Data Center Workload Migration for Power Variability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data centers face challenges with growing energy demands, high costs, inefficiencies, and variability in renewable power sources, which can lead to inadequate power supply and increased operational expenses, particularly due to episodic nature of renewable energy sources like wind and solar.
Innovation Solution
A data center system utilizing a management system that dynamically adjusts the execution of computational tasks by moving data components to locations with available and cost-effective power sources, incorporating temporal and spatial movement of data components, and leveraging a combination of renewable and traditional power sources to optimize resource utilization and pricing based on supply and demand curves.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data centers rely on on-grid power sources, then power supply stability is improved, but power costs increase and energy efficiency deteriorates
Solution Approach 1:
The patent implements dynamic workload migration between data centers based on real-time power availability and cost conditions. The management system continuously monitors power sources and adjusts computational task distribution dynamically, moving workloads from high-cost/low-efficiency data centers to low-cost/high-efficiency data centers, thereby resolving the contradiction between power supply stability and energy efficiency.
Solution Approach 2:
The system changes operational parameters by adjusting workload distribution across multiple data centers with different power source compositions. By modifying the parameter of workload location and timing, the system optimizes the balance between reliable power supply and energy efficiency, selecting optimal data centers based on current power conditions.
2Loss of energy
If data centers use renewable power sources, then power costs decrease and energy efficiency improves, but power supply reliability deteriorates due to episodic availability
Solution Approach 1:
The patent applies preliminary action by performing computational tasks in advance when renewable power is available and storing results, so that tasks can be completed even when power is later unavailable. The system predicts power availability and proactively executes workloads during favorable conditions, thereby maintaining reliability while utilizing efficient renewable energy sources.
Solution Approach 2:
The management system acts as an intermediary between renewable power sources and computational workloads. It buffers the episodic nature of renewable power by coordinating workload execution with power availability, using prediction algorithms and scheduling mechanisms to match computational demands with intermittent supply, thus resolving the reliability-efficiency contradiction.
3Reliability
If data centers operate continuously to meet service demands, then service availability is improved, but operational costs increase due to idle resources during power unavailability
Solution Approach 1:
The patent segments the data center system into multiple independent data centers with different power source profiles. This segmentation allows the system to distribute workloads across different locations, ensuring that while one data center may be idle due to power unavailability, others can continue operating, thereby maintaining overall service availability while improving operational efficiency.
Solution Approach 2:
The system implements multi-functionality by enabling data centers to serve different purposes at different times based on power availability. Data centers can alternately handle primary workloads, backup workloads, or remain idle depending on conditions, creating a flexible universal system that maintains service availability while optimizing operational efficiency across the distributed infrastructure.
4Loss of energy
If data centers migrate computational tasks to accommodate episodic power, then energy costs decrease, but system complexity increases due to task movement management
Solution Approach 1:
The patent implements self-service by enabling data centers to automatically monitor their own power conditions and make migration decisions without extensive external coordination. Each data center reports its power status and workload capacity to the management system, which then autonomously determines optimal task distribution, reducing the complexity burden while achieving energy cost savings.
Solution Approach 2:
The system uses feedback mechanisms where data centers continuously report power availability and workload status to the management system. This feedback loop enables dynamic, automated workload migration decisions based on real-time conditions, managing system complexity through structured information flow and automated response rather than manual intervention.
Data Source
AI summary
A data center system is described which includes multiple data centers powered by multiple power sources, including any combination of renewable power sources and on-grid utility power sources. The data center system also includes a management system for managing execution of computational tasks by moving data components associated with the computational tasks within the data center system, in lieu of, or in addition to, moving power itself. The movement of data components can involve performing pre-computation or delayed computation on data components within any data center, as well as moving data components between data centers. The management system also includes a price determination module for determining prices for performing the computational tasks based on different pricing models. The data center system also includes a “stripped down” architecture to complement its use in the above-summarized data-centric environment.