Behavioral Sensor for Processor Event Stream Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern processors generate millions of events per second, making it difficult for devices or processes to consume the information due to the high number of events produced, and existing aggregation methods discard fine temporal information and the order of events.
Innovation Solution
A behavioral sensor that extracts features from an event stream, associates them with time, and classifies events into predefined categories, creating a consumable time series that maintains temporal fidelity and order, enabling better understanding of processor behavior by diagnostic and security systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If events are aggregated using counters or accumulators to reduce the number of events, then the quantity of events is reduced, but fine temporal information and the order of events are discarded
Solution Approach 1:
The patent segments the event stream into fixed-time windows (e.g., 1-second intervals) and processes each window independently. Within each window, events are grouped by type but their temporal sequence is preserved. This segmentation allows aggregation of event counts while maintaining the temporal structure, resolving the contradiction between reducing event quantity and preserving temporal information.
Solution Approach 2:
The patent adds a temporal dimension to the aggregated event data by organizing events into time-series structures. Instead of simply counting events, the system creates multi-dimensional data structures that include event type, count, and temporal positioning. This dimensional transformation allows both aggregation and temporal preservation simultaneously.
2Measurement precision
If all events are tracked at high speed to maintain temporal fidelity, then temporal information is preserved, but the difficulty of consuming the information increases
Solution Approach 1:
The patent divides the continuous high-speed event stream into discrete, manageable time windows. Each window contains a subset of events that can be processed and consumed independently. This segmentation reduces the cognitive and computational burden on consuming systems while preserving temporal fidelity within each window, making the information more accessible and easier to operate with.
Solution Approach 2:
The patent applies partial action by selectively processing and highlighting significant events within each time window rather than treating all events equally. The system identifies and emphasizes events that meet certain criteria (e.g., error events, performance anomalies), allowing consuming systems to focus on critical information without being overwhelmed by the complete event stream.
3Loss of time
If events are processed in real-time without aggregation, then temporal order is maintained, but the complexity of processing increases
Solution Approach 1:
The patent introduces fixed-time windows as intermediate processing units that break down the complex real-time processing task into smaller, more manageable segments. Each window independently processes its subset of events, maintaining temporal order within the window while reducing overall processing complexity through parallelization and modularization of the processing pipeline.
Data Source
AI summary
A behavioral sensor for creating consumable events can include: a feature extractor coupled to receive an event stream of events performed by a circuit, wherein the feature extractor identifies features of a particular event of the event stream and associates the particular event with a time; and a classifier coupled to receive the features of the particular event from the feature extractor, wherein the classifier classifies the particular event into a classified event associated with the time using predefined categories based on the received features of the particular event; whereby the classified event and subsequent classified events extracted from the event stream within a time frame are appended in a time series forming the consumable events.