Service-Level Latency Anomaly Detection for Cloud Provisioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network-accessible computing platforms face challenges in quickly detecting and diagnosing latency anomalies, leading to inefficient resource usage and user interface delays due to unsatisfactory latency-related performance during provisioning actions like creating or deleting virtual machines.
Innovation Solution
A technique for detecting latency-related anomalies on a service-level granularity, using a machine-trained language model to confirm incidents and adjust control settings, generating latency profile information, and consolidating reports to reduce false positives, thereby expediting anomaly resolution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If operation-level granularity reporting is implemented, then detailed diagnostic information is provided, but the volume of reports becomes unmanageable and consumes excessive system resources
Solution Approach 1:
The patent segments the monitoring system into two distinct layers: operation-level monitoring that collects detailed diagnostic data, and service-level reporting that presents aggregated, actionable information. This segmentation allows the system to maintain comprehensive diagnostic capabilities while delivering manageable report volumes to users.
Solution Approach 2:
The patent introduces a hierarchical dimension to the reporting structure, organizing reports from operation-level details up to service-level summaries. This dimensional organization transforms the flat, overwhelming volume of operation-level reports into a structured hierarchy where only essential service-level anomalies are presented to users.
2Ease of operation
If service-level granularity reporting is implemented, then actionable insights are provided, but detailed diagnostic information for specific operations is reduced
Solution Approach 1:
The patent segments information delivery by function: service-level reports provide actionable insights for operational decision-making, while operation-level reports provide detailed diagnostic information for technical troubleshooting. Each segment serves its specific purpose without interfering with the other.
Solution Approach 2:
The patent introduces service-level anomalies as an intermediary layer between operation-level details and user decision-making. This intermediary aggregates and filters operation-level data into meaningful service-level insights, making the information both actionable and diagnostically useful.
3Measurement precision
If comprehensive anomaly detection is implemented, then detection accuracy is improved, but false positives increase requiring manual verification
Solution Approach 1:
The patent performs preliminary filtering and aggregation of operation-level anomalies before presenting them as service-level anomalies. This preliminary action reduces false positives by combining multiple operation-level signals into cohesive service-level patterns that are more reliable indicators of actual problems.
Solution Approach 2:
The patent implements feedback mechanisms where service-level anomaly detection results are used to adjust operation-level monitoring parameters. This feedback loop continuously improves detection accuracy while reducing false positives through learned patterns from historical data.
4Measurement precision
If manual analysis of anomalies is performed, then detection thoroughness is improved, but time to resolve provisioning actions increases
Solution Approach 1:
The patent enables the system to perform self-service anomaly analysis by automatically detecting service-level anomalies from operation-level data without requiring manual intervention. This self-service capability maintains thorough analysis while dramatically reducing the time needed to identify and respond to provisioning action anomalies.
Solution Approach 2:
The patent performs preliminary anomaly analysis automatically at the service-level before escalating to manual investigation. This preliminary action filters out many anomalies that can be automatically resolved, reserving manual analysis only for complex cases that truly require human expertise.
Data Source
AI summary
A technique detects latency-related anomalies that occur in performing provisioning actions in a network-accessible computing platform. Illustrative provisioning actions include creating a resource (e.g., a virtual machine), updating the resource, and deleting the resource. The technique operates by detecting and reporting the latency-related anomalies on a service-level granularity. The technique also provides access to operation-level latency-related data to assist in diagnosing the causes of the anomalies. In some implementations, the technique consolidates incidents that potentially reveal a common source of failure into a single report. In some implementations, the technique interacts with a machine-trained language model to confirm whether a reported incident is a false positive which does not warrant further action. The technique performs this function by sending a prompt to the language model that expresses a current incident and one or more prior incidents that have been determined to match the current incident.