1 1 00:00:00,590 --> 00:00:04,600 Welcome to Layer 2 of the OSI model, the data link layer. 2 2 00:00:04,600 --> 00:00:05,690 In the data link layer, 3 3 00:00:05,690 --> 00:00:08,340 we're going to package up the bits we got from Layer 1 4 4 00:00:08,340 --> 00:00:09,880 and put those into frames 5 5 00:00:09,880 --> 00:00:11,130 and then, we're going to take those frames 6 6 00:00:11,130 --> 00:00:13,430 and transmit them throughout the network 7 7 00:00:13,430 --> 00:00:16,260 while performing some error detection, correction, 8 8 00:00:16,260 --> 00:00:19,520 identifying unique network devices using MAC addresses, 9 9 00:00:19,520 --> 00:00:21,690 and we're going to provide some flow control. 10 10 00:00:21,690 --> 00:00:24,810 Now, a MAC address is a media access control address 11 11 00:00:24,810 --> 00:00:27,920 which is a means for identifying a device physically 12 12 00:00:27,920 --> 00:00:30,780 and allowing it to operate on a logical topology. 13 13 00:00:30,780 --> 00:00:33,150 So, when we started talking about physical topologies 14 14 00:00:33,150 --> 00:00:35,460 in the last lesson in the last layer, 15 15 00:00:35,460 --> 00:00:37,040 we dealt with things physically, 16 16 00:00:37,040 --> 00:00:40,270 but now, we have to deal with things on a logical level. 17 17 00:00:40,270 --> 00:00:42,680 These MAC addresses are incredibly important 18 18 00:00:42,680 --> 00:00:45,710 for dealing with switches and other Layer 2 devices. 19 19 00:00:45,710 --> 00:00:48,020 When it comes to identifying MAC addresses, 20 20 00:00:48,020 --> 00:00:49,880 every manufacturer of a network card 21 21 00:00:49,880 --> 00:00:53,440 assigns a unique 48-bit physical addressing system 22 22 00:00:53,440 --> 00:00:56,190 to every network interface card they produce. 23 23 00:00:56,190 --> 00:00:57,220 As you can see here, 24 24 00:00:57,220 --> 00:00:59,580 we have 12-digit hexadecimal numbers 25 25 00:00:59,580 --> 00:01:02,390 that are used to represent these MAC addresses. 26 26 00:01:02,390 --> 00:01:05,210 These MAC addresses are always written hexadecimally 27 27 00:01:05,210 --> 00:01:06,940 wherein each of the letters or numbers 28 28 00:01:06,940 --> 00:01:09,060 is considered four bits. 29 29 00:01:09,060 --> 00:01:12,750 The first 24 bits or the six letters as you can see here 30 30 00:01:12,750 --> 00:01:16,230 identifies the particular vendor who made that card. 31 31 00:01:16,230 --> 00:01:20,470 In our example, we have D2:51:F1 32 32 00:01:20,470 --> 00:01:22,250 and this is going to uniquely identify 33 33 00:01:22,250 --> 00:01:24,700 whichever person made this card. 34 34 00:01:24,700 --> 00:01:26,920 I like to think about this like a social security number 35 35 00:01:26,920 --> 00:01:28,200 in the United States. 36 36 00:01:28,200 --> 00:01:29,980 If you look at the first three digits, 37 37 00:01:29,980 --> 00:01:31,470 it's going to identify the state 38 38 00:01:31,470 --> 00:01:33,560 and the year that person was born in. 39 39 00:01:33,560 --> 00:01:36,767 For instance, if my social security number was 123456789, 40 40 00:01:38,870 --> 00:01:41,980 the first three digits might say when and where I was born. 41 41 00:01:41,980 --> 00:01:45,090 Maybe it was California in the year 1955 42 42 00:01:45,090 --> 00:01:49,720 and those other six digits are going to uniquely identify me. 43 43 00:01:49,720 --> 00:01:51,300 Well, this is the same thing that happens 44 44 00:01:51,300 --> 00:01:52,360 with a MAC address. 