Workload Shifter for IT Resource Utilization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing capacity planning and server consolidation tools treat workload demands as fixed and do not consider time-shifting or broader IT resources like energy sources and sustainability factors when mapping workloads to IT resources, leading to inefficiencies and environmental impacts.
Innovation Solution
The proposed solution involves shifting workload demands in time and considering a broader range of IT resources, including energy sources and sustainability factors, to determine optimal workload assignments, using methods like stochastic processes and algorithms to prioritize and reassign workloads dynamically.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If workload demand profiles are treated as fixed and mapped directly to IT resources, then the mapping process is simple and deterministic, but resource utilization efficiency is suboptimal and environmental impacts increase
Solution Approach 1:
The patent transforms fixed workload demand profiles into dynamic, shiftable profiles by introducing temporal flexibility. Workloads are no longer mapped to fixed time slots but can be shifted across different time periods based on resource availability and environmental conditions, enabling optimized resource utilization while maintaining manageable complexity through systematic evaluation methods.
Solution Approach 2:
The patent changes the temporal parameter of workload execution by introducing workload shifting capabilities. Instead of executing workloads at fixed times, the system allows adjustment of execution timing parameters based on resource capacity and environmental factors, thereby improving resource utilization efficiency without significantly increasing operational complexity.
2Object-affected harmful factors
If traditional capacity planning tools are used that only consider IT resources, then the planning process is straightforward, but environmental impacts and operational costs are not optimized
Solution Approach 1:
The patent extends the functionality of capacity planning tools by integrating multiple evaluation dimensions beyond traditional IT resources. The system simultaneously evaluates environmental impacts, operational costs, and resource utilization, creating a multi-functional planning approach that addresses multiple objectives without requiring separate planning processes.
Solution Approach 2:
The patent introduces an intermediary evaluation layer that mediates between workload requirements and resource allocation decisions. This intermediary system considers environmental and cost factors alongside technical requirements, translating multiple competing objectives into optimized workload assignments through systematic evaluation rather than direct mapping.
3Loss of energy
If workloads are shifted in time to optimize resource utilization, then resource efficiency improves, but the complexity of workload management and assignment increases
Solution Approach 1:
The patent applies preliminary action by pre-evaluating multiple workload assignment configurations before final deployment. The system systematically assesses different time-shifting scenarios and their impacts on energy efficiency and resource utilization, allowing optimized assignments to be determined in advance rather than through complex real-time adjustments.
Solution Approach 2:
The patent incorporates feedback mechanisms that evaluate the effectiveness of workload shifting decisions. By monitoring resource utilization and environmental impacts, the system provides feedback on assignment outcomes, enabling continuous optimization of workload timing while maintaining manageable complexity through data-driven adjustments.
Data Source
AI summary
Methods, apparatus and articles of manufacture to shift information technology workload demands are disclosed. An example method disclosed herein comprises obtaining demand profiles for workloads in an information technology infrastructure, each demand profile representing resource utilization over time to support a respective workload, and evaluating the demand profiles to shift, in time, assignment of at least one of the workloads to resources of the information technology infrastructure based on a target demand profile for an aggregation of the workloads.