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 during provisioning actions, leading to inefficient resource usage and user interface delays.
Innovation Solution
A technique for detecting latency-related anomalies on a service-level granularity, utilizing a machine-trained language model to confirm incidents and consolidate reports, reducing false positives, and providing operation-level data for diagnosis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If anomaly detection is performed at operation-level granularity, then detection precision is improved, but the quantity of reports increases voluminously creating noise and wasting system resources
Solution Approach 1:
The patent merges multiple operation-level anomaly reports into consolidated service-level reports. The service-level anomaly report aggregates anomalies from multiple operations within the same service, reducing the total number of reports while preserving detection precision through service-level attribution.
Solution Approach 2:
The service-level anomaly report serves multiple functions simultaneously: it provides anomaly detection, service-level localization, and consolidation of multiple operation-level issues into a single actionable report, reducing noise while maintaining diagnostic value.
2Ease of operation
If service-level reporting is implemented, then ease of operation is improved by reducing report volume, but manufacturing precision deteriorates by losing operation-level detail
Solution Approach 1:
The service-level anomaly report contains nested operation-level anomaly information within it. Each service-level report aggregates multiple operation-level anomalies, allowing operators to drill down from service-level overview to operation-level details when needed, thus maintaining both ease of operation and diagnostic precision.
3Reliability
If machine-trained language model is used to filter false positives, then reliability is improved by reducing false alarms, but use of energy increases due to additional processing
Solution Approach 1:
The language model is applied selectively rather than universally - it processes only service-level anomaly reports that meet certain criteria or require verification, rather than every single anomaly report. This partial application reduces energy consumption while maintaining reliability where it matters most.
Data Source
Figure 1
Figure 2
Figure 3
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.