Secure Multiparty Computing via Restricted Operating Environment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud-based multiparty computing environments face challenges in securing data from unauthorized access by untrusted parties, with existing cryptographic solutions being slow and impractical.
Innovation Solution
A secure multiparty computing framework is implemented using a restricted operating environment with a guest agent, where a token representing user data is processed within a sandboxed Docker container, isolating the computation from external components and ensuring that only the processing result is returned to the user, preventing data leakage to untrusted parties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If cryptographic based multiparty computation solutions are used, then data security is improved, but computation speed deteriorates
Solution Approach 1:
The patent introduces a trusted execution environment (TEE) as an intermediary layer between the cryptographic protocol and the underlying hardware. The TEE provides a secure enclave that performs cryptographic operations in isolation, mediating between the need for security and the requirement for performance. This allows cryptographic computations to be executed within a protected boundary that guarantees security while leveraging hardware acceleration for speed.
Solution Approach 2:
The patent replaces software-based cryptographic implementations with hardware-accelerated cryptographic operations. By substituting mechanical/software processing with dedicated hardware modules (such as AES-NI instructions, SHA extensions, or specialized cryptographic accelerators), the system achieves both high security (through hardware-enforced protection) and high speed (through parallel processing and optimized circuits).
2Adaptability or versatility
If cloud based servers are used for multiparty computing, then accessibility and scalability are improved, but vulnerability to unauthorized access by untrusted parties worsens
Solution Approach 1:
The patent segments the cloud computing environment into isolated execution contexts, where each multiparty computation task runs in a separate, sandboxed container or virtual machine. This segmentation ensures that even if one execution context is compromised, untrusted parties cannot access data or processes in other contexts. The segmentation is enforced through hardware virtualization and memory protection mechanisms that are inherent to cloud infrastructure.
Solution Approach 2:
The patent creates an inert or protected execution environment using trusted execution environments (TEEs) that provide a secure boundary around sensitive computations. Within this inert environment, data and processes are protected from the surrounding untrusted cloud infrastructure. The TEE acts as a protective atmosphere that isolates the computation from potential attacks by untrusted parties while still allowing the computation to leverage cloud resources for accessibility and scalability.
Data Source
AI summary
According to one embodiment, in response to a request received at a host agent of a server from a user device of a user over a network to process user data, a data processing system launches a restricted operating environment within the server. The system transmits a token representing the request to a guest agent executed within the restricted operating environment, where an executor associated with the token is launched by the guest agent within the restricted operating environment, where the executor, when executed, is configured to process the user data to generate a processing result without accessing an external component external to the restricted operating environment. The system returns the processing result back to the user device.