Overlay Network QoS Enforcement for Branch Connectivity

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing wide area network (WAN) solutions, such as MPLS, are expensive and have limited flexibility in supporting branch offices, while enterprises seek a cost-effective, globally accessible, and secure network service that optimally delivers cloud experiences to users, including branch offices, data centers, and mobile users over the Internet.

Innovation Solution

The implementation of an Internet-based overlay network with quality-of-service (QoS) management and enforcement mechanisms, allowing for dynamic capacity allocation and traffic optimization across multiple levels (global, regional, and branch) to ensure predictable and secure network performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If MPLS is used for WAN connectivity, then network performance and security are improved, but cost increases significantly

Engineering Contradiction:
Improvenetwork performanceVSAvoidcost
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The network is segmented into overlay and underlay layers. The overlay network provides virtual dedicated connections with QoS guarantees, while the underlay uses shared public Internet infrastructure. This segmentation allows enterprises to get MPLS-like performance where needed while using cost-effective Internet for other traffic.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

An overlay network with QoS enforcement mechanisms acts as an intermediary between enterprise applications and the public Internet. This intermediary layer provides traffic optimization, capacity enforcement, and performance guarantees without requiring direct MPLS infrastructure.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If MPLS providers place VPN concentrators at edge locations, then branch office connectivity is improved, but the number and location of concentrators is limited resulting in varying performance

Engineering Contradiction:
Improvebranch office connectivityVSAvoidperformance consistency
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The overlay network infrastructure provides universal access points distributed globally, allowing any branch office location to connect without being limited by physical concentrator locations. The system serves multiple functions including VPN termination, QoS enforcement, and traffic optimization at each overlay node.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The solution moves from a limited physical dimension (specific concentrator locations) to a virtual dimension (overlay network addresses). Branch offices can be assigned overlay addresses that route through optimal paths regardless of physical concentrator locations, enabling consistent performance across all locations.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Loss of energy

If Internet packet switching is used instead of MPLS, then cost is reduced, but QoS control and performance predictability worsen

Engineering Contradiction:
ImprovecostVSAvoidQoS control
Core Design Contradiction:
Loss of energyVSReliability

Solution Approach 1:

The overlay network dynamically adjusts routing paths and applies QoS policies based on current network conditions. Capacity enforcement mechanisms adaptively control traffic rates and prioritize flows, providing predictable performance over the dynamic public Internet infrastructure.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes key network parameters including routing paths, traffic rates, and priority levels through QoS policies. Capacity enforcement mechanisms modify these parameters in real-time to maintain performance guarantees while utilizing cost-effective Internet infrastructure.

Inventive Principle:
Principle #35Parameter changes

4Ease of operation

If an overlay network is implemented without QoS enforcement, then ease of operation is improved, but network performance and capacity utilization worsen

Engineering Contradiction:
Improvenetwork operationVSAvoidnetwork performance
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The QoS system operates autonomously with automated capacity enforcement, policy application, and traffic optimization. The overlay network self-manages resource allocation and performance control without requiring manual intervention, maintaining ease of operation while delivering guaranteed performance.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system implements continuous monitoring of network conditions, capacity utilization, and performance metrics. This feedback drives automated QoS policy adjustments and capacity enforcement actions, optimizing network performance while maintaining simple operation through centralized management.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS11082334B2Distributed quality-of-service (QoS) in an overlay network using capacity enforcement
Publication Date: 2021.08.03 AKAMAI TECHNOLOGIES INC
  • US11082334B2 patent drawing
  • US11082334B2 patent drawing
  • US11082334B2 patent drawing

AI summary

Techniques for enhanced overlay network-based transport of traffic, such as IPsec traffic, e.g., to and from customer branch office locations, are facilitated through the use of the Internet-based overlay routing infrastructure. This disclosure describes managing and enforcing quality-of-service (QoS) in an Internet-based overlay network shared by a set of content provider customer entities. For each entity having a customer branch, the customer branch is coupled to the Internet-based overlay routing network. A quality-of-service (QoS) policy is configured for the customer. Utilization of the Internet-based overlay network against the configured QoS policy is then monitored. The QoS is then enforced for the customer and at least one other customer, based in part on the QoS policies. Capacity is enforced for a customer entity according to the QoS policy at one of: a global level, a geographical region level, and at the customer branch level.