Veeningen, Meilof Geert

The invention relates to a cryptographic system (010) for performing a privacy-preserving computation on a dataset comprising multiple rows. Multiple cryptographic devices perform the computation as a cryptographic secure multi-party computation. A cryptographic device has a memory for storing the dataset, and a processor subsystem comprising multiple cores. The processor subsystem applies an operation of the cryptographic secure multi-party computation to the dataset by distributing the dataset across the multiple cores by assigning respective subsets of rows to the respective cores, wherein the distribution is determined from the number of rows; and applying the operation of the cryptographic secure multi-party computation to the dataset in parallel comprising, at a respective core, applying the operation to the subset of rows assigned to the respective core.

The invention relates to a cryptographic system (010) for performing a privacy-preserving computation on secret data. Multiple cryptographic devices perform the computation as a cryptographic secure multi-party computation. The privacy-preserving computation comprises performing an integer division of a numerator and a secret denominator. The secret denominator is normalized into a secret normalized denominator and a secret normalization factor. From the secret normalized denominator, a secret normalized denominator reciprocal is determined. A bias is added to the secret normalized denominator reciprocal to obtain a secret biased normalized denominator reciprocal. The secret division result is determined by performing a deterministic truncation based on the numerator, the secret biased normalized denominator reciprocal, and the secret normalization factor.