OSPF Metric –Cost

1. Calculate OSPF Best Routes

Okay, the next thing we'll talk about is the best food calculation in OSPF. We have seen how the OSPF process uses some kind of LS, the links to advertisements which contain information about each and every device network and all these LSS to build up the database, which we call a link state database. But now the question is, okay, once it builds the database,is it going to copy the complete database into the routetable or is it going to be more specific? So once it builds a database, it's not going to just copy that particular database into the routing table. So if you remember, we can verify the routing table by using show IP route or show IP route OSPF. And basically, the routing table contains the best route. But again, the database table contains information about each and every advertisement. So it's not simply going to copy that. So what happens is that whenever a router receives, like in this example, let's say the router one, the router one is receiving information about this network. Let's say this router eight is actually advertising this information to its neighbors, and from there, those particular neighbours will pass on the information to their respective neighbors, and finally the router one. Now the router one receives the information from this side when it receives it. So basically, it is also going to include the cost information. We'll talk about OSPF cost or metric, how it is going to be calculated probably in our next sections. So let's assume these are the cost values here. Now it is going to calculate from the router one to reach this particular route. Like here, I got the same information here. So if you go from the router one to reach this particular subnet, as I'm receiving this information from multiple neighbors, and from this neighbor, I received that the overall cost to reach from here is 180. So if you just calculate the overall cost here toreach, this is cost ten, which is the cost of this link, and let's assume the cost of this link is 180, then of course, the cost will calculateoff that particular link at the cost of ten. So it's going to become ten plus 180 plus ten. So it's going to be $200. The cost is similar in that it will calculate the cost from the other side of where it is receiving the harassment for that particular subnet. So it's going to calculate the cost of 20 plus 30 plus 40 plus ten. So the overall cost will be $100 on this particular path. And likewise, it is going to also calculate the cost from the third path, that is from Rone, R2, three, four, and eight. So the overall cost is like 30 plus 60, it's 1990 plus 20, it's 115, and then plus ten, 125. So the cost on the third route is 125. So depending upon this, the cost here is $200, this is $100, and this is $125. So it is going to figure out the least cost. Now, the least cost is nothing but more bandwidth. Again, we'll talk about that in the next topic. So, in this example, this will be the best route. So it is going to copy the best route in the routing table. So if you verify the Show IP route in the routing table, even though the router has multiple paths to reach this particular subnet, it's going to write down the best route from the routing table. So into the routing table. So that's what it's going to do. Like here, I said, the sum of all the interfacecosts to that outgoing interface will be calculated and then the router is going to add that particular best route in the routing table, which we can verify by using the Show IP out command. So if you try to verify in our practical example here, which we did earlier with a single area like the router to reach this particular 192-1681 dot subnet here. So the overall cost you can see in the rounding table. It clearly writes down here the cost. Again, we'll see how to calculate as Isaid next or how exactly the calculation goes. So the cost on this link is 64, and this is one. So the overall score becomes 65. So the cost of the serial link is 64, and this is one. So the overall cost to reach the three networks will also be 65. So if you want, you can just go through with our topology. I'll try to verify from the router one to reach the remaining networks, like if I say Show IP route or if I say Show IP route OSPF, you can see the cost values here. Again, you can see in the routing table,it will automatically copy the cost values here from the router one to reach this network. So these are the cost values. I'm just writing here. So if you go from here to here, from router one to three networks, the overall cost is 64 plus 64,128 plus 1129, and that's what you can see there, similar way from the router one to reach these two networks. So the overall cost is 64 plus 165, and that's what you can see over there.

