Virtual Machine Security via Intermediary Policy Enforcement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing distributed computer systems face challenges in securing inter-enterprise applications, as conventional firewall technologies are not well-suited for inter-enterprise environments, and process isolation methods like virtual machines are resource-intensive and not universally feasible.
Innovation Solution
Implementing a computer system that uses multiple virtual machines, each with its own access control mechanism, where message handling programs are executed in a separate virtual machine to perform security actions on intercepted messages, and local components of a distributed application are hosted in another virtual machine, allowing for secure message handling and resource isolation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If processes are isolated using virtual machines, then security is improved, but power consumption and resource usage increase
Solution Approach 1:
Multiple virtual machines are co-located on the same physical computer node, merging their resource usage. The message-handling VM and local-component VM share the same physical hardware resources (CPU, memory, storage), reducing overall power consumption compared to physically separate systems while maintaining virtual isolation for security.
Solution Approach 2:
A policy-enforcement point acts as an intermediary between the message-handling VM and local-component VM, enabling secure communication without requiring full process isolation. This intermediary enforces security policies that allow controlled information flow while maintaining virtual machine isolation.
2Productivity
If message handling programs are co-located with local components on the same computer, then resource utilization is improved, but security risks increase
Solution Approach 1:
The system segments message handling and local component execution into separate virtual machines. This segmentation maintains security boundaries while allowing both VMs to co-locate on the same physical computer, utilizing shared hardware resources efficiently.
Solution Approach 2:
The policy-enforcement point serves as an intermediary that enables secure interaction between the message-handling VM and local-component VM. It enforces security policies that control information flow between the segregated VMs, allowing resource sharing while maintaining security.
3Reliability
If conventional firewalls are used to secure enterprise networks, then security is improved, but they are not well-suited for inter-enterprise distributed applications
Solution Approach 1:
The policy-enforcement point acts as a distributed security intermediary that replaces conventional firewalls in inter-enterprise applications. It provides policy-based access control that is specifically designed for distributed application environments, enabling secure message passing between components across different enterprise boundaries.
Solution Approach 2:
The security system uses dynamic policy enforcement rather than static firewall rules. Security policies can be dynamically adjusted based on the specific distributed application context, message type, and security requirements, making the system adaptable to various inter-enterprise scenarios.
4Reliability
If processes are isolated on different computers, then security is improved, but power consumption increases
Solution Approach 1:
Multiple isolated processes run in separate virtual machines that are merged onto the same physical computer node. This consolidation maintains process isolation through virtualization while reducing power consumption by eliminating the need for multiple separate physical systems.
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
A distributed computer system is disclosed in which computers co-operate with one another by sending messages over a network such as the Internet in order to perform a distributed application. In order to improve the security of such system, each web service involved in the distributed application runs in a separate virtual machine. Furthermore, the virtual machines on a web server dedicated to respective web service instances utilise the same policy enforcement point - running in another virtual machine on the web-server - in order to handle messages for or from the web server. To increase security still further, each virtual machine provides virtual cryptoprocessor functionality which is used in the processing of messages sent in the performance of the distributed application.