Market-Based Energy Allocation in Elastic Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional networks face challenges in efficiently managing energy consumption while maintaining service level agreements (SLAs), as overprovisioning leads to idle resources and increased energy usage.
Innovation Solution
The implementation of an AI-driven elastic network that dynamically adjusts its architecture and resource allocation based on traffic demand, using a market-based approach to allocate energy-aware rights-to-operate, thereby optimizing energy consumption without compromising SLAs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If overprovisioning is used to meet user demands and SLAs, then service reliability is improved, but energy consumption increases
Solution Approach 1:
The patent implements dynamic rights-to-operate allocation where network devices can transition between operational states (fully on, partially on, fully off) based on real-time traffic demand and auction outcomes. This dynamic adjustment allows the network to maintain reliability when needed while reducing energy consumption during low-demand periods, resolving the contradiction between service reliability and energy usage.
Solution Approach 2:
The system changes the operational parameter of network devices from static (always on) to variable (rights-to-operate levels), allowing devices to operate at different power states. By adjusting these parameters based on traffic conditions and auction results, the system maintains service reliability while optimizing energy consumption.
2Use of energy by moving object
If dynamic scaling is implemented to conserve energy, then energy consumption is reduced, but network performance may deteriorate
Solution Approach 1:
The patent incorporates feedback mechanisms where the central controller monitors traffic patterns, auction outcomes, and device performance. Based on this feedback, the system adjusts rights-to-operate allocations dynamically, ensuring that energy-saving actions do not compromise network performance. The feedback loop allows continuous optimization of the energy-performance tradeoff.
Solution Approach 2:
The system performs preliminary actions by conducting auctions and allocating rights-to-operate before traffic demands materialize or before energy-saving actions are taken. This proactive approach ensures that the network is properly provisioned ahead of time, preventing performance degradation while still enabling energy savings during predictable low-demand periods.
3Loss of energy
If all devices are made eligible for downscaling, then energy-saving opportunities increase, but network reliability decreases
Solution Approach 1:
The patent applies partial action by allowing only certain devices to be eligible for downscaling based on their criticality and traffic patterns, rather than making all devices eligible. The auction mechanism selectively applies energy-saving actions to non-critical devices while protecting critical infrastructure, thus capturing energy-saving opportunities without compromising network reliability.
4Ease of operation
If the decision process for device downscaling is simplified, then operational ease is improved, but energy optimization capability decreases
Solution Approach 1:
The patent implements self-service by allowing network devices to participate in auctions and bid for their own rights-to-operate based on their operational needs and traffic patterns. Each device autonomously evaluates its own criticality and makes informed decisions about its operational state, simplifying the overall decision-making process while maintaining sophisticated energy optimization through market-based mechanisms.
Data Source
AI summary
In one implementation, a device assigns a budget to a bidding agent associated with a networking entity in a computer network. The device obtains a bid from the bidding agent selected by the bidding agent based on the budget. The device conducts an auction using the bid. The device excludes the networking entity from being eligible for performance of an energy-saving action, based on a result of the auction.