Advanced EIGRP: Unequal-Cost Load Balancing and Route Summarization

Basic EIGRP configuration, covered earlier in this series, only scratches the surface of what the protocol can do -- EIGRP can distribute traffic across paths with genuinely different costs, and can summarize routes at any arbitrary point in the network rather than only at fixed area boundaries. This article explains EIGRP's variance-based unequal-cost load balancing, covers flexible route summarization at any interface, and walks through EIGRP stub routing for hub-and-spoke topologies.

EIGRP VarianceEIGRP SummarizationEIGRP Stub Routing

~6 min read · Updated Sep 12, 2026

Beyond Equal-Cost Load Balancing

Every routing protocol covered so far in this series, including OSPF and basic EIGRP, discussed earlier in this series, automatically load-balances across multiple paths only when those paths have exactly equal cost. EIGRP offers something more flexible: the ability to load-balance across paths with genuinely different costs, proportionally distributing traffic based on how close those costs actually are.

The Variance Command: Enabling Unequal-Cost Load Balancing

Recall from earlier in this series that EIGRP's DUAL algorithm identifies a Feasible Successor -- a backup path guaranteed loop-free, but which by definition has a higher metric than the successor. The variance command allows EIGRP to actually use one or more feasible successors for forwarding traffic simultaneously, not just hold them in reserve as backups.

Router(config)# router eigrp 100
Router(config-router)# variance 2

-- This allows any feasible successor whose metric
-- is at most 2 times the successor's metric to
-- also be used for active forwarding

Example: if the successor's metric is 1000,
a feasible successor with metric up to 2000
now also becomes eligible for forwarding traffic

How Traffic Gets Distributed Proportionally

Router# show ip route 192.168.2.0

Routing entry for 192.168.2.0/24
  Known via "eigrp 100", distance 90, metric 1000
  * 10.0.0.2, via GigabitEthernet0/1
    10.0.1.2, via GigabitEthernet0/2

-- Both paths are installed. EIGRP distributes
-- traffic across them in inverse proportion to
-- their metrics -- the path with metric 1000
-- receives proportionally more traffic than
-- a path with metric 1500, rather than an
-- even 50/50 split

This proportional distribution, rather than an even split, is an important distinction from simple equal-cost load balancing: EIGRP sends more traffic down the genuinely better path and proportionally less down the worse (but still loop-free) path, making efficient use of both links without treating a significantly worse backup path as equally preferable to the primary one.

The Risk of Setting Variance Too High

Router(config-router)# variance 100

-- An excessively high variance value can allow
-- paths with dramatically worse metrics to
-- participate in forwarding -- for example, a
-- congested, low-bandwidth backup link might
-- become eligible alongside a high-bandwidth
-- primary link, potentially causing poor
-- overall performance if traffic is proportionally
-- sent down a genuinely unsuitable path

Variance should be configured deliberately based on actual link characteristics in a given network, rather than set to an arbitrarily high value simply to maximize the number of paths used -- a feasible successor existing does not automatically mean it should be actively used for forwarding traffic.

Manual Route Summarization at Any Interface

Unlike OSPF, where summarization is only configurable at Area Border Routers, discussed earlier in this series, EIGRP allows manual summarization to be configured at essentially any interface on any router.

Router(config)# interface gigabitethernet 0/1
Router(config-if)# ip summary-address eigrp 100 192.168.0.0 255.255.0.0

-- This router will advertise a single summarized
-- /16 route out this specific interface,
-- regardless of how many individual /24 subnets
-- actually exist behind it -- and this can be
-- configured on any router, not just designated
-- boundary routers

This flexibility is a genuine advantage over OSPF's more rigid area-boundary-only summarization -- an EIGRP network can summarize routes at exactly the point in the topology where it makes the most sense, rather than being constrained to summarize only where an administrative area boundary happens to exist.

The Automatically Created Discard Route

Router# show ip route

D    192.168.0.0/16 is a summary, 00:05:12, Null0

-- Whenever a summary route is configured, EIGRP
-- automatically installs a route to Null0 (a
-- discard interface) for the summarized range
-- itself -- this prevents a routing loop:
-- if a packet arrives destined for an address
-- within the summarized range but not matching
-- any more specific actual subnet, it is
-- discarded here rather than being sent back
-- out toward whichever router originally
-- advertised the summary, which could create
-- an infinite loop between the two routers

This automatic discard route is an important safety mechanism that is easy to overlook -- without it, a packet destined for an address that falls within the advertised summary range but does not actually correspond to any real subnet could bounce endlessly between routers, each believing the other has a more specific route.