45 45 00:01:52,360 --> 00:01:55,440 The first half of the MAC address, the first six digits, 46 46 00:01:55,440 --> 00:01:57,050 is going to tell us who made it. 47 47 00:01:57,050 --> 00:02:00,660 Was it made by Apple, Dell, Raw Link, or whatever? 48 48 00:02:00,660 --> 00:02:02,310 The second half is going to represent 49 49 00:02:02,310 --> 00:02:04,740 the exact machine it belongs to. 50 50 00:02:04,740 --> 00:02:06,940 This is important for our logical topology 51 51 00:02:06,940 --> 00:02:08,680 because we can look at the MAC address 52 52 00:02:08,680 --> 00:02:11,540 and observe the flow of data going through our networks. 53 53 00:02:11,540 --> 00:02:13,270 And at this point, we don't really care 54 54 00:02:13,270 --> 00:02:15,550 how these devices are physically connected. 55 55 00:02:15,550 --> 00:02:18,410 The issue at that point is a Level 1 issue. 56 56 00:02:18,410 --> 00:02:19,530 But now at Layer 2, 57 57 00:02:19,530 --> 00:02:22,460 we care about who's turn it is to talk and transmit 58 58 00:02:22,460 --> 00:02:25,390 so other devices aren't talking over each other. 59 59 00:02:25,390 --> 00:02:27,250 For example, when I teach this course 60 60 00:02:27,250 --> 00:02:28,580 in a classroom environment, 61 61 00:02:28,580 --> 00:02:29,810 instead of all of the students 62 62 00:02:29,810 --> 00:02:31,540 shouting out their answers at once, 63 63 00:02:31,540 --> 00:02:33,750 we use the system of raising our hands. 64 64 00:02:33,750 --> 00:02:35,910 We wait for the teacher to call on one of the students 65 65 00:02:35,910 --> 00:02:38,430 and then, we can let them ask a question. 66 66 00:02:38,430 --> 00:02:40,540 This is how we control the information flow 67 67 00:02:40,540 --> 00:02:42,610 so that everyone can hear each other. 68 68 00:02:42,610 --> 00:02:45,130 In a network, we use electronic mechanisms 69 69 00:02:45,130 --> 00:02:46,700 to do this same thing. 70 70 00:02:46,700 --> 00:02:48,090 Now, logical link control 71 71 00:02:48,090 --> 00:02:49,870 is going to provide connection services 72 72 00:02:49,870 --> 00:02:52,030 and allow your recipients to acknowledge 73 73 00:02:52,030 --> 00:02:53,480 the messages have actually gotten 74 74 00:02:53,480 --> 00:02:55,040 where you thought they were going. 75 75 00:02:55,040 --> 00:02:57,100 So, for example, if I called up 76 76 00:02:57,100 --> 00:02:58,590 and I asked if you got my phone call, 77 77 00:02:58,590 --> 00:02:59,690 you could say yes 78 78 00:02:59,690 --> 00:03:01,400 and that would acknowledge the receipt of that 79 79 00:03:01,400 --> 00:03:03,780 and then, we can move on to the next message. 80 80 00:03:03,780 --> 00:03:06,270 Logical link control does this for our networks. 81 81 00:03:06,270 --> 00:03:07,220 And because of this, 82 82 00:03:07,220 --> 00:03:10,030 it's the most basic form of flow control. 83 83 00:03:10,030 --> 00:03:12,470 Essentially, it's going to limit the amount of data 84 84 00:03:12,470 --> 00:03:14,100 that a sender can send at once 85 85 00:03:14,100 --> 00:03:17,040 and allow the receiver to keep from being overwhelmed. 