1
00:00:02,703 --> 00:00:09,087
[music]

2
00:00:09,119 --> 00:00:11,912
Okay, so up until
this point in time, we've

3
00:00:11,912 --> 00:00:17,951
had this concept where some
device was running TCP/IP,

4
00:00:17,951 --> 00:00:21,421
the protocol stack, and part
of TCP/IP meant that that

5
00:00:21,421 --> 00:00:25,158
device had to have a 32-bit
number assigned to it - an

6
00:00:25,158 --> 00:00:28,862
IP address. And how that
device looked at that 32-bit

7
00:00:28,862 --> 00:00:32,399
number was, was it a really
old style device, where it

8
00:00:32,399 --> 00:00:35,903
just looked at the first
eight bits and that was it?

9
00:00:35,903 --> 00:00:38,906
Or, was it a newer device
that used classful

10
00:00:38,906 --> 00:00:40,207
addressing, where it said,
Well, I'm going to look at

11
00:00:40,207 --> 00:00:43,210
the first two or three bits
there and determine if my

12
00:00:43,210 --> 00:00:46,580
address is A, B, or C. And
then based on that, I'll

13
00:00:46,580 --> 00:00:48,582
figure out where the
dividing line is between my

14
00:00:48,582 --> 00:00:54,421
network bits and everything
left over - my host bits.

15
00:00:54,421 --> 00:00:57,691
Well, even though this
classful addressing did end

16
00:00:57,691 --> 00:01:00,394
up giving us thousands and
thousands of more available

17
00:01:00,394 --> 00:01:05,132
networks. Even that very
soon became apparent that it

18
00:01:05,132 --> 00:01:07,901
was not going to be enough.
It wasn't going to be

19
00:01:07,901 --> 00:01:14,474
enough. So it soon became
apparent that "Hey look," you

20
00:01:14,474 --> 00:01:18,312
know, if a company goes to the
Regional Internet Registry or a

21
00:01:18,312 --> 00:01:21,248
service provider and that
company says, "Look I'm

22
00:01:21,248 --> 00:01:24,818
really big, and I'm going to
need a thousand networks. You

23
00:01:24,818 --> 00:01:29,356
know, I've got 800 different
departments of employees. In

24
00:01:29,356 --> 00:01:32,092
addition to that, I want all
my printers on one network.

25
00:01:32,092 --> 00:01:34,995
I want all my voice phones
on another. I want all my

26
00:01:34,995 --> 00:01:36,897
fax machines on other one,"
so I actually need a

27
00:01:36,897 --> 00:01:41,034
thousand broadcast domains,
a thousand networks. Well if

28
00:01:41,034 --> 00:01:42,769
the Regional Internet
Registry or the service

29
00:01:42,769 --> 00:01:46,907
provider gave them one
thousand public classful

30
00:01:46,907 --> 00:01:51,411
networks, at that rate those
classful networks were

31
00:01:51,411 --> 00:01:57,951
soon going to be gone. So
this is where in 1985 the

32
00:01:57,951 --> 00:02:03,090
concept of subnets came into
play. So originally we have

33
00:02:03,090 --> 00:02:06,727
this. We have a building,
four floors in the building.

34
00:02:06,727 --> 00:02:09,563
Let's just assume that each
floor requires its own

35
00:02:09,563 --> 00:02:12,799
unique network. So back in
the classful days, we said,

36
00:02:12,799 --> 00:02:15,435
Okay, we'll give each one of
those floors its own unique

37
00:02:15,435 --> 00:02:19,506
classful network - the 20,
the 21, the 22, and the 23

38
00:02:19,506 --> 00:02:23,644
networks. But that wasn't
very scalable. Even that was

39
00:02:23,644 --> 00:02:25,879
going to run out. So
instead, they said, Okay,

40
00:02:25,879 --> 00:02:28,115
here's what we're going to
do. We will give you a

41
00:02:28,115 --> 00:02:32,786
single classful network, the
20 network, the class A 20

42
00:02:32,786 --> 00:02:36,623
network, and you will now
divide that one network into

43
00:02:36,623 --> 00:02:43,330
multiple smaller pieces -
what we call subnetworks.

44
00:02:43,330 --> 00:02:47,134
Now in a classful scheme, if
all those devices in that

45
00:02:47,134 --> 00:02:49,236
building in the right
were doing classful IP

46
00:02:49,236 --> 00:02:53,240
addressing, they would say,
We're all in one network. We

47
00:02:53,240 --> 00:02:57,611
looked at the first byte--
and the number 20 in binary

48
00:02:57,611 --> 00:03:01,615
starts out with zero, right?
The 128 bit is off. So

49
00:03:01,615 --> 00:03:04,017
they'd say, Oh The 120 bit
is off, which means that

50
00:03:04,017 --> 00:03:07,754
this is a Class A address
which means the first byte

51
00:03:07,754 --> 00:03:10,123
tells us our network. Oh,
the first byte's number 20.

