Automated Virtual Instance Restart with Entitlement Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In Cloud computing environments, existing solutions fail to effectively manage virtual instance failures, leading to unavailability of resources, unauthorized access, and unauthorized costs, as they assume a level of trust that is not guaranteed in public Clouds and do not adequately monitor or restart instances across physical nodes.
Innovation Solution
A system that automates virtual instance restarts by detecting failed instances, authenticating user entitlement, and using a countdown timer to initiate a new instance, integrating with OSS and BSS for monitoring and charging, allowing for pay-as-you-go pricing and ensuring secure resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple system instances are deployed to monitor each other for failure detection, then instance failure detection capability is improved, but system complexity and resource consumption increase
Solution Approach 1:
The virtual instance performs self-monitoring through health check endpoints and automatic failure detection mechanisms. Instead of requiring multiple instances to monitor each other, the system enables instances to autonomously detect their own failure states and trigger restart workflows, eliminating the need for complex inter-instance monitoring infrastructure.
Solution Approach 2:
An intermediary orchestration service is introduced to manage failure detection and restart coordination. This central mediator handles the complexity of monitoring multiple instances, processing failure notifications, and coordinating restart operations, thereby reducing the complexity burden from individual instances while maintaining reliable failure detection across the system.
2Reliability
If virtualized view abstracts physical assignment information, then security and multi-tenancy are improved, but physical node failure monitoring capability deteriorates
Solution Approach 1:
The system adds a new abstraction dimension by introducing virtual instance identifiers and logical node concepts that operate independently of physical hardware assignments. Failure detection and restart operations are performed at this virtualization layer, allowing the system to maintain security through abstraction while enabling effective monitoring and management of instance lifecycles without exposing physical node details.
3Productivity
If automated restart service is implemented without entitlement verification, then service speed and automation are improved, but security and unauthorized access prevention deteriorate
Solution Approach 1:
Entitlement verification and authentication checks are performed in advance as preliminary actions before the automated restart process is initiated. The system validates user permissions, verifies instance ownership, and authorizes restart requests before executing the actual restart operation, ensuring security requirements are met while maintaining automation efficiency through pre-validation.
Solution Approach 2:
The system implements feedback mechanisms that continuously verify entitlement and authorization status throughout the restart workflow. Authentication tokens are validated, permission checks are performed at critical decision points, and the orchestration service receives feedback regarding authorization status to dynamically adjust the restart process, thereby maintaining both security and automation.
4Reliability
If countdown timer with threshold is used before instance restart, then unauthorized operation prevention is improved, but instance availability time deteriorates
Solution Approach 1:
The system dynamically adjusts the countdown timer threshold parameter based on instance priority, failure type, and service level agreements. Critical instances may have shorter or zero-wait thresholds for immediate restart, while non-critical instances use longer thresholds to prevent unauthorized operations. This parameter flexibility allows the system to balance security prevention with minimizing instance availability loss according to specific operational requirements.
Data Source
AI summary
Embodiments of the present invention provide a system that leverages the Operational Support System(s) (OSS) and Business Support system(s) (BSS) of a (e.g., public) computing Cloud with a service to automate virtual instance restarts. For example, under embodiments of the present invention, a failed virtual instance is detected within the Cloud computing environment, and a request for a new virtual instance is received in response thereto. Upon receiving the request, an entitlement of a user associated with the failed virtual instance will be tested. Specifically, a set of authentication calls and checks are deployed in accordance herewith to ensure the integrity of the requests, as well as the authorization of the requester for the resource use. Assuming testing is passed, a countdown timer associated with the failed virtual instance will be decreased. When the countdown timer reaches a predetermined threshold (e.g., zero), the new virtual instance will be started, the failed virtual instance will be terminated, and the countdown timer will be reset/restarted by instance. Integration with BSS further allows for monitoring and charging of service usage as well as opening the option for pay-as-you-go charges for the restart service itself.