Fast Data Masking™ for Oracle databases

The world's fastest data masking tool for Oracle!

Masking, obfuscating and de-identifying sensitive and personal data for use in non-production environments is quickly becoming a hot topic, as using "live" production data in development and testing is slowly becoming more and more unacceptable. Increased legislation, governing bodies like HIPAA, the PCI DSS and the GLBA, company fines, highly embarrassing data loss and data theft are forcing all companies to look at how they use production data for development and testing. In addition, the increased use of offshore, near shore and virtual images opens up all sorts of problems for securing your data from prying eyes.

Fast Data Masking™ for Oracle is an easy to use, no hassle solution which allows you to mask, reduce, compress and reorganize your test data in one operation. Using a powerful array of data masking functions, seed tables and high performance techniques you will quickly be able to create legally compliant databases for testing and development.

Grid-Tools data masking solution has an array of functionality to help you manage your large Oracle test databases.

Key features of the Grid-Tools data masking software

• Data masking - you can assign a rich variety of out‐of‐the‐box data masking to columns within a database and either extract clean data sets for external usage OR update the data in situ.
• Data compression - as the masked database is built the data can be compressed reducing your data footprint by up to 80%.
• Data reorganization - as the masked database is being built, reassign tables and indexes to new tablespaces allowing DBAs to balance their database disk access.
• Data reduction - a powerful database subsetting tool allows you to selectively extract sets of rows and their related tables. For example, only use a 5% sample of active customers and their associated tables.

Example run times

The example runtimes shown below demonstrate the difference in performance that can be achieved utilizing fairly modern hardware. This also demonstrates the inbuilt parallelism supported by Oracle and Unix. This explains the performance difference for the first test (Oracle) with the second, as scrambling for 180 tables were submitted in parallel sets of 5 (this can be adjusted by the user).

The bottom test (100 million rows scrambled in 58 minutes) is the likely scenario of running against SQL Server on a equivalent multi-processor machine running Windows (i.e. no operating system parallelism employed, but using multi-processor parallelism).

      

 

Back to the top