52
00:03:10,123 --> 00:03:13,427
Okay, anything beginning
with the number 20 is in my

53
00:03:13,427 --> 00:03:17,264
network, belongs to my
group. But clearly, that's

54
00:03:17,264 --> 00:03:20,000
not the case here. Here,
we've got four different

55
00:03:20,000 --> 00:03:23,937
broadcast domains. So with
subnetting we're saying,

56
00:03:23,937 --> 00:03:28,208
"Hey look, don't worry about
your class. Your class is

57
00:03:28,208 --> 00:03:31,244
kind of irrelevant. Don't
look at the first bit or two

58
00:03:31,244 --> 00:03:35,982
bits or three bits. Instead,
look at something else."

59
00:03:35,982 --> 00:03:37,284
So now you're not going to
look at your class

60
00:03:37,284 --> 00:03:39,553
identifier anymore. Now, I'm
going to give you a new

61
00:03:39,553 --> 00:03:42,956
piece of information that
tells you, of that 32 bit

62
00:03:42,956 --> 00:03:45,725
number where to draw that
dividing line between the

63
00:03:45,725 --> 00:03:49,529
network portion and the
host. This is where a subnet

64
00:03:49,529 --> 00:03:53,433
mask comes into play. Can't
do subnetting without a

65
00:03:53,433 --> 00:04:01,208
subnet mask. So we know, as
a review that when you get a

66
00:04:01,208 --> 00:04:04,211
network from an ISP you're
going to have the network

67
00:04:04,211 --> 00:04:10,183
part and the host part. Now,
from the ISP's perspective,

68
00:04:10,183 --> 00:04:13,687
the service provider, they
don't care what you do with

69
00:04:13,687 --> 00:04:16,656
your host bits. They're
going to assume that you're

70
00:04:16,656 --> 00:04:18,658
just going to assign those
host bits in unique

71
00:04:18,658 --> 00:04:21,261
combinations to your
laptops, in your PCs, in

72
00:04:21,261 --> 00:04:24,564
your servers, and so on and
so forth. But what they do

73
00:04:24,564 --> 00:04:28,301
care about is, they say,
"Look! We just gave you,

74
00:04:28,301 --> 00:04:32,506
let's say, an 8-bit network.
We gave you the number 20."

75
00:04:32,506 --> 00:04:36,710
So the ISP says, "That is
now registered to you. No

76
00:04:36,710 --> 00:04:40,247
other company in the entire
world can have that 8-bit

77
00:04:40,247 --> 00:04:43,650
pattern on their network.
That belongs to you, and we

78
00:04:43,650 --> 00:04:46,253
will make sure we advertise,
while the other ISPs

79
00:04:46,253 --> 00:04:49,790
throughout the entire world,
that the 20 network belongs

80
00:04:49,790 --> 00:04:53,393
right here. So in Germany,
Mongolia, China, if you guys

81
00:04:53,393 --> 00:04:55,262
ever need to send a packet
to something that's starting

82
00:04:55,262 --> 00:04:59,232
with 20, send it here, and
we'll get it to this company,

83
00:04:59,232 --> 00:05:02,169
which is, you know, our
customer." So it's very

84
00:05:02,169 --> 00:05:05,605
important that the network they
gave you, the network bits, you

85
00:05:05,605 --> 00:05:07,407
don't mess around with
those. So if you start

86
00:05:07,407 --> 00:05:09,809
changing your network bits
now, you're infringing on

87
00:05:09,809 --> 00:05:16,616
some other company. So let's
say for example, an ISP

88
00:05:16,616 --> 00:05:22,789
leases you this network,
129.1.0.0. Now if we're

89
00:05:22,789 --> 00:05:25,425
dealing with a classful
architecture, classful IP

90
00:05:25,425 --> 00:05:27,360
addressing, you would look
at that and you'd say, "Okay

91
00:05:27,360 --> 00:05:31,064
let's see, 129. Let's
convert that into binary."

92
00:05:31,064 --> 00:05:38,772
128 bit is on, 64 bit is
off. Boom, I'm done. That is

93
00:05:38,772 --> 00:05:43,443
a class B address which
tells me that the first half

94
00:05:43,443 --> 00:05:47,714
of that address, the first 2
bytes, 129.1, is the

95
00:05:47,714 --> 00:05:51,184
network. So this is what the
ISP is thinking. The ISP is

96
00:05:51,184 --> 00:05:53,153
thinking, "Okay, Keith's
company over here that we

97
00:05:53,153 --> 00:05:59,993
just leased this network to,
they own 129.1." Nobody else

98
00:05:59,993 --> 00:06:06,333
in the whole world owns 129.1.
So there's what it looks

99
00:06:06,333 --> 00:06:11,538
like in binary. So the ISP
says, Don't change those

100
00:06:11,538 --> 00:06:13,640
bits in red. If you do,
we're going to be very

101
00:06:13,640 --> 00:06:16,810
unhappy with you. But you
can do whatever you want

102
00:06:16,810 --> 00:06:22,882
with those last 16 bits.
Those are host bits. So the

103
00:06:22,882 --> 00:06:25,418
ISP's assumption - the
service provider's

104
00:06:25,418 --> 00:06:29,990
assumption - is that I have
one network, one broadcast

105
00:06:29,990 --> 00:06:33,226
domain, that uses that
16-bit pattern to identify

106
00:06:33,226 --> 00:06:37,897
itself, 129.1 but that
raises the question: what if

107
00:06:37,897 --> 00:06:41,134
I have other broadcast
domains? What if I need two