86 86 00:03:17,040 --> 00:03:19,170 So, if I go back to my classroom example, 87 87 00:03:19,170 --> 00:03:21,640 if I'm sitting there and I'm moving too quickly, 88 88 00:03:21,640 --> 00:03:23,277 a student might raise their hand and say, 89 89 00:03:23,277 --> 00:03:25,267 "Hey Jason, I don't understand this. 90 90 00:03:25,267 --> 00:03:27,190 "Can you slow down and repeat it?" 91 91 00:03:27,190 --> 00:03:28,440 In the case of this video, 92 92 00:03:28,440 --> 00:03:31,130 you can just pause or go back and watch that part again, 93 93 00:03:31,130 --> 00:03:32,270 but in a classroom, they can't 94 94 00:03:32,270 --> 00:03:34,570 so they may ask me to repeat it. 95 95 00:03:34,570 --> 00:03:37,260 Logical link control, it similarly does the same thing, 96 96 00:03:37,260 --> 00:03:39,350 allowing a device to make this request 97 97 00:03:39,350 --> 00:03:41,540 for either less information at a time 98 98 00:03:41,540 --> 00:03:43,910 or to replay that information. 99 99 00:03:43,910 --> 00:03:45,760 Logical link control also gives us 100 100 00:03:45,760 --> 00:03:47,750 some basic error control functions 101 101 00:03:47,750 --> 00:03:49,890 such as allowing the receiver to inform the sender 102 102 00:03:49,890 --> 00:03:51,970 if their data frame wasn't received 103 103 00:03:51,970 --> 00:03:53,860 or if it was received corrupted 104 104 00:03:53,860 --> 00:03:56,220 and it does this by using a checksum. 105 105 00:03:56,220 --> 00:03:57,670 Now, since everything it receives 106 106 00:03:57,670 --> 00:03:59,750 is just a series of ones and zeroes, 107 107 00:03:59,750 --> 00:04:01,590 the receiver is going to add all of these up 108 108 00:04:01,590 --> 00:04:04,320 and the last bit will either be even or odd. 109 109 00:04:04,320 --> 00:04:06,880 If it matches, they add them all up and they're even, 110 110 00:04:06,880 --> 00:04:08,770 then, it's going to assume that this was good 111 111 00:04:08,770 --> 00:04:11,760 if you have received a zero, meaning it was even. 112 112 00:04:11,760 --> 00:04:14,910 If the last bit was odd, meaning it was a one, 113 113 00:04:14,910 --> 00:04:16,070 and they added up all the numbers 114 114 00:04:16,070 --> 00:04:17,650 and they got an odd number, 115 115 00:04:17,650 --> 00:04:19,360 that means it was good, as well. 116 116 00:04:19,360 --> 00:04:21,940 But if not, they can figure that something was bad 117 117 00:04:21,940 --> 00:04:25,130 and then ask for a retransmission of the frame. 118 118 00:04:25,130 --> 00:04:27,940 Now, communication can be synchronized across Layer 2 119 119 00:04:27,940 --> 00:04:30,410 according to three different schemes. 120 120 00:04:30,410 --> 00:04:32,680 We have something known as isochronous mode 121 121 00:04:32,680 --> 00:04:35,380 which happens when the networks use a common reference clock 122 122 00:04:35,380 --> 00:04:36,750 similar to synchronous 123 123 00:04:36,750 --> 00:04:39,520 yet they also create time slots for transmissions, 124 124 00:04:39,520 --> 00:04:42,240 much like we did with time division multiplexing. 125 125 00:04:42,240 --> 00:04:44,850 This has less overhead than either of the other two modes 126 126 00:04:44,850 --> 00:04:47,310 because both devices know when they can communicate 127 127 00:04:47,310 --> 00:04:49,490 and for exactly how long. 128 128 00:04:49,490 --> 00:04:52,330 The second method we can use is known as synchronous method 129 129 00:04:52,330 --> 00:04:54,900 and this is much like we use back in Layer 1. 130 130 00:04:54,900 --> 00:04:57,610 It's going to involve devices using the same clock. 131 131 00:04:57,610 --> 00:05:00,030 But the reason it's different from isochronous 132 132 00:05:00,030 --> 00:05:01,240 is that this is going to allow us 133 133 00:05:01,240 --> 00:05:02,910 to have beginning and ending frames 134 134 00:05:02,910 --> 00:05:04,630 and special control characters 135 135 00:05:04,630 --> 00:05:06,290 to tell us when we're going to start 136 136 00:05:06,290 --> 00:05:09,410 and when we're going to end based on those beats. 