Assured Lazy Rollback for Multi-Tenant Identity Cloud Service
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current identity management systems in cloud environments face challenges in ensuring secure access and consistent security across diverse devices and user types, particularly in hybrid cloud deployments, where inconsistencies between on-premise and cloud environments can lead to security breaches and unauthorized access.
Innovation Solution
A multi-tenant cloud system with a microservices-based architecture provides secure identity and access management through a cloud-scale Identity and Access Management (IAM) platform, supporting secure access across various devices and user types by integrating with both on-premise and cloud applications, using standards-based services like OpenID Connect, OAuth2, and SCIM for unified identity management and secure token enforcement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the system continuously retries failed rollback operations, then the reliability of account creation is improved, but the resource consumption and system load increase significantly
Solution Approach 1:
The system implements periodic retry of failed rollback operations at predetermined time intervals (e.g., every 5 minutes) instead of continuous retry. The server schedules rollback retries to occur at specific periodic moments, reducing resource consumption while maintaining reliability through repeated attempts over time.
Solution Approach 2:
The system uses the client's existing status check requests to trigger rollback retries. When a client polls for account creation status, the server opportunistically initiates failed rollback operations at that moment, leveraging the client's necessary system interactions to perform additional maintenance tasks without requiring separate dedicated retry mechanisms.
2Stability of the object's composition
If the system implements comprehensive rollback mechanisms for failed account creation, then the data consistency is improved, but the system complexity increases
Solution Approach 1:
The system performs preliminary tracking of account creation operations and their associated resources. When a create request is received, the system pre-identifies what resources were allocated (user accounts, email routes, calendar accounts) and prepares rollback logic in advance, so that when failure occurs, the cleanup process is straightforward and systematic rather than requiring complex analysis of what needs to be cleaned up.
Solution Approach 2:
The system implements a feedback mechanism where the server monitors the status of asynchronous account creation operations and automatically triggers rollback operations when failures are detected. The server sends status check requests to clients and processes responses to determine whether rollback is needed, creating a closed-loop control system that maintains data consistency through automated feedback-driven correction.
3Speed
If the system uses asynchronous account creation operations, then the processing speed is improved, but the difficulty of detecting and measuring operation status increases
Solution Approach 1:
The system establishes a feedback loop where the server periodically sends status check requests to clients to query the state of asynchronous account creation operations. The client responds with current status information, allowing the server to detect completion or failure conditions and take appropriate actions (such as triggering rollbacks) based on this feedback information.
Solution Approach 2:
The system uses the client as an intermediary for status detection. Instead of the server directly and continuously monitoring the asynchronous operation state, the server leverages the client's status check requests as an intermediary mechanism to indirectly obtain operation status information, reducing the server's direct monitoring burden while maintaining visibility into operation outcomes.
Data Source
AI summary
Embodiments operate a multi-tenant cloud system for a plurality of user accounts. Embodiments receive a create request from a client system to provision a new account for a user. Embodiments determine that the create request has failed and initiates a rollback of the create request. Embodiments determine that the rollback has failed and initiates a new rollback in response to a status request from the client system.