Memory Buffer Modal Encryption with Adaptive IME Granularity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern computer systems face challenges in efficiently managing memory security and bandwidth due to limited metadata space for in-line memory encryption (IME) and error correction codes (ECC), particularly when using protocols like Compute Express Link (CXL), which can restrict the use of Chipkill technology and increase power consumption and latency.
Innovation Solution
A memory buffer device employing modal encryption and compression logic that adjusts encryption granularity based on data compression, allowing for a single MAC calculation across compressed data, reducing AES calculations and optimizing power usage by matching IME granularity to compression, and using side-band metadata for ECC check symbols.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If in-line memory encryption (IME) is implemented with traditional fixed granularity, then security is improved, but metadata space is exceeded and bandwidth is reduced
Solution Approach 1:
The patent implements dynamic encryption granularity that adapts to compression ratios. When compression is high, encryption operates at cache line granularity; when compression is low, it operates at larger block granularity. This dynamic adjustment optimizes the balance between security, metadata space usage, and bandwidth efficiency.
Solution Approach 2:
The system changes the encryption granularity parameter based on compression performance. By monitoring compression ratios and adjusting the encryption block size accordingly, the system can fit encryption metadata within available space while maintaining security requirements.
2Reliability
If Chipkill technology is used for error correction, then reliability is improved, but CXL protocol compatibility is reduced and device complexity increases
Solution Approach 1:
The patent implements a universal error correction approach using CRC codes that works across multiple protocols including CXL. This single error correction mechanism replaces protocol-specific implementations like Chipkill, providing broad compatibility while maintaining reliability.
3Reliability
If encryption granularity is reduced to cache line level, then security is improved, but AES calculation overhead increases and power consumption rises
Solution Approach 1:
The system dynamically adjusts encryption granularity based on workload characteristics and compression ratios. For compressed data with high security requirements, it uses cache line granularity; for other cases, it uses larger blocks, thereby reducing AES calculation overhead and power consumption when maximum security is not critical.
Solution Approach 2:
By changing the encryption block size parameter based on data characteristics, the system optimizes the trade-off between security strength and computational power consumption, avoiding unnecessary fine-grained encryption when it is not required.
4Reliability
If larger MACs are used for cryptographic integrity, then security is improved, but metadata space requirements increase
Solution Approach 1:
The patent implements dynamic MAC size selection based on available metadata space and security requirements. When metadata space is abundant, larger MACs provide stronger integrity verification; when space is constrained, smaller MACs are used, optimizing the balance between security and space utilization.
Data Source
AI summary
Technologies for modal encryption are described. One memory buffer device includes a compression block and an in-line memory encryption (IME) block. The compression block can output compressed data. The IME block can encrypt uncompressed data at a first granularity and encrypt the compressed data at a second granularity, wherein the second granularity is larger than the first granularity.