EIGRP Unequal Cost Load Balancing


(Rene Molenaar) #42

Hi Lokesh,

Both will be feasible successors yes. However, the route with the lowest feasible distance will become the next successor in case the successor fails.

Rene


(Paul B) #43

Just a note - you probably want to delete your second sentence, it is sort of a repeat of your first sentence.


(Networklessons Admin) split this topic #44

19 posts were merged into an existing topic: EIGRP Unequal Cost Load Balancing


(AZM U) #45

Hello Rene,
In OSPF COST is the key to manipulate traffic. What is the best way to manipulate traffic in EIGRP? Thank you so much.

Azm


(Lazaros Agapides) #46

Hello Azm

It is important to clarify what COST is. The proper term for routing protocols is metric but we often use the word cost to refer to that. Each routing protocol uses a different system for evaluating the metric as you know very well. So when implementing OSPF, when you want to manipulate the way that traffic is distributed among multiple paths to a specific destination, you can tweak the metric or cost by adjusting the bandwidth parameter on the interfaces that form each path, thus changing the OSPF metric to that destination.

The logic is similar for EIGRP. Change the metrics for each path and you can manipulate how traffic is routed. Because EIGRP evaluates the metric differently however, the procedure is a little bit more involved. You can choose from multiple parameters to tweak to get it to function as you like.

According to Cisco:

EIGRP updates contain five metrics: minimum bandwidth, delay, load, reliability, and maximum transmission unit (MTU). Of these five metrics, by default, only minimum bandwidth and delay are used to compute best path. Unlike most metrics, minimum bandwidth is set to the minimum bandwidth of the entire path, and it does not reflect how many hops or low bandwidth links are in the path. Delay is a cumulative value which increases by the delay value of each segment in the path. For more information on EIGRP metrics refer to the Enhanced Interior Gateway Routing Protocol White Paper.

So, in the default configuration, to manipulate EIGRP, you can:

  1. Tweak the bandwidth parameter of one of the links in the path in question (since minimum bandwidth of a path to a destination is used to calculate metric)
  2. Tweak the delay parameter of several of the links in the path in question (since delay is cumulative on the whole link to calculate metric)

If you want to go further, you can activate the other three metrics and use those for more granular manipulation. You can find out more about this at Cisco’s documentation: Setting a preferred route by influencing EIGRP metrics.

I hope this has been helpful!

Laz


(AZM U) #47

Hello Laz,
Thank you so much as usual. In OSPF, manipulation of cost(ip ospf cost …) is used widely to manipulate traffic even though changing bandwidth can also be used to manipulate path. That is why I was wondering what is the best practice to manipulate path in EIGRP. Thanks once again.

Azm


(Lazaros Agapides) #48

Hello Azm

Thanks for clarifying. Yes, the ip ospf cost command on an interface will always override the the formulated cost (using bandwidth etc).

As for EIGRP, there is no such counterpart command, so manipulating path selection must be implemented as stated in my previous post.

I hope this has been helpful!

Laz


(AZM U) #49

Thanks a lot…you the man…

Azm


(florian k) #50

Hi guys,

the FS normally is not inserted in the routing table, only the successor route can be found there.
If you do unequal-cost LB, the FS routes will be put in the routing table, right?

And if there are multiple FS in the EIGRP topology table, what is the deciding factor to choose one to replace the successor route in case it fails? The one with the lowest AD becomes the next successor?

Thanks
Florian


(florian k) #51

Hi guys,

just read the next article “EIGRP variance command example” and that answered already my first question: “the FS normally is not inserted in the routing table, only the successor route can be found there. If you do unequal-cost LB, the FS routes will be put in the routing table, right?”

So only this question is still current: “And if there are multiple FS in the EIGRP topology table, what is the deciding factor to choose one to replace the successor route in case it fails? The one with the lowest AD becomes the next successor?”

Thanks
Florian


(Lazaros Agapides) #52

Hello Florain

The FS is a FS only if it is not in the routing table. If it is in the routing table it has the best metric and is thus considered a successor, even if there are more than one such as is the case with EIGRP load balancing.

If you do unequal cost LB, then the routes in the routing table are also considered successors and not feasable successors. But putting aside terminology, the answer is yes, the routes that were FS (at least the next one(s) in line) is inserted into the routing table and becomes a successor.

