Dynamic Application Placement for Datacenter Energy Cost Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Energy costs in datacenters are rising due to the increasing scale of operations, and existing technologies fail to efficiently manage energy usage across multiple hosting sites with varying energy costs and availability, leading to substantial expenditures.
Innovation Solution
A dynamic optimization framework that places application instances across multiple hosting sites based on energy costs, availability, and network bandwidth, using a closed-loop control system integrated with a reliable language runtime mechanism to migrate applications between datacenters in response to energy availability and pricing variations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If application instances are placed in datacenters with lower energy costs, then energy expenditure is reduced, but network bandwidth costs and migration complexity increase
Solution Approach 1:
The patent implements dynamic application placement that continuously monitors and responds to changing energy costs, availability, and network conditions. The system automatically relocates application instances between datacenters based on real-time optimization criteria, transforming a static placement problem into a dynamic adaptive system that resolves the contradiction between energy savings and operational complexity.
Solution Approach 2:
The optimization framework incorporates feedback mechanisms that monitor energy costs, availability, and network bandwidth consumption. This feedback drives automated decision-making about application placement and migration, allowing the system to learn from past decisions and continuously improve its optimization strategy while managing complexity through data-driven control.
2Loss of energy
If application instances are frequently migrated between datacenters to optimize energy costs, then energy expenditure is reduced, but service level agreement compliance and system reliability may deteriorate
Solution Approach 1:
The system performs preliminary assessments of migration impacts before executing relocation decisions. It evaluates whether proposed migrations would violate service level agreements or compromise reliability, and only approves migrations that meet all constraints. This preventive approach ensures energy optimization does not come at the cost of service quality.
Solution Approach 2:
The optimization framework dynamically adjusts placement decisions based on multiple parameters including energy cost, availability, network bandwidth, and service level agreement constraints. By considering the full spectrum of parameters simultaneously, the system finds optimal placement configurations that balance energy savings with reliability requirements.
3Loss of energy
If a global network of hosting sites is used to optimize energy placement, then energy cost reduction is improved, but system complexity and coordination overhead increase
Solution Approach 1:
The global network of hosting sites is segmented into independent but coordinated units, each capable of autonomous decision-making within its local context. The optimization framework divides the complex global optimization problem into manageable sub-problems that can be solved independently and then integrated, reducing overall system complexity while maintaining global optimization benefits.
Solution Approach 2:
The patent creates a universal optimization framework that can be applied across diverse datacenter environments with different energy costs, availability characteristics, and network conditions. This multi-functional system handles various placement scenarios, migration types, and optimization objectives through a unified approach, reducing complexity by avoiding environment-specific customizations.
Data Source
AI summary
An optimization framework for hosting sites that dynamically places application instances across multiple hosting sites based on the energy cost and availability of energy at these sites, application SLAs (service level agreements), and cost of network bandwidth between sites, just to name a few. The framework leverages a global network of hosting sites, possibly co-located with renewable and non-renewable energy sources, to dynamically determine the best datacenter (site) suited to place application instances to handle incoming workload at a given point in time. Application instances can be moved between datacenters subject to energy availability and dynamic power pricing, for example, which can vary hourly in day-ahead markets and in a time span of minutes in realtime markets.