Execution Mode Selection for Sustainable Function Computing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing application performance optimization techniques, such as horizontal scaling, often prioritize speed over sustainability, leading to increased costs and carbon emissions without considering user-specific needs.
Innovation Solution
A sustainable application execution framework that allows users to selectively choose between performance and sustainability modes, using a task orchestrator to adjust compute resources based on user preferences and resource utilization, enabling developers to intentionally slow down execution for designated use cases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If horizontal scaling and multi-threaded frameworks are employed to improve application performance, then execution speed is improved, but costs and carbon emissions increase
Solution Approach 1:
The system dynamically adjusts execution modes between performance mode and sustainability mode based on user preferences and real-time resource utilization conditions. The task orchestrator can switch between executing functions with maximum compute resources (performance mode) and reduced compute resources (sustainability mode), allowing the system to adapt its energy consumption profile to match actual needs rather than always operating at maximum capacity
Solution Approach 2:
The system changes the parameter of compute resource allocation by introducing execution modes that modify the amount of compute resources allocated to function execution. By adjusting this parameter based on user-selected sustainability preferences and system conditions, the system can reduce energy consumption and carbon emissions while still meeting performance requirements when sustainability mode is selected
2Productivity
If maximum compute resources are allocated to execute functions, then productivity is improved, but sustainability deteriorates
Solution Approach 1:
The task orchestrator dynamically adjusts the degree of parallelism and compute resource allocation based on the selected execution mode. In sustainability mode, the system reduces the number of concurrent threads and compute resources allocated to function execution, directly lowering productivity but also reducing carbon footprint. This dynamic adjustment allows the system to optimize the trade-off between productivity and sustainability based on user preferences
Solution Approach 2:
The system introduces a sustainability parameter that modifies compute resource allocation. When sustainability mode is selected, the parameter changes to allocate fewer compute resources to function execution, thereby reducing both productivity and carbon footprint simultaneously. This parameter change enables the system to operate at lower resource utilization levels when maximum performance is not required
3Adaptability or versatility
If execution mode selection is implemented per software component, then adaptability is improved, but device complexity increases
Solution Approach 1:
The system segments the application into individual software components or functions, allowing execution mode selection to be applied at the granular level of each component. The task orchestrator evaluates each function independently and applies the appropriate execution mode based on user preferences configured for that specific component. This segmentation enables fine-grained adaptability while keeping the complexity management localized to individual components rather than requiring system-wide configuration changes
Data Source
AI summary
Techniques are provided for function execution using selected execution modes. One method comprises obtaining information characterizing execution modes, selected by a given user, for software components of an application, wherein the selected execution modes are selected from a number of available execution modes, wherein at least two of the available execution modes are associated with different amounts of compute resources to allocate to an execution of functions associated with one or more software components; obtaining a function, of a given software component of the application, to be executed for the given user; and initiating an execution of the 10 function using the amount of compute resources associated with the selected execution mode of the given user for the given software component. The available execution modes may comprise a performance execution mode, a sustainability execution mode and/or an automatic execution mode.