108
00:06:41,134 --> 00:06:44,204
networks or three or four?
Well certainly, the ISP would be

109
00:06:44,204 --> 00:06:48,108
more than happy to take my
money to buy another

110
00:06:48,108 --> 00:06:51,011
network, and another, and
another. The ISP will say,

111
00:06:51,011 --> 00:06:56,249
Oh sure, yeah, we'd love to
sell you 129.2 and 129.3 and

112
00:06:56,249 --> 00:06:59,352
129.30. Come to us with your
money. We'll give you as

113
00:06:59,352 --> 00:07:04,691
many networks as you need. So
I say, "Well, you know, if I'm

114
00:07:04,691 --> 00:07:10,363
spending $100 a month on that
one network, I don't necessarily

115
00:07:10,363 --> 00:07:14,334
have another $2,000 a month to
spend to get a couple

116
00:07:14,334 --> 00:07:17,937
dozen more networks."
So remember, the ISP says,

117
00:07:17,937 --> 00:07:21,141
"Look, as long as you don't
change those red bits,

118
00:07:21,141 --> 00:07:23,877
that's all we care about."
So I say, "Well, wait a

119
00:07:23,877 --> 00:07:27,514
second. We know that all the
devices in the same

120
00:07:27,514 --> 00:07:31,718
broadcast domain all have to
agree on what the name of

121
00:07:31,718 --> 00:07:35,255
their group is, in binary."
What the binary pattern is

122
00:07:35,255 --> 00:07:37,724
for their broadcast domain.
What their binary network

123
00:07:37,724 --> 00:07:44,964
is. I know I can't change
the first 16 bits. Well, why

124
00:07:44,964 --> 00:07:49,602
don't I do this. Why don't I
give them a unique 18 bit

125
00:07:49,602 --> 00:07:53,873
pattern like this? I'll keep
the first 16 bits the same

126
00:07:53,873 --> 00:07:55,275
because I can't change that.
That's what my service

127
00:07:55,275 --> 00:07:58,545
provider gave me. But, of
the last 16 bits, I'm going

128
00:07:58,545 --> 00:08:02,949
to take those two bits there in
green. Instead of telling my

129
00:08:02,949 --> 00:08:05,585
laptops, Hey, look at those
two bits as part of your

130
00:08:05,585 --> 00:08:08,121
host pattern. I'm going to
say, No, I don't want you to

131
00:08:08,121 --> 00:08:12,058
do that. All laptops on that
segment on the left there,

132
00:08:12,058 --> 00:08:16,363
of your 32-bit number, for
all of you, the first 18

133
00:08:16,363 --> 00:08:20,433
bits represent your network,
represent your group. So

134
00:08:20,433 --> 00:08:22,168
basically what I've done is
I've taken these two green

135
00:08:22,168 --> 00:08:27,006
bits that used to be host
bits and I've repurposed

136
00:08:27,006 --> 00:08:30,643
them. I've converted them
into networking bits. I've

137
00:08:30,643 --> 00:08:34,581
added them to the existing
16 networking bits I already

138
00:08:34,581 --> 00:08:38,918
had that I couldn't change.
And because the ISP doesn't

139
00:08:38,918 --> 00:08:43,423
care what I do with these 16
bits on the end, I can do

140
00:08:43,423 --> 00:08:46,259
whatever I want with that.
So I'll say, Okay, all you

141
00:08:46,259 --> 00:08:51,297
guys here on the left, this
is your 18-bit network. And

142
00:08:51,297 --> 00:08:54,567
all you guys on the right,
I'll use a different 18-bit

143
00:08:54,567 --> 00:08:59,272
network for you. Still
keeping the 16 bits I got

144
00:08:59,272 --> 00:09:02,942
from my ISP consistent, but
now this is what's called

145
00:09:02,942 --> 00:09:06,679
subnetting. Subnetting is
taking one network and

146
00:09:06,679 --> 00:09:10,850
dividing it into two or
three or four or five or

147
00:09:10,850 --> 00:09:14,721
hundreds of subnetworks,
keeping the original bit

148
00:09:14,721 --> 00:09:18,224
pattern of your original
network consistent across

149
00:09:18,224 --> 00:09:22,429
all of them. So in this
particular example here of

150
00:09:22,429 --> 00:09:27,100
my 32-bit number now, we can
classify those 32 bits into

151
00:09:27,100 --> 00:09:30,637
one of three buckets. We've
got our original networking

152
00:09:30,637 --> 00:09:33,173
bits which are the ones in
red. Those are the ones I

153
00:09:33,173 --> 00:09:35,775
can't change that I've
bought and paid for from my

154
00:09:35,775 --> 00:09:39,379
regional Internet registry
or from my service provider.

155
00:09:39,379 --> 00:09:41,748
And then we have some host bits
which are no longer host

156
00:09:41,748 --> 00:09:44,984
bits which I've now
repurposed or reused as

157
00:09:44,984 --> 00:09:48,788
networking bits. I'm now
telling my laptops, Add that

158
00:09:48,788 --> 00:09:51,357
to your existing network
section. That's part of your

159
00:09:51,357 --> 00:09:55,662
group name and those are
called subnet bits. And I've

160
00:09:55,662 --> 00:10:02,702
got my host bits on the end.
So here's the question.