2. OSPF Metric –Cost

Okay, so the next thing we'll try to understand is the OSP of metrics. Like previously, I discussed an example where it says OK, each and every interface it will derive the cost, and it is going to calculate the overall cost to reach that particularization based on some of all the cost values. But now the question is, how are these values derived? What is the formula? What is the process behind this one? So, by default, OSPF, like in the case of Ciscorouters, on which we are focusing here, has three different methods for calculating cost values. So the first one is based on the default calculation for the interface. This is the first method. Again, I'll just explain that in the next slide. So the default one, which is called interface bandwidth, So, based on the interface bandwidth. similar to ten to power divided by bandwidth It will take, let's say, the bandwidth on this interface is going to automatically be ten to the power of eight. At 80, you can write down nothing but 100 MPS, which means again, you can write something like this: if you write down one, five. Four Kbps.Kbps means three zeros and if you just derive this, you get 64. So I can see the table here. There is a default reference table. So, depending upon the bandwidth you have, the more bandwidth you have, the more the cost will come down. So it is like inversely proportional and, by default, the cost value will be derived based on the bandwidth settings you make on the interface. So that is one option, which is the default option, we can say. But again, there are two or more other options where we can change the cost values by directly going to the interface and we can change it.So that is another option. Again, we will verify these options in a little bit of detail in the next few slides. And the third option is that we can also change the reference value. The default reference value is ten to the power of eight divided by bandwidth. We can change that. That is the third option. The calculations will be performed based on the configuration commands you enter into the OSPF. And let's say if I'm not making any changes, the default will be the first option. So let's try to understand the first option, the default. And then we'll also try to see the other two options which can influence the cost values in your OSPF domain or the OSPF configurations. So let's start with the first option. So the first option is the default reference bandwidth. That's what we call the defaultreference bandwidth, which means that by default, the SPF algorithm is going to calculate the cost for each route based on the interface bandwidth, right? So we know that the best rate is calculated based on the least cost, and that particular cost is derived based on the bandwidth. Like here, you can see the table which indicates the default formula is like 100 Mbps, which means, as I said,ten to the power of eight means you can write eightzeros, probably ten to the power of eight. Okay, so 800 MPs, so probably 500MPs, and then divided by whatever bandwidth you have on that particular interface. Like, if I'm using the bandwidth of one five double-four KBS, one five double 40 zero, then based on the calculation, you get the cost as 64. So similarly, if I'm using a link of an Ethernet link,let's say this link is Ethernet and by default it supports ten MPS, that is the default interface bandwidth. The cost will automatically come to ten. If the link is a fast alternate link, which is 100 MPs, it will be one. And again, the main drawback with this first option is that if you are using anything over 100 MPS, the cost value will be delayed as well, because the default reference is 100 MPS. So even if I'm writing one gig that is divided by interface bandwidth, still, you know, it will counter as one, which is again the limitation. So there is no mechanism to differentiate the links which are higher speed than 100 MPS, which means that whether you're using 100 MPS or a 100 gigabit interface, the cost value will be derived as one. So, that is one kind of limitation. Okay, so by default, this is the way it will calculate. And if you want to verify this, you can use Show IP, OSPF interface Brief, where you can see the interfaces you have and what the cost is present on those interfaces. I'll attempt to validate the cost values on both routers and both interfaces. So either I can use Show IP OSTF Interfacebrief brief command or it doesn't work in the packet racer. So you can try this in June on the physical devices. So I just use the OSPF interface which lists all the interfaces and you can see this is my gig Ethernetlink which is showing the cost of one. And if you come down a little bit, you can see the serial link. The cost is $64.

