Temporally Normalized Processor Utilization for Data Center Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current utilization data for processors does not accurately reflect demand due to variations in clock rates, leading to misrepresentation of periodicities and trends, which affects planning accuracy in data centers.
Innovation Solution
Temporal normalization of utilization traces is applied by adjusting utilization data based on clock rate changes, allowing for a more accurate representation of demand by converting un-normalized traces into normalized traces, which accounts for clock rate variations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If utilization data is collected without temporal normalization, then data collection is simple, but the accuracy of demand representation deteriorates due to clock rate variations
Solution Approach 1:
The patent replaces manual or simple collection methods with an automated normalization system that uses clock rate data to adjust utilization traces. The system automatically retrieves clock rate information, calculates normalization factors, and applies corrections to utilization data without manual intervention, thereby improving accuracy while managing complexity through automation.
Solution Approach 2:
The patent introduces clock rate data as an intermediary element that mediates between raw utilization measurements and accurate demand representation. By using clock rate as a normalization factor, the system bridges the gap between simple utilization counting and accurate demand measurement, allowing for precise demand representation through a systematic intermediary calculation process.
2Reliability
If clock rate variations are not accounted for, then data processing is simpler, but planning accuracy deteriorates due to misrepresentation of periodicities and trends
Solution Approach 1:
The patent applies temporal normalization as a preliminary action before utilization data is used for planning purposes. By pre-processing the utilization traces with clock rate-based normalization factors, the system ensures that demand patterns, periodicities, and trends are accurately represented before any planning decisions are made, thereby improving reliability in advance.
Solution Approach 2:
The system uses clock rate data as feedback to continuously adjust and normalize utilization measurements. By incorporating real-time or historical clock rate information into the normalization process, the system creates a feedback loop that ensures utilization data accurately reflects actual demand conditions, thereby improving planning reliability through continuous correction.
Data Source
AI summary
A temporally normalized processor utilization trace is generated.