Cloud Load Optimization via Thermal and Power Data Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud solutions lack a mechanism to retrieve thermal and power information from hardware in a RESTful, cloud-friendly manner, making it difficult for hybrid cloud systems to access on-premises hardware information for intelligent load deployment.
Innovation Solution
A computer-implemented method that receives data from hardware abstract layers (HAL) associated with servers in multiple data centers, filters data based on highest power usage and thermal state conditions, aggregates the data, creates two-ratio statistics, generates a data center score, selects sites with the lowest scores, and initiates a request to transfer server-based loads to those sites.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If cloud infrastructure abstracts hardware layer to relieve customer concerns, then customer ease of operation is improved, but access to hardware information for hybrid cloud deployment is lost
Solution Approach 1:
The patent introduces a hardware abstraction layer (HAL) as an intermediary component that sits between the cloud infrastructure and hardware resources. The HAL collects, standardizes, and exposes hardware information (power usage, thermal data, inventory) through uniform APIs, allowing customers to access detailed hardware information without dealing with low-level hardware complexity directly. This resolves the contradiction by maintaining operational simplicity while enabling information access.
Solution Approach 2:
The patent segments the hardware information access into distinct functional layers: the HAL layer that collects and processes raw hardware data, the API layer that standardizes access patterns, and the application layer that consumes the information. This segmentation allows each layer to handle specific tasks, maintaining overall system simplicity while enabling detailed hardware information access where needed.
2Device complexity
If cloud solutions lack thermal and power information retrieval mechanism, then system complexity is reduced, but intelligent load deployment capability is impaired
Solution Approach 1:
The hardware abstraction layer automatically collects, processes, and makes available hardware information (power usage, thermal conditions, inventory status) without requiring manual configuration or complex system integration. The HAL performs self-service functions by continuously monitoring hardware and exposing information through standardized APIs, enabling intelligent load deployment while keeping the system architecture simple.
Solution Approach 2:
The patent creates a universal hardware abstraction layer that can access and standardize information from diverse hardware sources (different data centers, various sensor types, multiple hardware vendors) through a single interface. This multi-functional HAL enables intelligent load deployment across heterogeneous hardware environments without increasing overall system complexity, as the same abstraction mechanism serves multiple purposes.
Data Source
AI summary
An approach for optimizing server-based loads between data centers. The approach receives data from a hardware abstraction layer (HAL) associated with servers in a plurality of data centers. The approach filters the data associated with a portion of the data centers having the highest power usage and thermal state conditions. The approach aggregates the filtered data into performance data groups based on association with a data center. The approach creates two-ratio statistics of the aggregated groups. The approach generates a data center score based on the two-ratio statistics. The approach selects data center sites with the lowest scores. The approach initiates a request to transfer server-based loads from the servers associated with the filtered data to the data center sites with the lowest scores.