Hello and welcome back once again. Today we're going to look at how your Internet gets messages where they need to go. So we're going to switch actually now and talk about technical networks for this lecture as well as the last one for the course which will be also on the Internet. and the recurring theme here, as you will see, is that the Internet is really a network of networks. So it's composed of a bunch of sub-networks. So, to start off, we, we're going to revisit the idea of sharing, which takes us all the way back to when we were discussing cellular networks, in the first lecture. So, we talked that FDMA, TDMA, and CDMA right? They're divisions in either frequency, time, or codes in terms of allocating users' sum of the network resources. So, those are dedicating resources, so the way that we're sharing the network medium is by dedicating resources. Right? Whether those resources are frequency, time, or code, it's the same thing. That's how two sessions are differentiated from one another or two links are differentiated from one another, in terms of the resources that they're assigned. And this is just a diagram to show that here, we have two senders, Senders 1 and 2, and then Receiver 1 and 2, which are the recipients, again, of whatever information that we're sending. We're showing this here as as a wire. Right, this is a wire. It could also be the air, it doesn't have to be a wire. but basically the idea is that over time, right? So time is actually coming down here. Um, [INAUDIBLE] first Sender 1 has the time slot, then Sender 2 has the time slot, then Sender 1 would have it again and Sender 2 would have it again. Right. So during that time, the entire link is allocated just to Sender 1, this is a TDMA scheme. We could also make this frequency or we can make it codes, as well. But TDMA sometimes just makes the most sense. So that's the idea here. But there's one negative to this, is that we're not using all the resources constantly, right? So, for instance, if you look at this right here. If Sender 1, this is Sender 1's term in time. Sender one doesn't have anything to send right now, okay? So this is just going to be an empty or a wasted link. So, Sender 1 doesn't need to send anything during this time slot. And Sender 2 has something to send then. Then why wouldn't you just have let center 2 send it? Well, we can't do that because the way we're distinguishing users is by dedicating resources. Okay. So the way that we distinguish them is by knowing that during that time period, the entire session is just allocated to that user. Right. This is what we call circuit switching. Okay? This is a circuit switch network, and each of these are circuits, so each, each of these time slots are circuits. There is a paradigm shift, around the time when people start to think about the Internet, to what's called packet switching. So packet switching is different from circuit switching, it's the way that we share the network is by sharing the resources. Under circuit switching, we share the network by dedicating the resources. Under packet switching, we let everyone share the resources all the time. This is how it looks, okay? We divide the messages into smaller pieces called packets, and we'll talk more about what packets are in the next lecture, actually. But just suffice it to say, it's just a really, it's just a small piece of a message. Under packet switching, we don't dedicate any circuits or anything like that. We allow everybody to use the network medium, right? And they can send, right? And they can use the time slots accordingly. So you see, we don't have any wasted space here, because any of the time that Sender 1 or Sender 2 isn't sending, they can be filled in by Sender 2. [INAUDIBLE] anything or Sender 1 could fill in the time slot for Sender 2. So we don't dedicate anything. We just allow everyone to share all the resources. And we call a session here because I've, I've been using the term session. Session is really an application level. So user and session, we'll use interchangeably but they're not exactly the same thing, right? So the session would be, you'd open up a session between, for instance, your computer and my computer. Like if we were on Facebook messaging so those should be a session between my Facebook application and your Facebook application. We could also have another session opened if we were talking on instant messenger somewhere else. then those would be two different sessions. Right. But we're not, we're going to say that would be one user with two sessions but we're just going to call users and sessions are the same thing here. And we're also going to assume the sessions are unicast, meaning there's one source and one destination. So I'm not going to send to multiple people at once which would be multicasting. we're just going to assume that everything is unicasting. So with packet switching here you can see that we're cutting everything up into finer grains and we're not dedicating the network at any given time to anyone or frequency or code or anything else. We're letting everybody share all of the network resources by allowing them to pour their packets, so to speak, into the network. And then, allowing the network really to deal with things like congestion, right or deal with how to get this packet. So how do I know this packet is meant for this receiver if there's a whole bunch of other packets in the network. Well, we'll look at all that. It requires us to do different ways in terms of addressing, right, to be able to actually address this receiver. And so that's the main idea behind packet switching.
