ARX Data Anonymization Tool vs Greenmask

ARX Data Anonymization Tool is open-source software for anonymizing sensitive personal data and examining privacy risk and data utility. Its graphical application supports data import and cleansing, transformation-rule setup, tailoring anonymized datasets, and analysis of utility and residual re-identification risk. It imports CSV, MS Excel, and relational database data, including MS SQL, DB2, MySQL, and PostgreSQL. Listed methods include k-anonymity, l-diversity, t-closeness, δ-presence, risk-based criteria, and semantic models such as differential privacy. ARX exports displayed tables to CSV and compares input with transformed data using information-loss models, descriptive statistics, and application-specific analyses. Its risk analysis reports prosecutor, journalist, and marketer attack risks, estimates population uniqueness, and can identify HIPAA Safe Harbor identifiers and further quasi-identifiers. A Java library and API can add anonymization capabilities to Java programs. The tool is free under Apache License 2.0 and supports Linux, macOS, Windows, API, and self-hosted options. The maker provides PGP/GPG-signed checksums for download integrity and states the software is provided without warranties or conditions.

Greenmask is a free, open-source utility for logical database dumps, anonymization, synthetic data generation, and restoration. PostgreSQL 11 and later is fully supported; MySQL support is in progress and beta. Its subsetting tools can produce smaller development databases while preserving referential integrity, including cyclic and polymorphic references. Transformations may be deterministic or random, use dynamic parameters and conditional rules, and inherit across partitions and foreign-key references. Users can make custom transformers that exchange data through standard input and output, or use predefined templates. Greenmask supports local directories, S3-compatible storage, Azure Blob Storage, and SSH/SFTP. It uses PostgreSQL's pg_dump and pg_restore utilities for schema dumping and restoration, and its dumps can be restored with pg_restore. The utility is designed for CI/CD automation of anonymization and restoration. Its stateless, in-flight approach can anonymize data before it is persisted without requiring persistent storage of unmasked datasets in its normal workflow. Installation scripts support Linux and macOS; Windows is not supported yet.