Cloud Health Checking Entity for Low-Latency Failure Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing cloud systems face challenges in detecting faults quickly enough, especially in latency-critical environments like industrial IoT and telecom, where delayed detection can lead to physical damage or user disruptions, as current monitoring solutions operate on a few seconds scale and may not detect faulty states without crashes.

Innovation Solution

A method involving a health checking entity that passively monitors health reports from applications, triggering active monitoring only when anomalies are detected, allowing for low-latency failure detection with low overhead, and enabling application-specific recovery mechanisms by informing control entities of incorrect operation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If periodical status message exchange (Keepalive or heartbeat) is used to monitor node liveness, then failure detection speed improves, but monitoring resource consumption increases

Engineering Contradiction:

Improvefailure detection timeVSAvoidmonitoring resource consumption

Core Design Contradiction:

Loss of timeVSUse of energy by moving object

Solution Approach 1:

The system uses periodic health report transmission from applications combined with periodic monitoring by the health checking entity. This periodic action allows the system to achieve fast failure detection through regular status checks while optimizing resource consumption by only triggering active monitoring when anomalies are detected in the periodic reports, rather than maintaining continuous high-frequency monitoring.

Inventive Principle:

Principle #19Periodic action

2Use of energy by moving object

If existing cloud monitoring solutions operate on minute scale checking CPU and resource usage, then system overhead is reduced, but failure detection latency increases

Engineering Contradiction:

Improvesystem overheadVSAvoidfailure detection latency

Core Design Contradiction:

Use of energy by moving objectVSLoss of time

Solution Approach 1:

The monitoring system dynamically adjusts its operation mode between passive monitoring of health reports and active monitoring. The system starts in a low-overhead passive mode, then transitions to active monitoring when anomalies are detected, and can return to passive mode after recovery. This dynamic adaptation allows the system to maintain low overhead during normal operation while enabling fast detection when needed.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

Applications self-report their health status through periodic health reports containing information about their operation status. This self-service mechanism reduces the overhead on the monitoring system, as applications autonomously provide their own status information rather than requiring external probing of CPU and resource usage.

Inventive Principle:

Principle #25Self-service

3Measurement precision

If monitoring systems establish network connections periodically to test application functionality, then fault detection accuracy improves, but reaction time increases due to centralized report collection

Engineering Contradiction:

Improvefault detection accuracyVSAvoidsystem reaction time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The monitoring function is segmented into distributed health checking entities that can operate independently on individual nodes. Each health checking entity monitors local applications and can trigger local active monitoring and recovery actions without waiting for centralized report collection and decision-making. This segmentation enables both accurate fault detection through active monitoring and fast local reaction times.

Inventive Principle:

Principle #1Segmentation

4Loss of time

If supervision solutions subscribe to kernel events for immediate crash notification, then crash detection speed improves, but detection of faulty states without crashes fails

Engineering Contradiction:

Improvecrash detection speedVSAvoidfaulty state detection capability

Core Design Contradiction:

Loss of timeVSReliability

Solution Approach 1:

The health report mechanism serves multiple functions: it provides regular status updates for passive monitoring, enables detection of faulty states through application-reported errors, and triggers active monitoring when anomalies are detected. This multi-functional approach allows the system to detect both crashes and non-crash faulty states using the same health report infrastructure.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS11334468B2Checking a correct operation of an application in a cloud environment

Publication Date: 2022.05.17 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

AI summary

The method relates to a method for checking a correct operation of an application running in a cloud environment, the method comprising at a health checking entity, monitoring a reception of health reports generated by the application, each health report comprising at least information allowing an operation status of the application to be determined, determining whether an anomaly is detected in the reception of the health reports, wherein, when an anomaly is detected in the reception, triggering a transmission of a new health report to be transmitted by the application, determining whether the application is operating correctly after triggering the transmission of the new health report, wherein, if the application is not operating correctly after triggering the transmission of the new health report, informing a control entity of the application that the application is not operating correctly.