SaaS Identity Management Isolation via Hardened Containers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud-based SaaS solutions face challenges in securing customer data from service providers, as attackers can compromise service providers to access cloud service data, and internal employees can also pose a threat, necessitating secure authentication and data isolation.
Innovation Solution
Implementing a SaaS model for identity management that allows customers to maintain sovereign control over their identity and access management instances, with the ability to pull maintenance updates without exposing data to the service provider, using a multi-tenant identity cloud with dedicated trust zones and hardened Docker containers to isolate customer environments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a service provider is given access to customer data in the cloud, then service delivery and maintenance can be performed, but the service provider becomes an attack vector that can compromise customer data security
Solution Approach 1:
The system segments customer data by creating isolated tenant environments where each customer's data is stored in separate database schemas or tables. The service provider's access is segmented to only allow maintenance operations on infrastructure, not direct access to customer data. This segmentation enables service delivery while preventing data compromise.
Solution Approach 2:
An intermediary security layer is introduced between the service provider and customer data. This includes identity management services that act as mediators, controlling and auditing all access requests. The intermediary ensures service providers can perform maintenance through controlled interfaces without direct access to sensitive customer information.
2Ease of repair
If service providers can access customer environments for maintenance, then updates and patches can be applied, but internal employees or compromised providers can access and potentially steal sensitive information
Solution Approach 1:
The system implements preliminary anti-action by establishing security controls and authentication mechanisms before any maintenance access is granted. Identity management services pre-validate all maintenance requests, and security policies are configured in advance to prevent unauthorized access. This preliminary protection counters internal threats before they can materialize.
Solution Approach 2:
The architecture enables self-service maintenance capabilities where customers can apply their own security patches and updates through controlled interfaces. This reduces the need for service provider access to customer environments, thereby eliminating the internal threat vector while maintaining ease of repair through automated update mechanisms.
3Productivity
If a multi-tenant cloud environment is used to deliver SaaS services, then resource efficiency and scalability are improved, but isolation between customer data becomes more difficult to ensure
Solution Approach 1:
The system applies local quality by implementing tenant-specific security configurations and data isolation policies within the shared multi-tenant environment. Each tenant's data is marked with unique identifiers and access controls tailored to that specific customer's requirements. This allows resource efficiency through sharing while maintaining reliable data isolation through localized security measures.
Solution Approach 2:
The architecture introduces an additional dimensional layer of isolation through virtualization and containerization technologies. While physically sharing infrastructure resources for efficiency, the system creates logical separation in a higher dimension through isolated execution environments, network segmentation, and separate database schemas, thereby ensuring both productivity and reliability.
Data Source
AI summary
The disclosed technology teaches initializing an application instance using a SaaS model in a project implemented on a cloud-based computing service, including running a configuration engine that links a service provider for SaaS application to set configuration parameters for the project and initializing the project in which an application instance will be built, then removing the authorization of the configuration engine to access the project and removing access to set the parameters. The technology also includes running a SaaS application infrastructure builder autonomously, without the service provider having access to the builder, to build the instance, and then delivering the application as a SaaS service.