161
00:10:02,702 --> 00:10:06,005
Original IP, the devices
just assume back in the

162
00:10:06,005 --> 00:10:10,477
1960s the first 8 bits
represented their group,

163
00:10:10,477 --> 00:10:13,880
represented their network.
Move forward to what was it?

164
00:10:13,880 --> 00:10:17,183
1985, I think it was, or
1981, it was in the slides

165
00:10:17,183 --> 00:10:20,220
back, to Classful
addressing. Now a device

166
00:10:20,220 --> 00:10:23,389
take a look at the first few
bits of the IP address and

167
00:10:23,389 --> 00:10:27,093
said, what class am I? A, B,
or C? That'll tell me where

168
00:10:27,093 --> 00:10:31,231
my network ends and where my
host portion begins. Now

169
00:10:31,231 --> 00:10:33,900
we're saying as a third
alternative, don't use

170
00:10:33,900 --> 00:10:37,170
either one of those methods.
I'm going to arbitrarily

171
00:10:37,170 --> 00:10:40,173
tell you that of your 32 bit
number, the first 18 bits,

172
00:10:40,173 --> 00:10:45,311
or the first 20 bits, or the
first 28 bits represent your

173
00:10:45,311 --> 00:10:48,715
group. But how do I give it
that information? How do I

174
00:10:48,715 --> 00:10:51,985
tell it of those 32 bits
where that dividing line can

175
00:10:51,985 --> 00:10:55,889
arbitrarily fall? And that's
where subnet mask came into

176
00:10:55,889 --> 00:11:03,863
place. So here we see prior
to subnet masks just the

177
00:11:03,863 --> 00:11:07,200
first few bits identified
how much of the address

178
00:11:07,200 --> 00:11:09,669
belong to your network, and
then you were done, that was

179
00:11:09,669 --> 00:11:12,238
it, you feel into an A, B,
or C, and that's all you

180
00:11:12,238 --> 00:11:17,710
had. So in this particular
case, the PC on the left

181
00:11:17,710 --> 00:11:19,812
says, "Okay, I've now
identified my network

182
00:11:19,812 --> 00:11:26,352
numbers being 129.1. So if I
need to create a packet, if

183
00:11:26,352 --> 00:11:31,024
the destination of that
packet is also 129.1, great!

184
00:11:31,024 --> 00:11:33,459
I know they're on the same
wire as me. I know that when

185
00:11:33,459 --> 00:11:35,862
I put my electrical energy
on this wire that represents

186
00:11:35,862 --> 00:11:38,464
1s and 0s, they will see it
because they're in the same

187
00:11:38,464 --> 00:11:41,868
broadcast domain as I am.
But if I need to create a

188
00:11:41,868 --> 00:11:44,137
packet going to a
destination that's not

189
00:11:44,137 --> 00:11:49,475
129.1, that's anything else,
they are not in my broadcast

190
00:11:49,475 --> 00:11:52,345
domain, which means, I need
to send it to my router to

191
00:11:52,345 --> 00:11:57,584
get it to wherever it needs to
go." So he says, "Okay, all

192
00:11:57,584 --> 00:12:04,490
of my particular network, my
network is 129.1.0.0 and if

193
00:12:04,490 --> 00:12:08,761
I ever need to send a
broadcast well, my group is

194
00:12:08,761 --> 00:12:14,701
129.1 and I'll set all the host
bits to 1s. So the broadcast

195
00:12:14,701 --> 00:12:20,139
address for this particular
network is 129.1.255.255.

196
00:12:24,577 --> 00:12:30,116
Okay, now we go into subnet
masks. So now on my laptop or

197
00:12:30,116 --> 00:12:33,519
my PC or my tablet, I have to
provide it two pieces of

198
00:12:33,519 --> 00:12:37,657
information: the IP address
and right next to that

199
00:12:37,657 --> 00:12:41,160
another 32-bit number in
binary which is called a

200
00:12:41,160 --> 00:12:44,097
subnet mask. And the two are
used to compare against

201
00:12:44,097 --> 00:12:47,000
each other. Now here I've put
them one on top of each

202
00:12:47,000 --> 00:12:49,902
other to make it real clear.
I said, Okay, here's your

203
00:12:49,902 --> 00:12:53,206
32-bit number which is your
IP address and just below

204
00:12:53,206 --> 00:12:58,344
that is your subnet mask. And
the way that your laptop or

205
00:12:58,344 --> 00:13:01,714
smartphone or tablet looks
at this, it says, Okay, I'm

206
00:13:01,714 --> 00:13:05,218
going to do a one-for-one
comparison. For every bit in

207
00:13:05,218 --> 00:13:09,322
my subnet mask that's a 1,
that means the corresponding

208
00:13:09,322 --> 00:13:12,558
bit in the IP address
belongs to my network,

209
00:13:12,558 --> 00:13:16,329
belongs to my group, my
broadcast domain. And for a

210
00:13:16,329 --> 00:13:19,866
bit in my subnet mask that's
a 0, that means the

211
00:13:19,866 --> 00:13:22,702
corresponding bit in my IP
address belongs to my own