3. Changing Interface Bandwidth

So by default, this is the way it will calculate. And if you want to verify this, you can use ShowIP OSPF Interface Brief where you can see the interfaces you have and what the cost return on those interfaces is. I'll try to double-check the cost values on both routers and both interfaces. So either I can use Show IP OSPF Interface brief briefcommand doesn't work in the packet register, so you can try this in the Juness or on the physical devices. So I just use the OSPF interface which lists all the interfaces and you can see this is my gig ethernetlink which is showing the cost of one. And if you come down a little bit youcan see the serial link, the cost is 64. So this is like the default cost because the default bandwidth on this interface is one gigabit per second and the default bandwidth here is like 1000Mbps, one gigabit per second. It's a gigabit link. So this cost, whatever the cost I'm writing here, is going to vary based on the bandwidth. Like the G zero by zero interface, the bandwidth is one gigabyte. So what if I change the bandwidth? Again, one more thing you need to know here is that whatever the bandwidth you have on the interface is going to be used in the calculation. Let's say if I'm changing the bandwidth on this interface, let me just quickly go to the interface zero and I'm going to change the bandwidth. Let's say the bandwidth command is one five double four. So the bandwidth value we queue is in Kbps. So one hundred fifty-double-four Kbps is nothing but equivalent to serial bandwidth. So when I do this automatically, it will influence the cost if you verify Show the Ipospf interface. Even though I'm using a gigantic link because that is the default transmission speed it supports. But again, in the production networks, even though you're using an internet link or maybe, let's say I'm using a thisvan link, maybe it's not exactly one gig. So it might be something like, let's say, it just supports 50 MPs, let's say. So in that scenario, you really want to change the bandwidth and, depending upon that, the cost values will automatically change. So the cost value, which is derived by default, will be based on the interface bandwidth. And if you don't define any bandwidth, again, as Isaid, let's say you're connecting my PC in my home and I'm connecting to my internet or the service portal from where I'm getting the internet, and even though I'm connecting my NIC card, let's say it supports 100 MPs port. So it's not like that. You get a speed of 100 mps. So whatever the speed of the bandwidth you will be offered depends upon the line you purchase. Let's say I'm paying for 15 MVPs, so I'll be getting a maximum transmission speed of not exceeding 15 MPs, right? But at the same time, it also means that this interface can support up to 100 MPs. So, but again, there is nothing likecalculation required here because we are not really deciding any best routes. But in the case of OSPF, when you are taking the same example in terms of OSPF, when I'm connecting gigs in alink, the default transition speed will be one gig. Like you can say, 1000 MPs. So by default, this will be the interface bandwidth. So it is going to be used when it is using the formula or calculating the cost. It is going to be based on the default bandwidth and interface bandwidth. But most likely, even though you are connecting to this giglink, maybe from the service portal, you are just getting 100 Mbps of speed or the van speed. So it's recommended you go to the interface and change the bandwidth, basically the same bandwidth that is being provided by the service provider. so that when it calculates the best route, it will always calculate correctly. I'm going to say that at 50 Mbps. So I can just go and say 1500 Kbps, which is basically what we give you in Kbps. Which means if I need to give 15 MPs, I probably need to say something like this in Kbps, so we need to change this accordingly. So if you don't change that, then that can influence and that can actually impact the decisions. So that's the reason, especially on the van links. It's not required on the landlines, but on the van links you generally need to do that. Okay, so this is how OSPF calculates cost values, and we can verify it with this show IP OSPF interface brief. Or you can use SHO IP OSPF interface, which will be sufficient in general.

4. Default Reference Bandwidth – Limitations

So one of the main limitations, as I said, with the default reference bandwidth, is that initially, when they started, this probably was very good with slower links. Even if your bandwidth is up to 100 Mps, this is really good. At current costs, it can extract up to 100 MPs. But when your link speed goes above 100 Mbps, then basically there is no differentiation. So, which means if I'm connecting two links,one is a 100 bps link and the other is, let's say, a ten gig link. So basically, in terms of calculation, it is going to consider these two links as the same speed or same cost. So that was one of the limitations. So any interface with 100 MPS or faster will be calculated with a cost value of one. So basically, it's going to round up all these calculations to the least cost of all 100 MPs or above that. So that's what, which is really not right,especially when you're deciding the best routes. Because when you're having 100 bpslink or when you are having.

ExamSnap's Cisco 350-501 Practice Test Questions and Exam Dumps, study guide, and video training course are complicated in premium bundle. The Exam Updated are monitored by Industry Leading IT Trainers with over 15 years of experience, Cisco 350-501 Exam Dumps and Practice Test Questions cover all the Exam Objectives to make sure you pass your exam easily.