EIGRP Metric And K Values

EIGRP Metric And K Values

EIGRP Metric And K Values are important concepts for configuration of EIGRP on Cisco Router. This post also explains how neighbor adjacencies are formed and the purpose of the metrics in EIGRP. The objective of any dynamic routing protocol is to discover remote networks of other routers and achieve convergence in the routing domain. Before any EIGRP update package can be exchanged between routers, EIGRP must discover its neighbors. Neighboring EIGRP are other routers that run EIGRP on directly connected networks.

How EIGRP NEIGHBORS and ADJACENCY Form

EIGRP uses greeting packages to establish and maintain neighbor adjacencies. For two EIGRP routers to become neighbors, several parameters must match between them. For example, two EIGRP routers must use the same EIGRP metric parameters and both must be configured with the same autonomous system number. Each EIGRP router maintains a neighbor table, which contains a list of the routers in the shared links that have an EIGRP adjacency with that router. The neighbor table is used to track the status of these EIGRP neighbors.
The illustration shows two EIGRP routers that exchange initial EIGRP hello packets. When a router with EIGRP enabled receives a hello packet in an interface, it adds that router to its neighbor table.

1. A new router (R1) appears on the link and sends an EIGRP hello packet through all its configured EIGRP interfaces.
2. Router R2 receives the hello packet on an interface with EIGRP enabled. R2 responds with an EIGRP update package that contains all the routes included in its routing table, except those discovered through that interface (split horizon). However, neighbor adjacency is not established until R2 also sends an EIGRP hello packet to R1.
3. Once both routers exchange greetings, neighbor adjacency is established. R1 and R2 update their EIGRP neighbor tables and add the adjacent router as a neighbor.

EIGRP TOPOLOGY TABLE

EIGRP updates contain networks that can be reached from the router that sends the update. As EIGRP updates are exchanged between neighbors, the receiving router adds those entries to its EIGRP topology table.
Each EIGRP router maintains a topology table for each configured routing protocol, such as IPv4 and IPv6. The topology table includes the route entries for each destination that the router discovers from its directly connected EIGRP neighbors.
The illustration shows the continuation of the initial route discovery process on the previous page. Now, the topology table update is displayed.

When a router receives an EIGRP routing update, it adds the routing information to its EIGRP topology table and responds with an EIGRP acknowledgment.

1. R1 receives the EIGRP update from neighbor R2 and includes information about the routes announced by the neighbor, including the metric to each destination. R1 adds all update entries to its topology table. The topology table includes all destinations announced by neighboring (adjacent) routers and the cost (metric) to reach each network.
2. EIGRP update packages use reliable delivery; therefore, R1 responds with an EIGRP acknowledgment packet that informs R2 that it received the update.
3. The R1 sends an EIGRP update to the R2 in which it announces the networks it knows, except those discovered from the R2 (split horizon).
4. R2 receives the EIGRP update from neighbor R1 and adds this information to its own topology table.
5. The R2 responds to the EIGRP update package of R1 with an EIGRP acknowledgment.

EIGRP CONVERGENCE

In the illustration, the last steps of the initial route discovery process are shown.

After receiving the EIGRP update packets from R2, R1 uses the information in the topology table to update its IP routing table with the best route to each destination, including the metric and router of the next hop.
In the same way as R1, R2 updates its IP routing table with the best routes to each network.
At this point, EIGRP is considered to be in convergent state on both routers.

EIGRP K Values & Metrics

By default, EIGRP uses the following values ​​in its composite metric to calculate the preferred route to a network:

• Bandwidth : the slowest bandwidth between all output interfaces, along the route from origin to destination.
• Delay : the accumulation (sum) of all interface delays along the route (in tens of microseconds).

The following values ​​can be used, but not recommended, because they usually result in frequent recalculations of the topology table:

• Reliability : represents the worst reliability between origin and destination, which is based on keepalives.
• Load : represents the worst load on a link between origin and destination, which is calculated based on the packet speed and the configured bandwidth of the interface.