212
00:13:22,702 --> 00:13:27,306
unique host identifier. So
now, here with this subnet

213
00:13:27,306 --> 00:13:30,443
mask, I'm telling it, Okay,
don't worry about the class

214
00:13:30,443 --> 00:13:33,579
anymore. Forget about the
fact that it says 1 0 at the

215
00:13:33,579 --> 00:13:36,182
beginning. That's
irrelevant. Instead, the

216
00:13:36,182 --> 00:13:40,353
subnet mask says, of your
32-bit number, the first 24

217
00:13:40,353 --> 00:13:46,993
bits represent your specific
group, your broadcast

218
00:13:46,993 --> 00:13:52,365
domain. So now the laptop
says, Okay, based on that,

219
00:13:52,365 --> 00:13:55,101
anybody else I want to talk
to who's in my broadcast

220
00:13:55,101 --> 00:13:57,737
domain, who's on my wire,
should also have the exact

221
00:13:57,737 --> 00:14:07,013
same 24-bit pattern as me,
which is 129.1.1. And so now

222
00:14:07,013 --> 00:14:08,881
if I need to send a packet
to anybody that does not

223
00:14:08,881 --> 00:14:12,585
match those first 24 bits,
that means they're in a

224
00:14:12,585 --> 00:14:15,588
different broadcast domain,
and I'll need to use my

225
00:14:15,588 --> 00:14:20,760
default gateway. So now my
network has changed a little

226
00:14:20,760 --> 00:14:25,164
bit. My network is still the
same 129.1.0. Remember

227
00:14:25,164 --> 00:14:27,867
previously he said, "Oh my
network is only the first

228
00:14:27,867 --> 00:14:33,172
half of the address. I'm
class B so it's 129.1." Now

229
00:14:33,172 --> 00:14:34,907
he's saying, "Oh the Subnet
mask is telling me it's

230
00:14:34,907 --> 00:14:39,112
129.1.0." I got to include
that 0 as part of my network

231
00:14:39,112 --> 00:14:43,149
address, and if I want to
send a broadcast, that would

232
00:14:43,149 --> 00:14:53,459
be 129.1.1.255. So this is
the purpose of the Subnet

233
00:14:53,459 --> 00:14:56,662
mask. It's a comparison tool
to help the device with the

234
00:14:56,662 --> 00:15:00,366
IP address to know
specifically of those 32

235
00:15:00,366 --> 00:15:03,836
bits where the dividing line
ends. Which portion of the

236
00:15:03,836 --> 00:15:07,373
32 bits are the network, and
which portion are unique

237
00:15:07,373 --> 00:15:11,244
host bits. And we represent
subnet masks in one of three

238
00:15:11,244 --> 00:15:13,679
different ways. Typically,
you won't see it in binary,

239
00:15:13,679 --> 00:15:16,315
but this is how your laptop
sees it. It's just a string

240
00:15:16,315 --> 00:15:21,988
of binary bits, but
typically we convert just

241
00:15:21,988 --> 00:15:25,324
like we convert a 32 bit IP
address into dotted decimal.

242
00:15:25,324 --> 00:15:27,293
We do the same thing to the
subnet mask. We take that

243
00:15:27,293 --> 00:15:30,763
entire 32 bit binary number
and convert it into dotted

244
00:15:30,763 --> 00:15:33,633
decimals. So this would be a
common representation

245
00:15:33,633 --> 00:15:39,238
255.255.0.0. And a lot of times
in papers and books and

246
00:15:39,238 --> 00:15:41,574
stuff, they'll shorten it
even further. They'll say,

247
00:15:41,574 --> 00:15:44,811
"Okay well, in this
combination here, I'm seeing

248
00:15:44,811 --> 00:15:48,548
that the first 16 bits are
all 1's so that's telling

249
00:15:48,548 --> 00:15:51,083
me, that of my 32 bit
number, the first 16 bits

250
00:15:51,083 --> 00:15:55,188
are the network portion. So,
shorthand notation would be

251
00:15:55,188 --> 00:16:01,160
/16. Well, guess what? I
have some more questions for

252
00:16:01,160 --> 00:16:04,597
you! So here we go. Let me
expand this a little bit. A

253
00:16:04,597 --> 00:16:08,134
few more quiz questions.
Same thing. I will give you

254
00:16:08,134 --> 00:16:13,339
guys 30 seconds to think
about it and then I will

255
00:16:13,339 --> 00:16:19,278
show you the answer. So here
we go, quiz number 3. Given

256
00:16:19,278 --> 00:16:22,515
the following subnet mask in
binary, what is the

257
00:16:22,515 --> 00:16:29,956
equivalent representation in
dotted-decimal and shorthand?

