1
00:00:00,090 --> 00:00:05,430
So let's start with a simple topology to illustrate how spending tree works.

2
00:00:05,939 --> 00:00:09,630
Why would you require spending tree in a switched network?

3
00:00:10,440 --> 00:00:14,370
Since topology we have to stay connected to switch one.

4
00:00:14,730 --> 00:00:20,760
Switch one in turn is connected to switch two and switch two has host be connected to it.

5
00:00:21,270 --> 00:00:23,220
So very simple topology.

6
00:00:24,010 --> 00:00:30,070
Now, if the link went down between switch one and switch to host, I wouldn't be able to communicate

7
00:00:30,070 --> 00:00:32,500
with host B and vice versa.

8
00:00:33,170 --> 00:00:38,900
So you probably going to want to implement some kind of redundancy between those two switches by adding

9
00:00:38,900 --> 00:00:40,250
an additional link.

10
00:00:40,790 --> 00:00:46,980
So that's great because you now have a network redundancy in case one of the links goes down.

11
00:00:47,000 --> 00:00:51,590
However, that introduces problems in a switched environment.

12
00:00:52,040 --> 00:00:57,830
It's generally recommended in networks today that you implement some type of redundancy.

13
00:00:57,860 --> 00:01:04,760
So in this example, you have two links between your two switches, but that will introduce additional

14
00:01:04,760 --> 00:01:06,620
problems, which will now discuss.

15
00:01:07,400 --> 00:01:13,960
Let's assume for the moment that the switches have just booted up and they mac address tables or camp

16
00:01:13,970 --> 00:01:15,200
tables or empty.

17
00:01:15,990 --> 00:01:24,030
And to help explain this issue, let's add Mac address tables to the topology so that you can see how

18
00:01:24,030 --> 00:01:28,920
the MAC address tables are updated when traffic is sent from one host to another.

19
00:01:29,830 --> 00:01:34,120
So let's assume that in this topology the switches have just come up.

20
00:01:34,330 --> 00:01:40,950
In other words, they've been rebooted or powered up and the Mac address tables or camp tables are empty

21
00:01:40,960 --> 00:01:42,280
on the two switches.

22
00:01:43,080 --> 00:01:51,180
Now, in a sense a frame to be the destination address in the frame will be B and the source address

23
00:01:51,180 --> 00:01:52,110
will be A.

24
00:01:52,770 --> 00:01:55,290
So are you sending a frame to be?

25
00:01:55,290 --> 00:02:02,820
And when it arrives at switch one switch one will read the source Mac address of the frame and the switch

26
00:02:02,820 --> 00:02:05,010
will see that the source addresses a.

27
00:02:05,880 --> 00:02:11,610
The switch will update its Mac address table to state that a can be found on Port one.

28
00:02:12,250 --> 00:02:12,940
Mac address.

29
00:02:12,940 --> 00:02:19,090
B, however, is not in the Mac address table, so the switch will flood the frame out of all ports

30
00:02:19,090 --> 00:02:21,340
except on the port on which it arrived.

31
00:02:21,880 --> 00:02:25,780
So in this example, the frame will go to port two as well as port three.

32
00:02:26,140 --> 00:02:30,160
It does that because it doesn't know where Mac Address B is.

33
00:02:30,900 --> 00:02:33,610
Now, this is obviously a very simple typology.

34
00:02:33,630 --> 00:02:39,030
In this example, the frame is only being sent out of two ports of the switch.

35
00:02:39,240 --> 00:02:47,190
However, if the switch had many ports, let's say 96 ports, an incoming frame on one port could be

36
00:02:47,190 --> 00:02:51,450
replicated out of over 90 ports on that switch.

37
00:02:52,430 --> 00:02:56,810
That increases the amount of traffic sent in your network quite dramatically.

38
00:02:57,500 --> 00:03:02,630
So in this topology, what does switch to do with the frame received on Port one?

39
00:03:03,600 --> 00:03:07,470
The source address once again is a and the destination address is be.

40
00:03:08,120 --> 00:03:10,370
What will the switch do with the frame?

41
00:03:10,730 --> 00:03:16,400
Well, firstly, it's going to update its Mac address table to state that a can be found on Port one

42
00:03:16,790 --> 00:03:19,670
and then it's going to flood the frame out of all ports.

43
00:03:19,880 --> 00:03:23,330
So it'll flooded out of port two as well as Port three.

44
00:03:24,160 --> 00:03:28,570
So in this example, host B will receive the frame from host A.

45
00:03:29,110 --> 00:03:35,590
However, the switch also receives the frame on Port three, and this is where it gets a bit confusing.

46
00:03:36,220 --> 00:03:38,750
Where is a from the switchers point of view?

47
00:03:38,770 --> 00:03:41,350
Is that on Port one or is it on port three?

