Cloud Component Health Model Prevents Cascading Failures

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Cloud-based applications are vulnerable to increased failure points across multiple cloud nodes, leading to potential cascading failures that can disrupt service availability and are difficult to diagnose and manage effectively.

Innovation Solution

A computer system models cloud components as health entities, establishes declarative safety conditions, and uses guarded interfaces to detect and prevent cascading failures by identifying and isolating errors, notifying users of the affected component, and ensuring that recovery actions do not exacerbate the issue.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If cloud-based applications are distributed across multiple cloud nodes, then functionality and scalability are improved, but the number of failure points increases leading to potential cascading failures

Engineering Contradiction:

Improvecloud application distribution capabilityVSAvoidsystem failure resistance

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The system segments the cloud infrastructure into discrete health entities (hardware components, software components, virtual machines, containers) that can be independently monitored and managed. Each component has its own health model that tracks specific safety conditions, allowing failures to be contained to individual segments rather than propagating system-wide.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary health monitoring system that acts as a mediator between cloud components and users. This intermediary establishes declarative safety conditions and continuously tests them, providing a buffer that detects and reports failures before they can cause cascading effects across the distributed system.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Difficulty of detecting and measuring

If comprehensive monitoring of cloud components is implemented, then failure detection capability is improved, but system complexity and diagnostic difficulty increase

Engineering Contradiction:

Improvefailure detection capabilityVSAvoidmonitoring system complexity

Core Design Contradiction:

Difficulty of detecting and measuringVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by establishing declarative safety conditions in advance that define what constitutes a failure state for each health entity. These safety conditions are predefined based on component characteristics and operational requirements, so when monitoring begins, the system already has clear criteria for detecting failures without needing complex real-time analysis.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent transforms the monitoring approach by changing parameters from complex multi-variable analysis to simple binary safety condition checks. Each health entity monitors specific parameters against predefined safety thresholds, converting complex diagnostic problems into straightforward parameter comparisons that are easy to implement and interpret.

Inventive Principle:

Principle #35Parameter changes

3Reliability

If rapid error notification and isolation are implemented, then system reliability is improved, but response time requirements increase system operational complexity

Engineering Contradiction:

Improvecascading failure preventionVSAvoiderror response time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system prepares isolation mechanisms in advance by establishing safety conditions that automatically trigger isolation actions when violated. The declarative safety conditions include predefined response actions, so when a failure is detected, the system can immediately execute pre-planned isolation procedures without requiring complex real-time decision-making or manual intervention.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS8996932B2Cloud management using a component health model

Publication Date: 2015.03.31 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

Embodiments are directed to establishing a model for testing cloud components and to preventing cascading failures in cloud components. In one scenario, a computer system models identified cloud components (including cloud hardware components and/or cloud software components) as health entities. Each health entity is configured to provide state information about the cloud component. The computer system establishes declarative safety conditions which declaratively describe cloud computing conditions that are to be maintained at the identified cloud components. The computer system then tests against the declarative safety conditions to determine which cloud components are or are becoming problematic. Upon determining that an error has occurred, the computer system notifies users of the error and the component at which the error occurred. Guarded interfaces are established to ensure that actions taken to fix the error do not cause further failures.