EIGRP Stub Routing: Optimizing Hub-and-Spoke Topologies

In a hub-and-spoke topology, discussed earlier in this series regarding SD-WAN's traditional WAN alternative, spoke routers typically have only a single path back to the hub and do not need to serve as a transit path for traffic between other spokes. EIGRP Stub Routing allows a spoke router to explicitly inform the hub of this limitation.

Spoke-Router(config)# router eigrp 100
Spoke-Router(config-router)# eigrp stub connected summary

-- Declares this router as a stub, advertising
-- only connected and summary routes, and
-- explicitly telling the hub router not to
-- query this spoke when searching for an
-- alternate path during convergence

The genuine benefit here relates directly back to DUAL's query process, discussed earlier in this series: when a route becomes unreachable and no feasible successor exists, EIGRP must query neighbors to find an alternate path. Without stub routing, the hub might query every spoke router in a large hub-and-spoke topology, even though spokes are known in advance to never provide a valid transit path -- stub routing eliminates this unnecessary querying, speeding up convergence in exactly this common topology.

Verifying Stub Router Configuration

Hub-Router# show ip eigrp neighbors detail

EIGRP-IPv4 Neighbors for AS(100)
H   Address     Interface   Hold  Uptime
0   10.0.0.2    Gi0/1        13   00:15:42
   Version 20.0/3.0, Retrans: 0, Retries: 0
   Stub Peer Advertising ( CONNECTED SUMMARY ) Routes

This output on the hub router confirms it correctly recognizes the spoke as a stub and knows precisely which route types to expect from it -- essential verification when troubleshooting why a spoke's routes are not appearing as expected elsewhere in the topology, or when confirming that query traffic is being appropriately limited during a convergence event.

Why These Advanced Features Matter for Real Deployments

Unequal-cost load balancing, flexible summarization, and stub routing represent exactly the kind of nuanced protocol behavior that separates basic EIGRP configuration competency, covered earlier in this series, from genuine advanced routing expertise. Each feature solves a specific real-world limitation: making full use of unequal-bandwidth redundant links, summarizing at whatever point in the network topology actually makes sense rather than being constrained to area boundaries, and optimizing convergence speed specifically for the extremely common hub-and-spoke design pattern.

Written & researched by Dr. Shahin Siami

Related Articles

Systematic Network Troubleshooting: A Methodology Tying Everything Together

Every protocol and technology covered throughout this series is only useful if a problem involving it can actually be diagnosed and fixed efficiently under real-world pressure. This article presents a systematic troubleshooting methodology built around the OSI layers, walks through applying it to a realistic connectivity problem, and shows how the specific verification commands covered throughout this entire series fit into a structured diagnostic process.

Continue

NETCONF, YANG, and Python: Programmatic Network Configuration at Scale

The REST APIs and JSON/YAML formats covered earlier in this series represent one approach to network automation, but NETCONF and YANG provide a more structured, standards-based alternative purpose-built for network device configuration. This article explains what distinguishes NETCONF from a simple REST API, covers how YANG models define exactly what configuration data looks like, and walks through using Python to programmatically interact with network devices.

Continue

IPsec VPN Fundamentals: Securing Traffic Across Untrusted Networks

Connecting two sites across the public internet exposes traffic to interception unless it is properly encrypted, and IPsec provides the standard framework for building secure, authenticated tunnels between sites. This article explains the two-phase IKE negotiation process, covers the distinction between AH and ESP protocols, walks through configuring a basic site-to-site IPsec VPN, and covers essential verification commands.

Continue

MPLS Fundamentals: Label Switching Explained

Traditional IP routing requires every router along a path to perform a full routing table lookup on every packet, but MPLS takes a fundamentally different approach by making that forwarding decision once and attaching a simple label that every subsequent router can use instead. This article explains the core label-switching concept, walks through how the Label Distribution Protocol builds the label forwarding tables that make this possible, and covers the practical benefits MPLS provides in real provider networks.

Continue

BGP Route Reflectors and Confederations: Scaling iBGP Beyond Full Mesh

The iBGP full-mesh requirement, briefly mentioned earlier in this series, becomes a serious scaling problem as an autonomous system grows, requiring a number of sessions that increases quadratically with router count. This article explains exactly why full mesh does not scale, walks through how route reflectors solve this by relaxing BGP's normal route-propagation rules, and covers confederations as an alternative approach that divides a single AS into smaller sub-autonomous systems.

Continue

OSPF Area Types Deep Dive: Stub, Totally Stubby, and NSSA

Multi-area OSPF, covered earlier in this series, already reduces database size by separating a network into areas, but OSPF offers further specialized area types that reduce routing table size even more aggressively by filtering out unnecessary external routes entirely. This article explains the LSA types that must be suppressed to create each specialized area type, walks through configuring stub, totally stubby, and not-so-stubby areas, and covers the specific trade-offs each design choice involves.

Continue