The deciding factor of choosing the next successor is always the same, regardless of whether or not you are using load balancing. EIGRP will keep up to six feasible successors in the topology table and only the one (or ones) with the best metric (or metrics) become the successor(s) and are placed in the routing table.

I hope this has been helpful!

Laz


(Dylan M) #53

In the picture it looks like the destination interface is missing it’s value of 5. Is this accurate?
image

Otherwise how would R1 have a F.D. of 15?


(Rene Molenaar) #54

Hi Dylan,

This is a bit confusing yes, I just updated the picture to include a destination value and I also included router names everywhere. The values and tables are also updated.

Rene


(Jermaine D) #55

I have a question for Unequal Cost Load Balancing in EIGRP, do we need to set the maximum paths like RIP
For example in RIP:

R1(config) router rip 
R1(config-router) maximum path 4

(Lazaros Agapides) #56

Hello Jermaine

EIGRP can be configured with equal cost and unequal cost load balancing. For equal cost load balancing, the maximum-paths command determines the maximum number of routes that the routing protocol can use.

When using unequal cost load balancing, using the variance command to determine how many routes will be installed in the routing table. More information about this can be found at the following Cisco documentation:


I hope this has been helpful!

Laz


(DIESSI B) #57

Hello Rene,

I’m new to networklessons.
In principle I would like to congratulate you for the material.
I have a question regarding the relationship of the CEF with the unequal balance of EIGRP.
Does EIGRP unequal balancing work with IP CEF load-sharing per-destination?
Or does it only work with IP CEF load-sharing load-sharing per-packet?

Regards,
Diéssi Borba.


(Lazaros Agapides) #58

Hello Diessi

Welcome to Networklessons! We’re glad to have you on board and to help you out with any and all your questions.

Concerning unequal load balancing using EIGRP, Cisco states the following:

By default, CEF uses per-destination load balancing. If it is enabled on an interface, per-destination load balancing forwards packets based on the path to reach the destination. If two or more parallel paths exist for a destination, CEF takes the same path (single path) and avoids the parallel paths. This is a result of the default behavior of CEF.

When the interfaces through which the paths are of the same type, the hash algorithm determines the path to be chosen. When the load sharing interfaces are of different types, such as a serial port and an Ethernet port, CEF chooses a single path .

However, if you are using interfaces of different types, using per-packet load sharing with CEF will allow load balancing to function correctly.

You can find more information from the following Cisco documentation:

I hope this has been helpful!

Laz


(DIESSI B) #59

Hello Lazaros,

Thanks for your response.
I had read this post from Cisco, but there is something that is not clear to me.

I have noticed that in situations of different links it is necessary to use the per-packet.

My question is this:
In situations that have equals types links. For example, 3 ethernet links, however with metrics differents than EIGRP, scenario to run UCMP. In this situation, will this unequal balancing work even with per-destination mode balance in CEF?

Thank you,
Regards,
Diéssi Borba


(DIESSI B) #60

Hello Lazaros,

Thanks for your response.
I had read this post from Cisco, but there is something that is not clear to me.

I have noticed that in situations of different links it is necessary to use the per-packet.

My question is this:
In situations that have equals types links. For example, 3 ethernet links, however with metrics differents than EIGRP, scenario to run UCMP. In this situation, will this unequal balancing work even with per-destination mode balance in CEF?

Thank you,
Regards,
Diéssi Borba


(Lazaros Agapides) #61

Hello Diessi

If you have unequal EIGRP load balancing across three Ethernet links, and you are using CEF, the default behaviour is to load balance based on destination. This will indeed function, but it is not always the best choice.

Let’s say you have load balancing on these three links with a ratio of 3:2:1. Let’s say you have 60 destinations being routed in a particular period of time. You will have 30 destinations going through one route, 20 going through another and 10 through the third regardless of what the actual traffic for each destination eventually becomes. Once each destination IP is encountered by the router for the first time, it will be routed based on this load balancing algorithm. This is how CEF functions by default. The more destinations there are, the more efficient this process is. You can also see that this is not very effective for traffic with very few destination IP addresses such as two or three.

You can of course modify this default behaviour of CEF by implementing per-packet load balancing as you suggested as well.