Cisco CCNA 200-301 – WAN – Wide Area Networks Part 3

7. PPPoE Point to Point Protocol over Ethernet

In this lecture, you’ll learn about PPPoE, the point to point protocol over Ethernet. In earlier lectures in the section, you learned about the different Wan options that a large company or large office would use for their Wan and their Internet connectivity. Less expensive options, which are often aimed at home user Internet access, are often used as Internet VPN Wan backup options in those corporate environments.

They can also be used as the primary Wan connection method to the corporate network from smaller offices and for home users. So the types of connections we’re talking about here are DSL, digital subscriber line, cable, and wireless. For example, 4G over cellular networks. With these connection types, there will typically be no corporate level SLA because they’re going over the Internet using VPN. So that could be going over multiple providers.

So there’s no way to give end to end guarantees. So looking an example of how this would be used, we’ve got a large company here, they’ve got an office in New York as we’re headquarters, and we’ve got a large branch in Boston. They’ve got a leased line to the Internet from New York, and there’s also a leased line directly between the New York and Boston offices.

So whenever there’s traffic between New York and Boston, it will go over that leased line and it’s got a good SLA there. Whenever users in New York go out to the Internet, it goes over that lease line. Whenever users in Boston want to go out to the Internet, it will go over the lease line to New York and then break out to the Internet from there.

So this is all good, but what if that leased line between New York and Boston goes down? The company are concerned about that, so they want to put in a backup option, but they’re not going to put in another leased line just for their backup because that’s expensive. So what they could do is put in a 4G, for example, connection from the Boston office out to the Internet.

Now, if that leased line between New York and Boston goes down, the users in Boston can still get out to the Internet over that 4G connection, and they can also set up an Internet VPN connection between the two offices as well. So if the lease line goes down, then they won’t get as good quality, but at least the connection is still there.

Another reason that they would want to use one of these kinds of options is they’ve got an office in London, which is a small branch office with just a handful of staff there. So they can’t justify the cost of putting in a leased line in London. They could put in a DSL connection there that gives the users their Internet access, and they could also configure an Internet VPN over the same connection for access to the corporate network PPPoE.

PPP over Ethernet is commonly used in DSL deployments, and PPPoE can be configured on either the DP DSL modem or on the router. So my experience is usually configured on the modem. So you don’t have to do anything special on the router. But maybe in your country it is common to do it on the router. Okay, that’s everything I needed to tell you here. See you in the next lecture.

8. WAN Topology Options

This is the last lecture in the Wan section and it’s going to be just a short lecture on the Wan topology options. Now for this, just imagine that all of your connections are point to point leased lines and we’re just thinking about our Wan connections between our offices here. Don’t think about internet connections connections or VPN connections because it will just get confusing if you do that. But you’re going to get questions probably on this in the CCNA exam.

Easy way to answer them is if you just think of point to point least lines between offices. So the first of our topology options is a hub and spoke, which is also known as a star. You see in the example here that New York is our hub site. Our spoke sites are Los Angeles, New Orleans and Boston. So probably New York is the headquarters and the other offices are the branch. All of the branch offices connect into the head office, which is the hub.

So if Los Angeles wants to send traffic to Boston, that traffic will have to go via New York. We don’t have a direct connection between Los Angeles and Boston or New York between any of the branch offices. The advantages of using a hub and spoke topology is it’s simple and you get simplified centralized security policy.

Because all of the traffic going between your branches goes through New York, you can secure it in that central location in New York. Disadvantages are there’s a single point of failure here? If we lose New York, we’ve lost connectivity between all of our different offices. Also it’s suboptimal traffic flow. Traffic from the branches to another branch is not going directly there, it’s having to go through the hub site. That’s going to add some delay.

The next option we can use is a redundant hub and spoke where we put in an extra hub. So in the example it’s the same as before, but now we’ve also got a hub site in Washington as well and all of the branches are connected to both New York and Washington as well. That second hub could be in a different city or it could be in the same city, even the same location as the main hub. Advantages of this are pretty obvious. The main one, it removes that single point of failure.

If we lose New York, all of the offices can still talk to each other via Washington. Again, we have that centralized security policy for traffic between all of the offices. It’s going to be secured in New York and Washington. Disadvantage is obviously the cost is going to go up. We’ve just doubled the amount of links that we’ve got here and it’s still suboptimal traffic flow because traffic between branches is still going via the hub site. The next topology option we have is a full mesh.

This is where we connect every office to every other office. The advantage is that we get the optimal traffic flow traffic is taking the direct path disadvantage is that it’s higher complexity and it’s going to be higher cost as well because we’ve got so many more links. Now the last topology that we’ve got is a partial mesh which is a trade off between the other options. So here we’ve got New York which is acting as a hub site, and we’ve got our branches in Los Angeles, New Orleans and Boston. But here Los Angeles and New Orleans have also got a direct connection between them as well. So this is going to be a bit higher cost than a pure hub and spoke, but we’re going to get some of the advantages of more direct connectivity between some of our branches by doing this.

Okay, so those were the topology options for Wan connections between offices. The last thing that I want to speak about here is a related topic which is Internet redundancy options. So in the offices for a company they’re going to want to have internet connectivity there. The first option is single home and you can see here that the customer router is connected to a service provider router with a single link. So that is a single home connection. Hopefully you can see the obvious potential issue there which is the single points of failure. If either router fails or if the link fails, then the customer is going to lose internet connectivity in that site. So for redundancy the customer may want to use dual homes instead. And you can see that what we’ve done is double up on everything here.

So there’s two customer routers connected to two service provider routers over separate links. So if any router or link fails, the customer will still retain the internet connectivity. Now it looks like there’s no single points of failure here, but there actually is one because what if the service provider has some kind of internal issue and they lose their internet connectivity? Not very likely I know, but it’s not completely unheard of. So if a customer is worried about that, what they can do is use multi homes. And at first glance this looks the same, but you see that now the customer is connected to two different service providers. So they’ve got a connection to service provider one and a connection to service provider two.

And now there’s no single points of failure. And the last option that the customer customer has is dual multi homes which again they are connected to two different service providers but the routers have got multiple connections to the different service providers. This might seem a little bit paranoid, but this does give you the highest level of redundancy for your internet connections. Okay, that is it for the Wan section. See you in the next section.

• Category: Uncategorized