While the MTU is included in the routing table updates, it is not a routing metric used by EIGRP.

EIGRP COMPOSITE METRIC

In Image 4, the compound metric formula that EIGRP uses is shown.

The formula consists of the values ​​K1 to K5, known as " EIGRP metric weights ". K1 and K3 represent bandwidth and delay, respectively. K2 represents load, and K4 and K5 represent reliability.
By default, K1 and K3 are set to 1, and K2, K4 and K5 are set to 0. As a result, only the bandwidth and delay values ​​are used in the calculation of the default composite metric. In EIGRP for IPv4 and EIGRP for IPv6 the same formula is used for the composite metric.
The method to calculate the metric (k values) and the EIGRP autonomous system number must match between EIGRP neighbors. If they do not match, the routers do not form an adjacency.
The default k values ​​can be changed with the metric weights command of the router configuration mode:
Router (config-router) # metric weights tos k1 k2 k3 k4 k5
Note : the modification of the value of metric weights is generally not recommended and exceeds the scope of this course. However, its importance is relevant to the establishment of neighborhood adjacencies. If one router modified the weights of the metric and another router did not, an adjacency is not formed.

VERIFICATION OF THE EIGRP K VALUES

The show ip protocols command is used to verify the k values. In Image 5, the result of the command for R1 is shown. Note that the k values ​​in R1 are set to the default setting.

INTERFACE VALUE ANALYSIS

The show interfaces command displays information about the interfaces, including the parameters used to calculate the EIGRP metric. The illustration shows the command show interfaces for the Serial 0/0/0 interface on R1.

• BW : interface bandwidth (in kilobits per second | kb / s = Kbit / sec ).
• DLY : interface delay (in microseconds | usec ).
• Reliability : interface reliability expressed as a fraction of 255 (255/255 is a 100% reliability), calculated as an exponential average for five minutes.

By default, EIGRP does not include its value when calculating the metric.

• Txload, Rxload : load transmitted and received through the interface expressed as a fraction of 255 (255/255 is completely saturated), calculated as an exponential average for five minutes. By default, EIGRP does not include its value when calculating the metric.

EIGRP BANDWIDTH METRIC

The bandwidth metric is a static value that some routing protocols, such as EIGRP and OSPF, use to calculate the routing metric .
The bandwidth is shown in kilobits per second (kb / s). Most serial interfaces use the default bandwidth value of 1544 kb / s or 1,544,000 b / s (1,544 Mb / s). This is the bandwidth of a T1 connection.

However, some serial interfaces use another default bandwidth value. In Image 7, the topology used in this section is shown. The types of serial interfaces and their associated bandwidths may not necessarily reflect the most frequent types of connections found in networks today.
Always verify bandwidth with the show interfaces command . (EIGRP Verification Commands Cisco)
The default bandwidth value may or may not reflect the actual physical bandwidth of the interface. If the actual bandwidth of the link differs from the default bandwidth value, the bandwidth value must be modified.
BANDWIDTH PARAMETER SETTING
On most serial links, the default bandwidth metric is 1544 kb / s. Because EIGRP and OSPF use bandwidth in default metric calculations, a correct value for bandwidth is very important for the accuracy of routing information.
Use the following interface configuration mode command to modify the bandwidth metric:
Router (config-if) # bandwidth  kilobits-bandwidth-value
Use the no bandwidth command to restore the default value.
The configuration used in the three routers to modify the bandwidth in the appropriate serial interfaces is shown below.
R1 (config) # interface s 0/0/0
R1 (config-if) # bandwidth 64
R2 (config) # interface s 0/0/0
R2 (config-if) # bandwidth 64
R2 (config-if) # exit
R2 (config) # interface s 0/0/1
R2 (config-if) # bandwidth 1024
R3 (config) # interface s 0/0/1
R3 (config-if) # bandwidth 1024

BANDWIDTH PARAMETER VERIFICATION

Use the show interfaces command to verify the new bandwidth parameters, as shown in Image. It is important to modify the bandwidth metric on both sides of the link to ensure proper routing in both directions.