48
00:03:42,030 --> 00:03:48,420
So in this example, it's going to update its Mac address table to state that A is on Port three because

49
00:03:48,420 --> 00:03:55,530
the frame in our example arrived on Port three a later, then on Port one.

50
00:03:55,530 --> 00:04:01,740
So it's going to update the Mac address table entry to state that A is now available and Port three.

51
00:04:02,690 --> 00:04:06,290
The switch will also flood the frame out of all ports.

52
00:04:06,290 --> 00:04:10,160
So it's going to flood it out of Port one and out of port two.

53
00:04:10,610 --> 00:04:18,860
So Host B has now received the frame twice, once from the original frame that arrived on Port One sent

54
00:04:18,860 --> 00:04:23,690
to OSB and secondly for the frame that arrived on Port three.

55
00:04:24,200 --> 00:04:30,080
So this can get confusing for end devices because they receiving the same frame multiple times.

56
00:04:30,380 --> 00:04:33,110
The Mac address table is also changing.

57
00:04:33,620 --> 00:04:39,800
The first frame that arrived on Port one allowed the switch to update its Mac address table to state

58
00:04:39,800 --> 00:04:41,720
that a can be found on Port one.

59
00:04:42,170 --> 00:04:48,410
However, the frame that arrived on Port three now indicates to the switch that A can be found on Port

60
00:04:48,410 --> 00:04:49,160
three.

61
00:04:49,160 --> 00:04:54,890
So the switch needs to update its Mac address table to state that A can be now found on Port three.

62
00:04:55,520 --> 00:04:58,520
So this introduces instability in the Mac address table.

63
00:04:59,210 --> 00:05:06,620
So we have end devices receiving frames multiple times and we have Mac address instability because the

64
00:05:06,620 --> 00:05:12,140
switch thought that A was available on Port one, but now sees that it's available on Port three.

65
00:05:12,350 --> 00:05:13,760
But it gets worse.

66
00:05:14,630 --> 00:05:20,360
When the frame arrived on Port one, the switch updated its Mac address table to state that A can be

67
00:05:20,360 --> 00:05:21,530
found on Port One.

68
00:05:21,680 --> 00:05:27,620
But it also flooded the frame out of both Port two and Port three in this topology.

69
00:05:28,460 --> 00:05:34,700
The frame was received by Host B, but in addition the frame was sent back to switch one.

70
00:05:35,090 --> 00:05:42,230
So switch one has received as a frame that it sent to switch two and switch one and now updates its

71
00:05:42,230 --> 00:05:47,390
Mac address table to state that A is available on Port three.

72
00:05:48,440 --> 00:05:50,980
Now when to switch one received the frame.

73
00:05:50,990 --> 00:05:57,250
It not only updates its Mac address table, but it also floods the frame out of all ports except to

74
00:05:57,260 --> 00:05:58,790
the port in which it arrived.

75
00:05:59,150 --> 00:06:05,330
So the frame arrived on Port three and is flooded out of Port one and out of port two.

76
00:06:05,750 --> 00:06:12,110
So this gets confusing for Host A because it's receiving the frame that it's originally sent.

77
00:06:12,590 --> 00:06:20,420
But not only is a receiving the frame that it sent switch one is also sending the same frame back to

78
00:06:20,420 --> 00:06:21,470
switch two.

79
00:06:21,950 --> 00:06:24,260
And what is switch two going to do with the frame?

80
00:06:24,500 --> 00:06:25,880
It's going to flood it.

81
00:06:25,880 --> 00:06:28,310
So it's going to send a copy to host B.

82
00:06:28,670 --> 00:06:35,450
Host P is now received the same frame three times, but it'll also send the frame back to switch one.

83
00:06:36,350 --> 00:06:41,390
As well as updating its Mac address table to now state that A is on Port one.

84
00:06:41,960 --> 00:06:46,760
So originally when it received the first frame, it thought that A was on port one.

85
00:06:47,240 --> 00:06:53,750
Then when it received the frame and port three, it thought that A was in port three and now it thinks

86
00:06:53,750 --> 00:06:55,370
that A is on port one.

87
00:06:55,640 --> 00:07:00,470
So we've got a lot of Mac address instability in the MAC address table.

88
00:07:01,100 --> 00:07:04,460
Post B's receiving the same frame multiple times.

89
00:07:04,730 --> 00:07:11,420
But the biggest issue here is that the frame gets sent back to switch one, gets flooded again, gets

90
00:07:11,420 --> 00:07:16,580
sent back to switch to, and this process continues over and over again.

91
00:07:17,180 --> 00:07:24,920
We have a loop in this topology with the frame being duplicated and sent round and round and round between

92
00:07:24,920 --> 00:07:26,150
these two switches.