Split Random Number Generator for CPU Core Entropy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing random number generators in computing systems often produce deterministic pseudo-random numbers, which can lead to security breaches due to low entropy, high correlations, and predictability, especially in critical applications like cryptography and encryption.
Innovation Solution
A split random number generator design where a master random number generator provides unique seed values to each CPU core, utilizing a cryptographic coprocessor with a noise source, entropy accumulator, and encryption block to generate high-quality, uniformly distributed random numbers, reducing predictability and improving entropy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a master random number generator provides unique seed values to each CPU core, then the quality of random number generation and entropy are improved, but the device complexity increases due to the need for a cryptographic coprocessor and seed distribution mechanism
Solution Approach 1:
The system segments the random number generation function by providing a single master RNG that generates unique seed values for each CPU core, rather than having each core with its own independent hardware RNG. This segmentation approach improves reliability through centralized entropy management while avoiding the complexity of duplicate hardware RNGs in every processor core.
Solution Approach 2:
The patent introduces a cryptographic coprocessor as an intermediary component that manages the seed value generation and distribution to CPU cores. This intermediary handles the complex cryptographic operations centrally, reducing the complexity burden on each individual core while maintaining high entropy and unpredictability in the random number generation.
2Productivity
If hardware random number generators are implemented in every processor core, then the productivity and performance of random number generation are improved, but the device complexity and area consumption increase
Solution Approach 1:
The master random number generator serves multiple CPU cores by providing unique seed values to each core, making a single hardware component serve multiple functions. This universal approach enables multiple cores to generate high-quality random numbers without requiring dedicated hardware RNGs in each core, thereby maintaining performance while reducing area consumption.
Solution Approach 2:
Instead of duplicating entire hardware random number generators in each processor core, the system creates virtual copies by distributing unique seed values from the master RNG to each core. Each core maintains its own random number generation capability through software implementation seeded with unique values, achieving functional duplication without physical duplication of the complex hardware RNG.
3Ease of operation
If software random number generators are used with dynamic seed values, then the predictability is reduced, but the reliability and entropy quality deteriorate compared to hardware RNGs
Solution Approach 1:
The system changes the parameter of seed value generation by using a master hardware RNG to generate cryptographically strong, unique seed values for each CPU core. This parameter change ensures that the entropy quality remains high while the software-based implementation provides flexibility and reduced predictability through dynamic seed updates and unique initialization for each core.
Data Source
AI summary
A computing system may implement a split random number generator that may use a random number generator to generate and store seed values in a memory for retrieval and use by one or more core processors to generate random numbers for secure processes within each core processor.