1
00:00:00,540 --> 00:00:03,690
We first talked about data flows back in IPv4

2
00:00:03,690 --> 00:00:05,070
earlier in this section,

3
00:00:05,070 --> 00:00:08,070
but they changed a little bit here in IPv6.

4
00:00:08,070 --> 00:00:09,780
Now there's still three data flows,

5
00:00:09,780 --> 00:00:11,460
but one of them has changed.

6
00:00:11,460 --> 00:00:12,660
We still have unicast,

7
00:00:12,660 --> 00:00:14,130
we still have multicast,

8
00:00:14,130 --> 00:00:16,290
but now we've done away with broadcast

9
00:00:16,290 --> 00:00:19,320
and we've introduced something called Anycast instead.

10
00:00:19,320 --> 00:00:22,470
Now, unicast works just like it did in IPv4,

11
00:00:22,470 --> 00:00:24,600
but instead of using IPv4 addresses,

12
00:00:24,600 --> 00:00:27,000
we're going to use IPv6 addresses.

13
00:00:27,000 --> 00:00:28,830
Now in this case, let's say my server wants

14
00:00:28,830 --> 00:00:31,680
to send a message to both PC one and PC two.

15
00:00:31,680 --> 00:00:33,570
Well, it's going to have to send out two copies

16
00:00:33,570 --> 00:00:36,330
of the same message, one intended for each host

17
00:00:36,330 --> 00:00:38,670
because it's going from one server to one host,

18
00:00:38,670 --> 00:00:41,400
and back from one server to the other host.

19
00:00:41,400 --> 00:00:44,250
Notice the only difference here is their IP addresses.

20
00:00:44,250 --> 00:00:47,040
We're using A, B, C, D, colon colon one,

21
00:00:47,040 --> 00:00:49,560
and all this space in between those two colons

22
00:00:49,560 --> 00:00:50,850
will be filled up with zeros

23
00:00:50,850 --> 00:00:53,880
to get us back to 32 hexa decimal digits.

24
00:00:53,880 --> 00:00:55,350
Next we have multicast.

25
00:00:55,350 --> 00:00:57,990
And in multicast, just like IPv4,

26
00:00:57,990 --> 00:00:59,910
we use multicast groups.

27
00:00:59,910 --> 00:01:01,410
The server's going to address the message

28
00:01:01,410 --> 00:01:02,820
to the multicast group.

29
00:01:02,820 --> 00:01:05,910
In this case FF zero zero colon colon A.

30
00:01:05,910 --> 00:01:07,320
Remember, anything that starts

31
00:01:07,320 --> 00:01:09,690
with FF is a multicast address,

32
00:01:09,690 --> 00:01:11,460
and then it's going to get to the switch.

33
00:01:11,460 --> 00:01:12,750
And when it gets to the switch,

34
00:01:12,750 --> 00:01:14,940
it distributes the additional copies to each

35
00:01:14,940 --> 00:01:17,280
of the recipients in the multicast group number

36
00:01:17,280 --> 00:01:20,040
based on this number one that we're using.

37
00:01:20,040 --> 00:01:22,830
Data is going to travel from a single source, our server

38
00:01:22,830 --> 00:01:25,920
to multiple but specific destination devices.

39
00:01:25,920 --> 00:01:28,620
The third type of data flow we have is known as anycast.

40
00:01:28,620 --> 00:01:31,440
And this is the different one from IPv4.

41
00:01:31,440 --> 00:01:33,660
This is the one that's unique to IPv6

42
00:01:33,660 --> 00:01:36,660
because we did away with broadcast in IPv4.

43
00:01:36,660 --> 00:01:37,950
Now any cast is designed

44
00:01:37,950 --> 00:01:40,410
to let one host initiate an efficient updating

45
00:01:40,410 --> 00:01:43,200
of router tables for a group of other hosts.

46
00:01:43,200 --> 00:01:46,020
IPV six is able to determine which gateway the host

47
00:01:46,020 --> 00:01:48,930
is closest to and send the packets to that gateway

48
00:01:48,930 --> 00:01:51,480
as though it was a unicast communication.

49
00:01:51,480 --> 00:01:54,821
The gateway is then going to broadcast that through the anycast

50
00:01:54,821 --> 00:01:56,940
to any other host in the group until all

51
00:01:56,940 --> 00:01:58,710
of the routing tables are updated.

52
00:01:58,710 --> 00:02:00,390
Now what does this really look like?

53
00:02:00,390 --> 00:02:02,700
Well, let's take a look at this diagram on the screen.

54
00:02:02,700 --> 00:02:05,430
Here we have a server, and the router there is gone left,

55
00:02:05,430 --> 00:02:07,710
is going to send things out through anycast,

56
00:02:07,710 --> 00:02:09,900
it sends it out to the internet, and it's going to go

57
00:02:09,900 --> 00:02:12,690
to whatever the closest place is that has a router.

58
00:02:12,690 --> 00:02:14,160
Now, in this case, it's the router

59
00:02:14,160 --> 00:02:16,050
that's at the top of the screen here.

60
00:02:16,050 --> 00:02:18,510
Now once it gets to that router on top, it's then going to go

61
00:02:18,510 --> 00:02:20,850
to the DNS server and it'll reroute down

62
00:02:20,850 --> 00:02:22,560
through the routers onto the bottom

63
00:02:22,560 --> 00:02:23,940
and keep pushing that around,

64
00:02:23,940 --> 00:02:25,470
but the original router on the left side

65
00:02:25,470 --> 00:02:28,500
of the screen only had to send out that message once,

66
00:02:28,500 --> 00:02:29,460
and then from there,

67
00:02:29,460 --> 00:02:32,010
all of the other routers will figure out the best way

68
00:02:32,010 --> 00:02:34,170
to update everybody else's tables.

69
00:02:34,170 --> 00:02:36,630
That's the benefit of using something like anycast

70
00:02:36,630 --> 00:02:39,210
as opposed to using another way to update your routers.

71
00:02:39,210 --> 00:02:40,950
And this becomes very efficient

72
00:02:40,950 --> 00:02:43,353
and it's one of the benefits of using IPv6.