258
00:16:29,956 --> 00:16:35,995
Okay, time's up.
So here is our answer. So if

259
00:16:35,995 --> 00:16:38,631
you take a look at all the 1s
and remember that your 1s

260
00:16:38,631 --> 00:16:40,466
represent your networking
bits, if you just add them

261
00:16:40,466 --> 00:16:47,306
up, you've got 8 + 8 + 4
which says of the 32 bits,

262
00:16:47,306 --> 00:16:50,276
the first 20 of them are
networking bits. So that gives

263
00:16:50,276 --> 00:16:54,113
us our shorthand answer of
/20. And if you convert this

264
00:16:54,113 --> 00:16:59,085
into a dotted-decimal
number, all 1s is 255.255

265
00:16:59,085 --> 00:17:02,889
and the first four 1s, if
you add those up, 128 + 64 +

266
00:17:02,889 --> 00:17:08,094
32 + 16 gives you 240. So
that's why answer D is the

267
00:17:08,094 --> 00:17:13,833
correct answer also. Let's do a
couple more of these. If the

268
00:17:13,833 --> 00:17:18,104
following subnet mask were to
applied to a PC's NIC, how many

269
00:17:18,104 --> 00:17:21,340
bits of the IP address would be
interpreted as networking

270
00:17:21,340 --> 00:17:30,449
bits? All right, and the
answer to this, 29 of those

271
00:17:30,449 --> 00:17:34,553
bits will be considered
networking bits. Let's just

272
00:17:34,553 --> 00:17:40,326
do that one on the
whiteboard real quick. So

273
00:17:40,326 --> 00:17:45,531
we had 255-- so let's just
make this a little bit

274
00:17:45,531 --> 00:17:54,874
larger. We had
255.255.255.248, so if we

275
00:17:54,874 --> 00:18:01,280
take a look at that, we know
the number 255 is that, so

276
00:18:01,280 --> 00:18:04,216
that tells me that all 8-bits
of the 1st octet are

277
00:18:04,216 --> 00:18:07,820
networking bits. All 8-bits
of the 2nd octet and all

278
00:18:07,820 --> 00:18:10,957
eight bits of the third
octet. And then if we take a

279
00:18:10,957 --> 00:18:19,198
look at the number 248,
that's the first five bits

280
00:18:19,198 --> 00:18:25,004
as being networking bits. So
if you add all those up,

281
00:18:25,004 --> 00:18:29,275
eight plus eight plus eight
is 24. 24 plus five gives

282
00:18:29,275 --> 00:18:33,579
us-- what subnet mask would
be appropriate in dotted

283
00:18:33,579 --> 00:18:37,550
decimal to indicate to a
host that the first 26 bits

284
00:18:37,550 --> 00:18:44,857
of its IP address should be
considered as the network?

285
00:18:44,857 --> 00:18:48,594
All right, the correct
answer for this one was

286
00:18:48,594 --> 00:18:57,503
answer B, the first 26 bits
of the IP address. So if

287
00:18:57,503 --> 00:19:02,208
we're looking at the first
26 bits, it looks like this.

288
00:19:02,208 --> 00:19:04,877
One, two, three, four, five,
six, seven, eight. One, two,

289
00:19:04,877 --> 00:19:07,013
three, four, five, six,
seven, eight. One, two,

290
00:19:07,013 --> 00:19:08,614
three, four, five, six,
seven, eight. That's the 1st

291
00:19:08,614 --> 00:19:14,487
24-bits, 1-2, so there's 26,
3-4-5-6-7-8, and when you

292
00:19:14,487 --> 00:19:22,228
convert those numbers,
you've got 255.255.255.192.

293
00:19:28,200 --> 00:19:31,871
Given the following IP
address and subnet mask,

294
00:19:31,871 --> 00:19:36,008
identify the sub network
address of this host. So

295
00:19:36,008 --> 00:19:41,280
what you have here is a full
32-bit host IP address, and

296
00:19:41,280 --> 00:19:43,683
when the host IP address,
when that host looks at that

297
00:19:43,683 --> 00:19:46,218
IP address that he's got, as
well as his corresponding

298
00:19:46,218 --> 00:19:50,189
subnet mask, what is the
network that he will assume

299
00:19:50,189 --> 00:19:52,525
that he is on, that he's
sitting on? Remember, the

300
00:19:52,525 --> 00:19:56,362
key thing here, about this
is, in a 32-bit IP address,

301
00:19:56,362 --> 00:20:01,834
if all the host bits are 0s,
that is your network number.

302
00:20:01,834 --> 00:20:05,738
So you're trying to extract
from this, what the network is,

303
00:20:05,738 --> 00:20:07,940
where all the host bits are
zeroed out.

304
00:20:11,677 --> 00:20:18,250
Okay, the answer to this one is
answer A. So let's work through

305
00:20:18,250 --> 00:20:34,033
the logic of how this one works.
So here we had 137.54.101.77,

306
00:20:34,033 --> 00:20:48,647
and a corresponding subnet
mask was 255.255.224.0. Now

307
00:20:48,647 --> 00:20:51,383
whenever you see a-- within
a subnet mask, if an entire

308
00:20:51,383 --> 00:20:55,688
byte is the number 255,
remember, that's real easy,

309
00:20:55,688 --> 00:21:01,427
right? That means that all eight
bits of that byte in the

310
00:21:01,427 --> 00:21:05,097
corresponding IP address
belong to the network. So that

311
00:21:05,097 --> 00:21:07,933
means that this entire byte
of 137 belongs to the

312
00:21:07,933 --> 00:21:12,438
network, and this entire
byte of 54 belongs to the

313
00:21:12,438 --> 00:21:15,040
network. Now where it may
have gotten a little

314
00:21:15,040 --> 00:21:17,643
difficult for you was in the
third byte because we don't