137 137 00:05:09,410 --> 00:05:11,550 For example, if I use it in music, 138 138 00:05:11,550 --> 00:05:13,810 I have songs that have various time signatures, 139 139 00:05:13,810 --> 00:05:16,900 things like 3/4 or 4/4 timing. 140 140 00:05:16,900 --> 00:05:19,850 This tells us how many beats are in each measure. 141 141 00:05:19,850 --> 00:05:22,050 Our networks operate much the same way 142 142 00:05:22,050 --> 00:05:23,790 in that our devices can only communicate 143 143 00:05:23,790 --> 00:05:27,630 at frequencies specified by these particular clock cycles. 144 144 00:05:27,630 --> 00:05:29,790 Because of this, there isn't a lot of gap time 145 145 00:05:29,790 --> 00:05:31,690 that isn't already properly utilized 146 146 00:05:31,690 --> 00:05:34,920 and this becomes a major drawback for synchronized mode. 147 147 00:05:34,920 --> 00:05:37,310 And finally, of course, we have asynchronous 148 148 00:05:37,310 --> 00:05:39,220 which is going to allow each of our network devices 149 149 00:05:39,220 --> 00:05:41,200 to reference their own clock cycles 150 150 00:05:41,200 --> 00:05:43,480 and use their own start and stop bits. 151 151 00:05:43,480 --> 00:05:45,550 In this way, there's no real control 152 152 00:05:45,550 --> 00:05:47,810 over when the devices are allowed to communicate, though, 153 153 00:05:47,810 --> 00:05:50,270 and that becomes the major drawback here. 154 154 00:05:50,270 --> 00:05:51,990 Now, when we look at Layer 2 devices, 155 155 00:05:51,990 --> 00:05:54,320 we have things like network interface cards, 156 156 00:05:54,320 --> 00:05:56,120 bridges, and switches. 157 157 00:05:56,120 --> 00:05:58,670 In contrast to how a hub is a dumb machine 158 158 00:05:58,670 --> 00:06:00,520 that simply relies on a message coming in 159 159 00:06:00,520 --> 00:06:02,070 and repeating it back out, 160 160 00:06:02,070 --> 00:06:03,930 switches are smarter. 161 161 00:06:03,930 --> 00:06:05,540 They can actually use logic 162 162 00:06:05,540 --> 00:06:09,030 to learn which physical ports are attached to which devices 163 163 00:06:09,030 --> 00:06:10,660 based on their MAC addresses. 164 164 00:06:10,660 --> 00:06:11,660 And in this way, 165 165 00:06:11,660 --> 00:06:14,530 they can send data to specific devices in the network, 166 166 00:06:14,530 --> 00:06:16,320 allowing us to pick up and choose 167 167 00:06:16,320 --> 00:06:19,310 different lines of communication to go to different areas. 168 168 00:06:19,310 --> 00:06:21,860 Now, we'll talk all about how this works 169 169 00:06:21,860 --> 00:06:23,510 and how these switches do these, 170 170 00:06:23,510 --> 00:06:26,400 including things like CAM tables using the MAC addresses 171 171 00:06:26,400 --> 00:06:28,980 and how they're doing the switching across the network 172 172 00:06:28,980 --> 00:06:31,610 in later lessons and we'll go into depth in that 173 173 00:06:31,610 --> 00:06:33,280 because you will need to understand that 174 174 00:06:33,280 --> 00:06:35,400 to understand how networks really work. 175 175 00:06:35,400 --> 00:06:36,440 But for right now, 176 176 00:06:36,440 --> 00:06:40,300 just remember that switches, bridges, and MAC addresses 177 177 00:06:40,300 --> 00:06:43,790 are three great examples of things that operate at Layer 2, 178 178 00:06:43,790 --> 00:06:44,973 the data link layer.