Joint Traffic Engineering and Server Selection System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traffic engineering (TE) and adaptive server selection (SS) processes in networks often conflict due to different time granularities and objectives, leading to inefficiencies in content delivery and end-user experience.
Innovation Solution
A joint TE and SS system that operates at multiple time scales, with a global TE component determining network paths and a local TE component handling link-level congestion, communicating through a SS component to optimize traffic routing and content distribution, using a receiver, logic circuit, and transmitter to compute and signal routing matrices across network nodes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If TE process designates paths on a daily or hourly basis to transport service efficiently, then network load balancing is improved, but SS process cannot respond quickly to changing content demands on a minutes or seconds basis
Solution Approach 1:
The system segments the traffic management function into two independent components: TE component that operates on a coarse time scale (daily/hourly) for load balancing, and SS component that operates on a fine time scale (minutes/seconds) for content delivery. This segmentation allows each component to operate at its optimal time granularity without interfering with the other, resolving the contradiction between load balancing efficiency and response speed.
Solution Approach 2:
The patent introduces a joint TE-SS optimization component that acts as an intermediary, coordinating the operations of TE and SS processes. This mediator enables the SS process to utilize multiple paths designated by TE while adapting quickly to content demands, thus maintaining both load balancing efficiency and fast response capability.
2Speed
If SS process obtains content from servers via multiple paths not designated by TE, then content delivery speed is improved, but network congestion and bottlenecks increase
Solution Approach 1:
The system dynamically adapts the SS process to operate within the constraints of TE-designated paths. The SS component can flexibly select from multiple available paths on a fine time scale, but these selections are guided by the broader load balancing objectives set by TE. This dynamic coordination allows fast content delivery while preventing network congestion through intelligent path selection.
Solution Approach 2:
The joint TE-SS optimization implements feedback mechanisms where SS performance metrics are fed back to the TE component. This feedback loop allows the system to learn from actual content delivery patterns and adjust TE path designations accordingly, enabling SS to deliver content quickly while TE prevents congestion through informed routing decisions.
3Adaptability or versatility
If TE and SS operate independently at different time granularities, then each process can optimize its own objectives, but joint optimization for service delivery is reduced
Solution Approach 1:
The patent merges the TE and SS optimization objectives into a unified joint TE-SS optimization framework. This combined approach allows both processes to maintain their operational independence and time scale differences while working toward a common goal of optimizing service delivery. The joint optimization coordinates path selection and content delivery to achieve better overall performance than independent operation.
Data Source
AI summary
An apparatus comprising a traffic engineering (TE) and server selection (SS) component configured to couple to an SS component and operate at a first time scale based on SS policies from the SS component, wherein the SS is configured to operate at a second timescale based on TE policies from the TE and SS component, and wherein the second timescale has a finer time granularity than the first time scale by at least one order of magnitude. Also disclosed is a network component comprising a receiver configured to receive one or more SS policies from a content provider or customer node, a logic circuit configured to compute a plurality of joint TE and SS policies for a plurality of network nodes based on the received SS policies, and a transmitter configured to send the joint TE and SS components to the content provider or customer node.