Modifying the bandwidth value does not change the actual bandwidth of the link. The bandwidth command only modifies the bandwidth metric that routing protocols use, such as EIGRP and OSPF.

EIGRP DELAY METRIC

The delay is the measure of the time it takes for a package to cross the route.
The delay metric (DLY) is a static value determined based on the type of link to which the interface is connected and is expressed in microseconds.
Delay is not measured dynamically. In other words, the router does not really track the time it takes for packets to reach the destination. The delay value, like the bandwidth value, is a default value that the network administrator can modify.
When used to determine the EIGRP metric, the delay is the accumulation (sum) of all interface delays along the route (measured in tens of microseconds).

In the table in Image, the default delay values ​​for various interfaces are shown. Note that the default value is 20,000 microseconds for serial interfaces and 10 microseconds for GigabitEthernet interfaces.

Use the show interfaces command to verify the delay value on an interface, as shown in Image 10.

While an interface with multiple bandwidths may have the same default delay value, Cisco recommends not changing the delay parameter, unless the network administrator has a specific reason for doing so.

HOW TO CALCULATE THE EIGRP METRIC

While EIGRP automatically calculates the routing table metric used to choose the best route, it is important that the network administrator understand how these metrics were determined.
The figure shows the composite metric used by EIGRP. By using the default values ​​for K1 and K3, the calculation can be simplified to the slowest bandwidth (or minimum bandwidth), plus the sum of all delays.

In other words, by analyzing the bandwidth and delay values ​​for all output interfaces on the route, we can determine the EIGRP metric as follows:

• Step 1 . Determine the link with the slowest bandwidth. Use that value to calculate the bandwidth (10 000 000 / bandwidth).
• Step 2 . Determine the delay value for each exit interface on the way to the destination. Add the delay values ​​and divide by 10 (sum of the delays / 10).
• Step 3 . Add the calculated bandwidth and delay values ​​and multiply the sum by 256 to obtain the EIGRP metric.

The result of the routing table for R2 shows that the route to 192.168.1.0/24 has an EIGRP metric of 3,012,096.

2.6. EIGRP METRIC CALCULATION
In Image , the topology of the three routers is shown. This example illustrates how EIGRP determines the metric shown in the R2 routing table for the 192.168.1.0/24 network.

EIGRP BANDWIDTH

EIGRP uses the slowest bandwidth in the calculation of its metric. The slowest bandwidth can be determined by analyzing each interface between R2 and the destination network 192.168.1.0.
The Serial 0/0/1 interface on R2 has a bandwidth of 1024 kb / s. The GigabitEthernet 0/0 interface on R3 has a bandwidth of 1 000 000 kb / s. Therefore, the slowest bandwidth is 1024 kb / s and is used in the calculation of the metric.
EIGRP divides a reference bandwidth value of 10,000,000 by the value in kb / s of the interface bandwidth. As a result, higher bandwidth values ​​receive a lower metric, and lower bandwidth values ​​receive a higher metric. 10 000 000 is divided by 1024.
If the result is not an integer, the value is rounded down. In this case, 10 000 000 divided by 1024 is equal to 9765.625. Decimals (625) are discarded, and the result is 9765 for the bandwidth portion of the composite metric, as shown in Image.

EIGRP TIME DELAY

As shown in Image, the same output interfaces are used to determine the delay value.

EIGRP uses the sum of all delays to the destination. The Serial 0/0/1 interface on R2 has a delay of 20,000 microseconds. The Gigabit 0/0 interface on R3 has a delay of 10 microseconds. The sum of these delays is divided by 10.
In the example, (20,000 + 10) / 10, results in a value of 2001 for the delay portion of the composite metric.

METRIC CALCULATION

Finally, use the calculated values ​​for the bandwidth and the delay in the metric formula. The result is a metric of 3,012,096, as shown in Image.

This value matches the value shown in the routing table for R2.