315
00:21:17,643 --> 00:21:25,484
have all 8 bits. 224 is this
number, that's 224 which

316
00:21:25,484 --> 00:21:31,056
tells me of the number 101
only the first 3 bits belong

317
00:21:31,056 --> 00:21:34,026
to the network, and then
everything else to the right

318
00:21:34,026 --> 00:21:38,397
needs to be zeroed out as
host bits. So to do this, we

319
00:21:38,397 --> 00:21:41,233
had to convert the number
101 - just the first 3 bits

320
00:21:41,233 --> 00:21:48,941
of it - into binary. So with
101, the 128-bit is off,

321
00:21:48,941 --> 00:21:55,915
64-bit is on, 32-bit is on,
64 plus 32 is 96. At this

322
00:21:55,915 --> 00:21:58,551
point, we can't go any
further because if we turned

323
00:21:58,551 --> 00:22:03,923
on the 16-bit, that would be
too high. And then

324
00:22:03,923 --> 00:22:19,138
everything else, needs to be
zeroed out. 12345678. So

325
00:22:19,138 --> 00:22:23,342
this is telling us here that
these bits here in the pink

326
00:22:23,342 --> 00:22:27,880
boxes all belong to the
network. Everything else are

327
00:22:27,880 --> 00:22:30,983
host bits. So when you
convert that entire thing

328
00:22:30,983 --> 00:22:34,353
into a dotted decimal
number, that's what gave you

329
00:22:34,353 --> 00:22:43,596
your final answer of
137.54.96.0 because remember

330
00:22:43,596 --> 00:22:47,199
to identify the network that
means all host bits have to

331
00:22:47,199 --> 00:22:53,305
be zeroed out. What if I
were to ask you using the

332
00:22:53,305 --> 00:22:58,344
same IP address and same
Subnet mask to identify the

333
00:22:58,344 --> 00:23:04,316
broadcast address of this
network? Remember what makes

334
00:23:04,316 --> 00:23:06,619
the network address
different than the broadcast

335
00:23:06,619 --> 00:23:09,388
address? And the network
address, all the host bits

336
00:23:09,388 --> 00:23:12,291
are 0 because you're
referring to the wire itself

337
00:23:12,291 --> 00:23:15,561
not any particular host on
that wire. You're referring

338
00:23:15,561 --> 00:23:18,797
to the broadcast domain in
general. Well in a

339
00:23:18,797 --> 00:23:22,801
broadcast, you're saying I'm
sending it to all the hosts,

340
00:23:22,801 --> 00:23:29,041
every single host there is.
So 011 ends my network, and

341
00:23:29,041 --> 00:23:35,781
if I'm sending it to all my
hosts, it looks like that

342
00:23:35,781 --> 00:23:41,120
because all my host bits are
set to ones. So to identify

343
00:23:41,120 --> 00:23:47,459
the broadcast address, it
would be 137.54, and then

344
00:23:47,459 --> 00:23:50,663
what's this entire byte
right here? Well that entire

345
00:23:50,663 --> 00:23:54,733
byte, what is it? That's 1
number shy of the number 128

346
00:23:54,733 --> 00:23:57,269
isn't it? Because the next
binary pattern bigger than

347
00:23:57,269 --> 00:24:03,242
that would be 100000 which
is 128 so that means that

348
00:24:03,242 --> 00:24:11,684
number is 127.254. So using
that 1 IP address and Subnet

349
00:24:11,684 --> 00:24:13,986
mask, we were able to
extract from that, and this

350
00:24:13,986 --> 00:24:16,755
is exactly the way your
laptop does it. It extracts

351
00:24:16,755 --> 00:24:19,992
from that the network
address says, "Okay, once my

352
00:24:19,992 --> 00:24:22,628
networking bits end, I'm
just going to zero out

353
00:24:22,628 --> 00:24:24,596
everything." That's the
network. That's the

354
00:24:24,596 --> 00:24:28,033
broadcast domain. That's my
group. And it says, "Okay

355
00:24:28,033 --> 00:24:31,236
and if I turn all those host
bits to 1's, that's how I

356
00:24:31,236 --> 00:24:35,607
send a broadcast to everyone
in my group." Let's do

357
00:24:35,607 --> 00:24:40,713
another one. Given the
following IP address and

358
00:24:40,713 --> 00:24:44,483
subnet mask in binary,
identify which of the

359
00:24:44,483 --> 00:24:47,386
answers below provide for
the broadcast address of

360
00:24:47,386 --> 00:24:50,289
this network. We just did
that. So let's see if you can

361
00:24:50,289 --> 00:24:56,995
do that for yourself. Okay,
time's up. So if we take a look

362
00:24:56,995 --> 00:25:04,570
at this. According to the
subnet mask, this right here

363
00:25:04,570 --> 00:25:08,407
belongs to our network
because the subnet mask says

364
00:25:08,407 --> 00:25:14,546
the first 16 bits belong to
the network. So what is that

365
00:25:14,546 --> 00:25:19,418
number? Well, just by
process of elimination

366
00:25:19,418 --> 00:25:21,487
you can pretty much
assume it's 67 to begin

367
00:25:21,487 --> 00:25:25,958
with and then the second
number is three so that

