Enterprise Network Agent Traffic Engineering Cloud Services
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems lack the ability for enterprise networks to effectively control incoming traffic from cloud-based networks, relying on the error-prone Border Gateway Protocol (BGP), which is not designed for this purpose, leading to instability and unpredictable outcomes.
Innovation Solution
An enterprise network agent determines and transmits traffic policies to a cloud service provider, allowing for the monitoring and management of incoming traffic across multiple ingress points, enabling load balancing, cost reduction, and improved performance by executing policies through a cloud agent that instruments routing tables and sets up tunnels between border routers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If ad-hoc solutions based on Border Gateway Protocol (BGP) are used to control incoming traffic, then operators can attempt to control traffic flow, but the system becomes error-prone, vulnerable to instabilities and with unpredictable outcomes
Solution Approach 1:
The patent introduces a traffic engineering system with specialized agents (enterprise network agent, cloud service provider agent, and optional internet service provider agent) that act as intermediaries between cloud-based networks and enterprise networks. These agents implement standardized protocols and procedures specifically designed for traffic engineering, replacing the inappropriate use of BGP and providing reliable, predictable traffic control without the instability and errors associated with ad-hoc BGP configurations
Solution Approach 2:
The system dynamically changes traffic engineering parameters such as routing policies, load balancing weights, and traffic distribution ratios through programmable interfaces. This allows operators to precisely control traffic flow characteristics (rate, manner, destination) in a stable and predictable way, avoiding the unpredictable behavior of BGP-based ad-hoc solutions while maintaining flexibility in traffic management
2Ease of operation
If cloud based networks control the rate and manner of data transmission to enterprise networks, then data flow is simple to manage from cloud side, but enterprise networks lack the ability to sufficiently control incoming traffic
Solution Approach 1:
The patent segments traffic control authority into multiple independent agents: the cloud service provider agent controls outbound traffic from the cloud, the enterprise network agent controls inbound traffic to the enterprise network, and optionally the internet service provider agent controls intermediate traffic flow. This segmentation allows each party to control traffic from their perspective while maintaining overall system coordination, solving the contradiction between cloud-side simplicity and enterprise-side flexibility
Solution Approach 2:
Instead of having cloud-based networks unilaterally control traffic flow (top-down approach), the patent enables enterprise networks to dictate their traffic requirements and have these requirements propagated back to cloud-based networks (bottom-up approach). The enterprise network agent receives policies from enterprise operators and transmits them to cloud agents, allowing enterprise networks to control incoming traffic while cloud networks continue to manage their outbound traffic
3Productivity
If traffic is distributed across multiple ingress points, then load balancing and congestion avoidance improve, but system complexity increases due to policy transmission and coordination requirements
Solution Approach 1:
The patent creates a universal traffic engineering framework where standardized agents with defined interfaces and protocols can be deployed at any ingress point or network boundary. These multi-functional agents handle policy reception, traffic monitoring, data collection, and routing control in a unified manner, enabling complex multi-path traffic distribution without proportionally increasing system complexity. The standardized architecture allows the same agent implementation to work across different network configurations and ingress points
Data Source
AI summary
Traffic engineering of cloud services include receiving, by an enterprise network agent of an enterprise network, a policy from an operator of the enterprise network and transmitting, by a processor, the policy to a cloud service provider. Aspects also include monitoring an incoming traffic to the enterprise network from the cloud service provider and collecting data regarding the incoming traffic and transmitting collected data regarding the incoming traffic to the cloud service provider.