Cabinet Power Over-Allocation for Safe Server Deployment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data centers face inefficiencies in server deployment due to underutilized power resources in cabinets, leading to idle capacity and suboptimal resource usage rates, despite servers operating below their power ratings.
Innovation Solution
A resource tapping method utilizing deep learning and machine learning models to determine a target power over-allocation value based on operation data in multiple dimensions, allowing for additional server deployment while ensuring safety and resource constraints are met.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If servers are deployed based on cabinet power rating divided by server power rating, then power capacity is ensured, but resource utilization rate deteriorates due to underutilized power resources
Solution Approach 1:
The patent changes the parameter of power allocation from a static rating-based approach to a dynamic operation-based approach. By monitoring actual power consumption parameters of deployed servers and comparing them against cabinet power capacity, the system determines an optimal number of additional servers that can be deployed without exceeding power limits, thus improving resource utilization while maintaining power capacity reliability.
2Productivity
If additional servers are deployed to improve resource utilization, then productivity increases, but power consumption exceeds cabinet power rating leading to reliability issues
Solution Approach 1:
The patent implements a feedback mechanism by continuously monitoring the actual power consumption of deployed servers and using this information to determine the maximum number of additional servers that can be safely deployed. The system calculates the remaining power capacity based on actual consumption patterns rather than theoretical ratings, providing feedback that guides safe server deployment decisions and prevents power overload.
3Reliability
If conventional power rating division method is used for server deployment calculation, then power safety is maintained, but measurement precision of available capacity deteriorates due to ignoring actual power consumption
Solution Approach 1:
The patent performs preliminary monitoring and assessment of actual power consumption patterns before making deployment decisions. By gathering data on how servers actually consume power in operation (rather than relying on theoretical ratings), the system precariously assesses the true available capacity of the cabinet, enabling more precise determination of additional server deployment limits while maintaining safety.
Data Source
AI summary
This disclosure provides a resource tapping method, a resource tapping apparatus and an electronic device, and relates to the field of computer technology, in particular to the technical field of artificial intelligence, such as deep learning and machine learning. A specific implementation is as follows: obtaining operation data in M resource dimensions of a target cabinet, the M resource dimensions including a power resource, where M is a positive integer; determining a target power over-allocation value of the target cabinet based on the operation data, the target power over-allocation value being used for indicating an allowable power increment on the basis of a power rating of the target cabinet; and determining, based on the target power over-allocation value, a first quantity of additional servers deployable in the target cabinet.