0:07:38.870,0:07:41.000 You can see the[br]target MAC address, 0:07:41.000,0:07:43.700 right now we don't know what[br]the MAC address is of the target, 0:07:43.700,0:07:46.950 so it's all zeros, and you can[br]see that I'm requesting the MAC 0:07:46.950,0:07:52.820 address for the device that has[br]the IP address of 10.1.10.210. 0:07:52.820,0:07:55.730 We very quickly get a response[br]from this device, which 0:07:55.730,0:07:58.640 happens to be a Cisco[br]switch, and the response 0:07:58.640,0:08:01.910 from the MAC address is[br]from the Cisco MAC address 0:08:01.910,0:08:06.260 with the sender's IP address,[br]which is 10.1.10.210, 0:08:06.260,0:08:08.390 and the target is the response back 0:08:08.390,0:08:11.750 to my Apple computer[br]and my local IP address. 0:08:11.750,0:08:15.570 You can see in the response[br]that it filled in the sender MAC address, 0:08:15.570,0:08:19.400 so instead of being all zeros,[br]I see this long MAC address 0:08:19.400,0:08:21.440 associated with this IP. 0:08:21.440,0:08:23.900 And if you remember[br]the IP address and MAC 0:08:23.900,0:08:26.240 address in my[br]local ARP cache, it 0:08:26.240,0:08:31.780 matches both of those that were[br]received by this ARP response. 0:08:31.780,0:08:34.570 That ARP process is what[br]we use an IP version 0:08:34.570,0:08:37.270 4 to be able to[br]identify a MAC address, 0:08:37.270,0:08:40.360 but we don't have[br]broadcasts in IPv6. 0:08:40.360,0:08:43.630 There's also a different[br]process for IPv6 0:08:43.630,0:08:47.770 to identify the MAC addresses of[br]devices on your local network. 0:08:47.770,0:08:51.610 In IPv6, we use in NDP,[br]which is Neighbor Discovery 0:08:51.610,0:08:57.280 Protocol, using multicast,[br]specifically with ICMPv6. 0:08:57.280,0:09:00.460 This replaces the ARP function[br]that we would commonly 0:09:00.460,0:09:04.390 see in IPv4 with[br]this Neighbor MAC Discovery. 0:09:04.390,0:09:06.550 This can also be[br]used in conjunction 0:09:06.550,0:09:10.900 with SLAAC, which is Stateless[br]Address Autoconfiguration, 0:09:10.900,0:09:14.200 which allows the system to[br]automatically configure itself 0:09:14.200,0:09:18.250 with an IP address without[br]using a DHCP server. 0:09:18.250,0:09:20.080 Neighbor Discovery Protocol is also 0:09:20.080,0:09:24.610 used to identify any[br]duplicate addresses using DAD, 0:09:24.610,0:09:26.980 or Duplicate Address Detection. 0:09:26.980,0:09:29.170 If you wanted to see[br]the conversation that 0:09:29.170,0:09:32.650 takes place in IPv6,[br]instead of using ARP, 0:09:32.650,0:09:37.930 we send a neighbor solicitation,[br]or NS, on a multicast address, 0:09:37.930,0:09:43.570 and that is the IPv6 multicast that's [br]used for this neighbor solicitation frame. 0:09:43.570,0:09:46.870 The response is sent back from[br]the other side with a neighbor 0:09:46.870,0:09:49.780 advertisement, or NA, and that NA 0:09:49.780,0:09:52.960 includes the MAC address[br]of that local device. 0:09:52.960,0:09:56.030 Although the protocols and the[br]method is slightly different, 0:09:56.030,0:10:01.330 you can see that the process is very [br]similar to the one that occurs in IPv4. 0:10:01.330,0:10:04.300 Not only are we sending[br]data over ethernet networks, 0:10:04.300,0:10:07.180 we can also send power[br]over those networks 0:10:07.180,0:10:11.860 at the same time using[br]Power over Ethernet, or POE. 0:10:11.860,0:10:15.610 This allows us to connect[br]devices such as access points, 0:10:15.610,0:10:18.520 voiceover IP phones,[br]and other devices 0:10:18.520,0:10:21.370 by simply plugging in[br]an ethernet connection. 0:10:21.370,0:10:23.920 You don't have to then[br]plug in a separate power 0:10:23.920,0:10:25.570 connection for that device. 0:10:25.570,0:10:28.120 That power is coming[br]from either the switch, 0:10:28.120,0:10:31.240 or another device that's[br]connected into the network. 0:10:31.240,0:10:34.512 If it's coming from the switch, [br]we call that an Endspan, 0:10:34.512,0:10:37.910 or if it's coming from an injector,[br]like the one you see here, 0:10:37.910,0:10:41.020 which sits in the middle of an[br]existing ethernet connection, 0:10:41.020,0:10:44.080 we refer to that as a Midspan. 0:10:44.080,0:10:47.260 If your ethernet network[br]is a 10 or 100 megabit 0:10:47.260,0:10:51.010 per second connection, then you[br]have some extra wires inside 0:10:51.010,0:10:53.110 of that cable that you[br]could use for power. 0:10:53.110,0:10:56.290 We refer to that as Mode[br]B power over ethernet, 0:10:56.290,0:10:58.990 where you're sending[br]power on the spare pairs. 0:10:58.990,0:11:01.030 But if you're using[br]gigabit connections, 0:11:01.030,0:11:04.660 you're using all of those wires[br]for your gigabit ethernet data. 0:11:04.660,0:11:06.970 And in those cases,[br]we're using Mode A, 0:11:06.970,0:11:11.108 where we're sending power[br]and data over the same wire. 0:11:11.108,0:11:13.400 You'll find there are a[br]number of different power 0:11:13.400,0:11:15.730 over ethernet standards,[br]and these standards 0:11:15.730,0:11:18.460 are being added to and[br]changed all the time. 0:11:18.460,0:11:24.370 Two very common standards are[br]the IEEE 802.3af from 2003. 0:11:24.370,0:11:28.000 We refer to that as the[br]original POE standard, which 0:11:28.000,0:11:32.320 provides 15.4 watts of[br]direct current power, 0:11:32.320,0:11:35.230 with a maximum current[br]of 350 milliamps. 0:11:35.230,0:11:38.290 An update to that standard[br]is what we call POE+. 0:11:38.290,0:11:42.640 This was updated[br]with 802.3at in 2009. 0:11:42.640,0:11:48.040 This also has been incorporated[br]into the existing 802.3 ethernet standard. 0:11:48.040,0:11:50.920 This provides a bit more[br]power on the ethernet network, 0:11:50.920,0:11:57.100 25.5 watts of DC power, with a [br]maximum current of 600 milliamps. 0:11:57.100,0:11:59.490 There are other power[br]over ethernet standards, 0:11:59.490,0:12:01.430 and these are being[br]updated all the time, 0:12:01.430,0:12:04.990 so make sure you check with the[br]ethernet standards from IEEE 0:12:04.990,0:12:09.570 to know exactly what options[br]may be available for you.