Multi-Tenant Workload Integrity via LDWINT Binary Loading
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory encryption techniques in multi-tenant computing environments provide insufficient defense against active attacks where an attacker with physical access can modify code/data in memory, leading to potential secrets leakage, and current integrity protection solutions introduce significant overheads in cloud-based implementations.
Innovation Solution
Implement a Load with Integrity (LDWINT) instruction and a Multi-Key Total Memory Encryption (MKTME) engine mode to load binaries with integrity, using an integrity-only mode that generates cryptographic responses without encryption, ensuring the binary is loaded correctly and securely, while utilizing IntelĀ® Virtualization Technology and Trusted Execution Technology to manage guest virtual machines and protect consumer data from host VMM access.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional memory integrity mechanisms with cryptographic MAC are implemented, then defense against active attacks is improved, but system overhead increases
Solution Approach 1:
The patent segments the integrity protection mechanism by separating MAC generation from traditional memory access paths. The MAC is generated only during binary loading operations via the LDWINT instruction, rather than continuously during all memory operations. This segmentation reduces overhead by limiting integrity computation to critical load operations only.
Solution Approach 2:
The patent applies preliminary action by generating the cryptographic MAC during the binary loading phase before execution begins. The integrity verification is performed upfront during load operations, so that once loaded, the binary can execute without continuous integrity checking overhead. This preliminary verification approach eliminates runtime overhead while maintaining security.
2Reliability
If memory encryption is implemented to protect confidentiality, then defense against passive attacks is improved, but no defense exists against active modification attacks
Solution Approach 1:
The patent merges memory encryption with integrity protection by combining MKTME (Multi-Key Total Memory Encryption) with cryptographic MAC generation in a unified memory management approach. The encryption provides confidentiality while the integrated MAC mechanism provides integrity verification, creating a combined security solution that defends against both passive eavesdropping and active modification attacks simultaneously.
Solution Approach 2:
The patent uses composite security mechanisms by layering cryptographic MAC verification on top of existing memory encryption (MKTME). This creates a composite protection scheme where encryption handles confidentiality and MAC handling handles integrity, similar to how composite materials combine different properties to achieve multiple functional requirements.
3Reliability
If current integrity protection solutions are implemented in cloud environments, then security is improved, but implementation difficulty increases significantly
Solution Approach 1:
The patent applies self-service by enabling the cloud customer to independently verify the integrity of their binary using the cryptographic response and shared secret key. The customer's software can autonomously compute the expected MAC and compare it against the cryptographic response provided by the cloud provider, eliminating the need for complex trusted execution environments or hardware-based verification mechanisms on the provider side.
Solution Approach 2:
The cryptographic response serves as an intermediary that bridges the trust gap between cloud customer and provider. Instead of requiring complex mutual authentication protocols or trusted hardware, the system uses a simplified intermediary mechanism: the cloud provider computes and returns a cryptographic response (MAC) that the customer can independently verify using their own key, establishing trust through cryptographic proof rather than complex protocol negotiation.
Data Source
AI summary
In one embodiment, a multi-tenant computing system includes a processor including a plurality of cores on which agents of tenants of the multi-tenant computing system are to execute, a configuration storage, and a memory execution circuit. The configuration storage includes a first configuration register to store configuration information associated with the memory execution circuit. The first configuration register is to store a mode identifier to identify a mode of operation of the memory execution circuit. The memory execution circuit, in a first mode of operation, is to receive encrypted data of a first tenant, the encrypted data encrypted by the first tenant, generate an integrity value for the encrypted data, and send the encrypted data and the integrity value to a memory, the integrity value not visible to the software of the multi-tenant computing system. Other embodiments are described and claimed.


