Workload Management via Carbon Intensity and Governance Policies
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cloud computing systems fail to optimize workload management effectively, leading to suboptimal performance and increased energy consumption due to neglect of real-time carbon intensity and governance policies, as well as complex batch job interdependencies.
Innovation Solution
A system that identifies candidate execution environments based on governance policies and carbon intensity values, generating recommendation information for workload processing to minimize carbon emissions and ensure compliance, using a combination of carbon intensity extract, discovery, recommendation, and carbon savings engines.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-generated harmful factors
If existing workload management systems are used, then basic resource allocation is achieved, but carbon emissions are not optimized and governance policies are not integrated
Solution Approach 1:
The system segments workload management into multiple independent modules: carbon intensity extraction module, discovery module, recommendation module, and carbon savings module. Each module handles a specific aspect of the problem, allowing the system to integrate external factors (carbon intensity, governance policies) without compromising overall functionality. This modular architecture enables the system to address carbon emissions while maintaining adaptability to various external constraints.
Solution Approach 2:
The patent introduces intermediary components that bridge internal workload management with external factors. The carbon intensity extract acts as an intermediary that fetches external carbon intensity data, while the discovery module serves as an intermediary that identifies suitable execution environments based on both internal workload requirements and external governance policies. These intermediaries enable integration of external factors without directly coupling the workload scheduler to external data sources.
2Productivity
If workload scheduling is optimized for performance, then processing speed is improved, but energy consumption and carbon footprint increase
Solution Approach 1:
The system dynamically adjusts workload scheduling decisions based on real-time carbon intensity values and governance policy constraints. Rather than using static scheduling rules, the recommendation module continuously evaluates candidate execution environments against current carbon intensity data and policy requirements, enabling the system to optimize for low carbon emissions while maintaining acceptable processing speeds. This dynamic approach allows flexibility in balancing performance and energy consumption.
Solution Approach 2:
The patent changes the optimization parameters from purely performance-based metrics to include carbon intensity and governance policy compliance. The recommendation module evaluates execution environments based on multiple parameters including carbon intensity values, policy constraints, and performance metrics. By incorporating carbon intensity as a key parameter in the scheduling decision process, the system can reduce energy consumption and carbon footprint while still meeting performance requirements through the balanced evaluation of multiple factors.
3Device complexity
If batch job scheduling is simplified, then scheduling complexity is reduced, but batch interdependencies are not properly managed
Solution Approach 1:
The discovery module performs preliminary analysis of batch job interdependencies and execution environment compatibility before final scheduling decisions are made. By pre-identifying suitable execution environments and their capabilities, the system can simplify the actual scheduling process while ensuring that batch job dependencies are properly managed. This preliminary action allows the recommendation module to make simpler scheduling decisions based on pre-validated environment selections, reducing scheduling complexity without compromising reliability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Aspects of the present disclosure provide systems, methods, and computer-readable storage media that support workload management. A computing device may generate recommendation information for processing of the workload, the recommendation information associated with a first execution environment selected from the set of candidate execution environments based on execution information that indicates, for each candidate execution environment of the set of candidate execution environments, a carbon intensity value associated with the candidate execution environment. The computing device may output a first indicator that indicates the recommendation information.