Processor Capacity Measurement Using CPU Time Units
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for measuring CPU usage in computer systems are inadequate when dealing with processors of different speeds and architectures, as they cannot accurately compare or sum performance across varying clock speeds and internal circuitry configurations.
Innovation Solution
Assigning processors to classes based on type and architecture, using conversion factors to scale performance into a common 'CPU time unit' for accurate capacity and usage measurement and reporting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If processors of different speeds and architectures are measured using absolute time units (processor-seconds), then the measurement method remains simple and consistent, but the accuracy of performance comparison and capacity reporting deteriorates because different processor-seconds are not equivalent
Solution Approach 1:
The patent applies parameter changes by introducing a new measurement unit (CPU time unit) that incorporates processor speed and architecture characteristics into the measurement parameters. Instead of using fixed time units, the system dynamically adjusts the measurement parameters based on processor class and conversion factors, allowing accurate comparison across different processor types while maintaining a unified measurement framework.
Solution Approach 2:
The patent uses conversion factors as an intermediary element between different processor measurements and the common CPU time unit. These conversion factors act as mediators that translate various processor-seconds into equivalent CPU time units, enabling accurate aggregation and comparison of processor capacity across heterogeneous processors without requiring direct comparison of dissimilar units.
2Ease of operation
If multiple processors of different types are aggregated into a single capacity metric, then the reporting is simplified, but the reliability of capacity planning deteriorates due to erroneous comparisons and summing of incompatible processor units
Solution Approach 1:
The system changes the measurement parameter from simple time units to class-specific CPU time units with conversion factors. This allows processors to be grouped by class (maintaining ease of operation) while applying appropriate conversion factors to each class (ensuring reliability). The patent reports capacity by processor class using the formula: total capacity = sum of (processor count × CPU time units × conversion factor) for each class, which preserves both simplicity and accuracy.
3Measurement precision
If processor capacity is measured in processor-seconds without considering clock speed variations, then the measurement approach remains straightforward, but the measurement precision deteriorates because one second on different processors represents different amounts of work
Solution Approach 1:
The patent transforms the measurement parameter from fixed time units to dynamic CPU time units that incorporate processor performance characteristics. By defining CPU time units based on processor class and applying conversion factors that reflect clock speed and architecture differences, the system achieves precise measurement of CPU resource consumption while accounting for variations in processor capabilities.
Solution Approach 2:
The patent creates equipotentiality by normalizing all processor measurements to a common reference frame (CPU time units with conversion factors). This allows different processor types to be compared on an equal footing, where each processor class has its measurements scaled to represent equivalent computational capacity, eliminating the inequality caused by different clock speeds and architectures.
Data Source
AI summary
In a computer system that includes multiple processors, each processor in a computer system is assigned a processor class. Processor capacity and usage are monitored according to the class assigned to the processor. Capacity and usage are reported on a class-by-class basis so that the capacity and performance of different classes of processors are not erroneously compared or summed. The capacity and usage are monitored and reported in an abstract unit of measurement referred to as a “CPU time unit”. Processors of the same type that run at different clock speeds or that have different internal circuitry enabled are preferably assigned the same class, with one or more conversion factors being used to appropriately scale the performance of the processors to the common CPU time unit for this class.