Telemetry Clustering for Missed Data Troubleshooting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Troubleshooting telemetry data issues in large-scale network switches, servers, and storage systems is challenging due to missed or delayed data, making it difficult to detect problems in a timely manner across numerous devices and locations.
Innovation Solution
The technology processes time series data to identify missing and delayed telemetry data, converting it into ternary values, and applies unsupervised clustering to determine the source of issues by filtering data by customers, products, and locations, using reverse peak time series clustering and Sankey charts for visualization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If telemetry data is collected from large numbers of devices across multiple locations and customers, then the quantity of data is increased, but the difficulty of detecting and measuring problems increases
Solution Approach 1:
The patent segments the large volume of telemetry data by organizing it into groups based on device type, location, customer, and time period. This segmentation allows the system to manage and analyze data in manageable chunks rather than as one overwhelming dataset, thereby maintaining the ability to detect problems even as data quantity increases
Solution Approach 2:
The patent introduces an intermediary processing layer that aggregates and pre-processes telemetry data before analysis. This intermediary layer consolidates data from multiple sources and applies initial filtering and grouping, making the subsequent problem-detection process more efficient and less complex
2Adaptability or versatility
If data is collected from multiple failure points including gateways and networks, then the coverage is improved, but the complexity of the system increases
Solution Approach 1:
The patent implements a universal data collection framework that can handle multiple data sources (devices, gateways, networks) through a common architecture. This multi-functional system uses standardized data formats and processing routines that work across all data types, thereby expanding coverage without proportionally increasing complexity
Solution Approach 2:
The patent changes the parameters of data organization by introducing hierarchical grouping levels (device → location → customer → time period). This parameter transformation allows the system to manage complexity through structured organization rather than through complex processing logic
3Speed
If telemetry data is processed in real-time to detect problems timely, then the speed of detection is improved, but the loss of time for processing large datasets increases
Solution Approach 1:
The patent applies preliminary actions by pre-aggregating and pre-grouping telemetry data according to device type, location, and time periods before analysis is needed. This preliminary organization reduces the processing burden during actual problem detection, enabling faster identification of issues without sacrificing thoroughness
Data Source
AI summary
The described technology is generally directed towards processing time series (e.g., device telemetry) data, including identifying missing data (gaps in the time series data), and delayed data. The time series data are converted to ternary data, e.g., zero if timely, one if delayed or two if missing, and counts are obtained for each. If the missing data and/or delayed counts are significant, e.g., exceed a threshold percentage of the total data, the time series data indicates a problem that can be narrowed down to a more specific cause. For example, the time series data can be filtered by customer products/offers and customer locations, and if a filtered dataset's ternary data are similar to the problematic data, as determined via unsupervised clustering as similarity data (occurring at a similar time), the potential problem or problems can be narrowed to a potential cause based on that filtered dataset's similarity.