Deterministic Next Hop Ordering in Multi-Tier ECMP Networks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Inconsistent next hop ordering in Equal Cost Multi-path (ECMP) forwarding equivalence classes across network devices in multi-tier networks leads to unpredictable traffic hashing and load imbalance, exacerbated by transient interface link-flap events.

Innovation Solution

Implementing a deterministic next hop ordering approach using unique network device identifiers (NDIs) to enforce consistent path selection across network devices, ensuring that network devices on the same tier with the same set of next hops maintain the same ordering within their ECMP groups, which is programmed into network device hardware using bitmaps to preserve the position of each next hop.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If unordered next hops are used in ECMP forwarding equivalence classes, then network devices can flexibly select paths, but traffic hashing becomes inconsistent and load imbalance occurs

Engineering Contradiction:
Improvepath selection flexibilityVSAvoidtraffic hashing consistency
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent changes the parameter of next hop ordering from unordered to ordered by introducing a deterministic ordering mechanism based on next hop identifiers. This transforms the ECMP FEC from having flexible but inconsistent path selection to having consistent and predictable ordering, resolving the contradiction between adaptability and reliability.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If deterministic next hop ordering is enforced, then traffic routing becomes predictable and load distribution improves, but network device complexity increases

Engineering Contradiction:
Improvetraffic routing predictabilityVSAvoidnext hop ordering mechanism
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements self-service by having network devices automatically generate and enforce deterministic ordering of next hops based on their own identifiers and the received ECMP FEC. The ordering mechanism is self-contained within each device, requiring no external coordination or complex centralized control, thus achieving predictability without proportionally increasing overall system complexity.

Inventive Principle:
Principle #25Self-service

3Reliability

If transient interface link-flap events occur, then network resilience is tested, but next hop ordering becomes unpredictable and load balance deteriorates

Engineering Contradiction:
Improvenetwork resilienceVSAvoidnext hop ordering stability
Core Design Contradiction:
ReliabilityVSStability of the object's composition

Solution Approach 1:

The patent applies preliminary action by pre-establishing a deterministic ordering mechanism for next hops before link-flap events occur. The ordering is based on stable identifiers and pre-configured ECMP FECs, so that even when link-flap events cause dynamic changes, the underlying ordering principle remains intact, ensuring that next hop selection remains predictable and load balance is maintained.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11700201B2Mechanism to enforce consistent next hops in a multi-tier network
Publication Date: 2023.07.11 ARISTA NETWORKS INC
  • US11700201B2 patent drawing
  • US11700201B2 patent drawing
  • US11700201B2 patent drawing

AI summary

In general, the disclosure relates to a method for forwarding a packet through a multi-tier network by establishing a routing protocol session with network devices in the multi-tier network, obtaining routing protocol information from network devices of the multi-tier network, determining a group using the routing protocol information, generating an ordered group listing using network device identifiers (NDIs) for the network devices in the group, and programming a network device hardware of a network device of the set of network devices of the multi-tier network using the ordered group listing. The group includes a set of the network devices of the multi-tier network.