End Hosts

Loading...

Из курса от партнера Princeton University

Networks Illustrated: Principles without Calculus

39 оценок

Princeton University

Networks Illustrated: Principles without Calculus

39 оценок

What makes WiFi faster at home than at a coffee shop? How does Google order its search results from the trillions of webpages on the Internet? Why does Verizon charge $15 for every GB of data we use? Is it really true that we are connected in six social steps or less? These are just a few of the many intriguing questions we can ask about the social and technical networks that form integral parts of our daily lives. This course is about exploring the answers, using a language that anyone can understand. We will focus on fundamental principles like “sharing is hard”, “crowds are wise”, and “network of networks” that have guided the design and sustainability of today’s networks, and summarize the theories behind everything from the social connections we make on platforms like Facebook to the technology upon which these websites run. Unlike other networking courses, the mathematics included here are no more complicated than adding and multiplying numbers. While mathematical details are necessary to fully specify the algorithms and systems we investigate, they are not required to understand the main ideas. We use illustrations, analogies, and anecdotes about networks as pedagogical tools in lieu of detailed equations. Please note that per Princeton University policy, no certificates, credentials or reports are awarded in connection with this course.

Из урока

Controlling Congestion in the Internet

The Internet has many important tasks to manage, like routing packets (discussed in the last module) and controlling congestion. This workload is modularized into different functional layers, each responsible for performing a different set of functions, as we will see in this lesson. We will also look at the principles of congestion control, managed at the transport layer.

Divide And Conquer2:10

Layered Protocol Stack4:13

Transport & Network Layers4:42

Processing Layers2:17

Controlling Congestion4:03

Traffic Jam & Bucket Analogy3:02

Sliding Window2:36

Cautious Growth of Window Size6:51

Inferring Congestion5:46

Congestion Control Versions2:16

Loss-Based Congestion Inference5:19

Delay-Based Congestion Inference: Part I6:43

Delay-Based Congestion Inference: Part II5:33

Delay-Based Congestion Inference: Part III4:51

Distributed Congestion Control5:14

Познакомьтесь с преподавателями

Christopher Brinton
Lecturer
Electrical Engineering
Mung Chiang
Professor
Electrical Engineering

So the end-hosts are where all this congestion control actually happens.

And the reason for that we did allude to

a little bit, but really the end-to-end principle was one,

some of the ideas that came behind the network, and

why TCP/IP is allocated in the way that it is.

the idea is to have really intelligent end hosts.

Right? End hosts that.

can process, for instance, all the way up the application

layer and really doing a lot of functions itself and

a dumb core network. Right?

So the idea is that the network just serves

to get the data where it needs to go.

Right?

Because it's not doing anything else.

The cellular network is quite the opposite actually.

The Our end, our end hosts are not as intelligent and

the the core, or the network that send through is, is much

smarter because it's allocating us

different sessions, and it's doing connection

establishment and so on so it's kind of the opposite mentality, really.

But the end-to-end principle

means, really saying.

in, in layman's terms, it's not, you know, it's very

complicated and I supposed much more in depth than this.

But, just have intelligent end hosts and dumb core network, so

a lot of the functionality should lie with the end host.

And so, the senders themselves are going to manage

the end-to-end connections, and they're going to manage congestion.

Right?

That's those are things happening at the transport layer, which nothing

in the network itself really except for that the ends are

going to go up to the transport layer.

So, how is this possible, though, because they're not the network themselves.

Right?

Wouldn't it make more sense to have the routers during

congestion rate because the routers themselves are monitoring the links.

Right?

or pretty closely at least and they can probably

tell how much congestion there is from link to link.

So on the other hand the end host which is just sitting at this one side.

Right?

Is going to have a much harder time to

tell which links within the network are very congested.

But the end-time principle says that we're going to do congestion control over here.

Well, we can do this as we have the

network actually provide hints to the end host, to say.

Say, well this is kind of what the network looks like right

now, and the simplest way to do that is through acknowledgement packets.

Right?

Which is what we talked about with

CSMA carrier sensing multiple access back with Wi-fi.

So, when you send a piece of data out.

once it gets to the receiver.

The way you know it gets to the

receiver is the receiver will send an acknowledgement back.

This is at least the case with TCP transfer control protocol.

With UDP, which is just another transport layer

protocol, you don't deal with acknowledgements or anything.

UDP doesn't really do much functionality at all besides just sending the data.

To where it needs to go.

TCP has a lot more functionality, which also creates

a lot more overhead, so there are trade offs.

so these acknowledgements were sent back upon the proper receipt, and

the, presence and the timing of them can help us infer congestion.

Alright? So first of all, are they even there?

Do we even get an acknowledgement? Right?

If we didn't get an acknowledgement, I mean,

it doesn't necessarily mean that there is congestion.

For instance, it could just be a bad link in the network, but we

could infer that maybe there is congestion and we should slow down our transmission.

Also if the packets come back really late. Right?

If they're taking a while to get back we can

infer that there's a lot of queuing delay within the network.

So,

[UNKNOWN]

things are spending lots of time in the queue.

So, just by the presence of the acknowledgement packet in analysing

the delay in the acknowledgement or

whether or not the acknowledgement arrives.

We can do a lot and infer congestion right at the sending side.

And so, here really recall, we should just recall CSMA.

We talked about acknowledgements in the sense

of knowing whether collisions had occurred or not.

Ознакомьтесь с нашим каталогом

Присоединяйтесь бесплатно и получайте персонализированные рекомендации, обновления и предложения.

Coursera делает лучшее в мире образование доступным каждому, предлагая онлайн-курсы от ведущих университетов и организаций.

© Coursera Inc., 2018 Все права защищены.

Загрузить из App Store

Загрузить в Google Play