368
00:25:25,958 --> 00:25:31,597
eliminates answer D, that
can't be right. So 67.3 and

369
00:25:31,597 --> 00:25:33,198
we're being asked to
identify the broadcast

370
00:25:33,198 --> 00:25:36,535
address so that means that
all of the host bits have to

371
00:25:36,535 --> 00:25:48,213
be set to 1's. So that means it
would 67.3.255.255 and that

372
00:25:48,213 --> 00:25:53,719
gives us our answer. Let's
do a couple more. I want to

373
00:25:53,719 --> 00:25:57,823
make sure you really have
this. Different permutations

374
00:25:57,823 --> 00:26:03,228
of the same question. Given
the following IP address and

375
00:26:03,228 --> 00:26:06,432
subnet mask, identify which
of the answers below provide

376
00:26:06,432 --> 00:26:08,634
for the broadcast address of
this network so a very

377
00:26:08,634 --> 00:26:11,370
similar but, a little bit
different. Give you 30

378
00:26:11,370 --> 00:26:16,875
seconds for this one. All
right, time's up. So in this

379
00:26:16,875 --> 00:26:20,579
particular case, the subnet
mask was real easy, right?

380
00:26:20,579 --> 00:26:27,453
Tells us that the first 3 octets
belong to the network, and

381
00:26:27,453 --> 00:26:30,389
so if we take our host bits
and we set them all to the

382
00:26:30,389 --> 00:26:34,326
ones to make a broadcast,
that gives us our correct

383
00:26:34,326 --> 00:26:51,276
answer of 130.54.6.255.
Let's do another one. Given

384
00:26:51,276 --> 00:26:53,679
the following IP address and
Subnet mask - now we're back

385
00:26:53,679 --> 00:26:56,148
in binary - identify which
of the answers below

386
00:26:56,148 --> 00:27:00,519
provides for the broadcast
address of this network.

387
00:27:00,519 --> 00:27:07,526
Thirty seconds beginning
now. Okay, so in this one

388
00:27:07,526 --> 00:27:18,737
right here, we've got all of
this belongs to the network

389
00:27:18,737 --> 00:27:23,509
because that's what matches
our Subnet mask pattern. And so

390
00:27:23,509 --> 00:27:25,677
if we're going to identify the
broadcast address that means

391
00:27:25,677 --> 00:27:28,413
all these remaining host
bits have to be set to 1's.

392
00:27:33,485 --> 00:27:35,854
All right, so this right
here, that's easy. If that's

393
00:27:35,854 --> 00:27:44,196
all set to 1's that's 255.
So we've got 67.3 and then in

394
00:27:44,196 --> 00:27:47,199
the third octet, if we have
011 and then we set the rest

395
00:27:47,199 --> 00:28:00,178
of it to all 1's, that would
give us 67.3. - I made a

396
00:28:00,178 --> 00:28:07,819
mistake here, it's not just
the first 3 bits, it's the

397
00:28:07,819 --> 00:28:12,124
first 4 bits. The network
mask did not quite line up

398
00:28:12,124 --> 00:28:17,262
underneath. That's where we
ended, the first 4 bits

399
00:28:17,262 --> 00:28:21,767
there. So now, if we set the
remaining stuff to 1's, that

400
00:28:21,767 --> 00:28:30,008
third octet if we convert
that into dotted decimal,

401
00:28:30,008 --> 00:28:41,553
will give us the number 111.
01101111 is the number 111.

402
00:28:41,553 --> 00:28:43,855
One more quiz question. One
more and then we'll keep

403
00:28:43,855 --> 00:28:49,461
going on. Same type of thing
- identify the broadcast

404
00:28:49,461 --> 00:28:58,403
address for this network.
Time's up. Correct answer

405
00:28:58,403 --> 00:29:04,910
was the bottom one. In this
particular case,

406
00:29:04,910 --> 00:29:12,618
255.255.224, that tells us
that all of 77, all of 42,

407
00:29:12,618 --> 00:29:15,554
and then of the number 200,
the first three bits because

408
00:29:15,554 --> 00:29:21,326
224 is 111 and then all five
0s. Kind of hard for me to

409
00:29:21,326 --> 00:29:27,833
draw it here so forgive my
sloppy drawing. That's 224.

410
00:29:27,833 --> 00:29:30,702
So the number 200, we've got
the first three bits we're

411
00:29:30,702 --> 00:29:37,843
concerned with. So the
128-bit is on, 64-bit is on,

412
00:29:37,843 --> 00:29:44,683
32-bit is off. That gives us
the number 192. Everything

413
00:29:44,683 --> 00:29:48,420
else is host bits. So if
we're trying to identify the

414
00:29:48,420 --> 00:29:52,190
broadcast address, we have
five more bits here so let's

415
00:29:52,190 --> 00:30:00,866
go ahead and draw those in.
One, two, three, four, five,

416
00:30:00,866 --> 00:30:04,903
and then one, two, three,
four, five, six, seven,

417
00:30:04,903 --> 00:30:15,147
eight. So 77.42, this is the
number 223.255. That is the

418
00:30:15,147 --> 00:30:17,380
broadcast address of that
network.

419
00:30:17,405 --> 00:30:21,976
[music]