Batched Format-Preserving Encryption Parallel Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Format-preserving encryption (FPE) techniques using block-oriented cryptographic primitives like AES are slow due to forced serialization, limiting optimization potential and performance in computing environments.
Innovation Solution
Batching FPE operations by performing multiple encryption calls in parallel on independent data fields, leveraging the same key and tweaks, to accelerate encryption processing without changing the underlying implementation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If format-preserving encryption is implemented using block-oriented cryptographic primitives like AES, then encryption security is maintained, but processing speed deteriorates due to forced serialization
Solution Approach 1:
The patent segments the encryption processing by identifying independent data fields within a batch that can be encrypted separately. Each field is processed as an independent unit through the FPE function, allowing parallel execution of encryption operations across multiple fields while maintaining security requirements. This segmentation enables the system to bypass the serialization bottleneck by processing multiple segments simultaneously.
Solution Approach 2:
The patent merges multiple independent encryption operations into a single batch processing call. By combining several independent data fields into one batch input for the FPE function, the system achieves economies of scale and reduces overhead. The merging of operations allows the cryptographic primitive to process multiple items in a unified operation that can be executed in parallel, improving overall throughput while maintaining the security properties of individual encryptions.
2Reliability
If multiple encryption calls are performed sequentially on data fields, then encryption accuracy is ensured, but productivity deteriorates
Solution Approach 1:
The system segments the batch of data fields into independent encryption units that can be processed in parallel. Each field within the batch is treated as a separate encryption task that maintains accuracy requirements while being executed simultaneously with other fields. This segmentation transforms the sequential processing model into a parallel one, dramatically improving productivity without compromising encryption accuracy.
Solution Approach 2:
The patent implements a universal batch processing interface that handles multiple independent encryption operations through a single function call. This multi-functional approach allows the same encryption logic to be applied uniformly across all fields in the batch, ensuring consistency and accuracy while achieving high throughput through parallel execution. The universal interface abstracts away the complexity of managing multiple individual encryption calls.
3Reliability
If format-preserving encryption uses tight loops of iterative chained calls, then encryption security is strengthened, but device complexity increases and optimization potential is limited
Solution Approach 1:
The patent merges multiple iterative chained encryption calls into a single batch processing operation. By combining what would traditionally require separate sequential calls into one unified batch operation, the system reduces the complexity of managing multiple function invocations, error handling, and state management. The merging preserves the security-strengthening iterative chains within each field while eliminating the overhead of repeated call setup and teardown.
Solution Approach 2:
The patent transitions from a single-dimensional sequential processing model to a multi-dimensional parallel processing model. Instead of processing fields one after another in a single thread, the system orchestrates multiple independent encryption operations across parallel execution threads. This dimensional change from sequential to parallel architecture reduces processing complexity by distributing the computational burden and eliminating serialization overhead, while maintaining the security properties of the iterative chained calls within each parallel stream.
Data Source
AI summary
Batched execution of encryption operations is performed. A batched set of data for which format-preserving encryption is to be performed is obtained. The batched set of data includes a plurality of fields of data, which are independent of one another. Multiple rounds of format-preserving encryption are performed on the plurality of fields of data to provide an output of format-preserved encrypted data. A round of format-preserving encryption includes calling an encryption function to perform one or more encryption operations on the plurality of fields of data in parallel.


