How Secure is Your Data?
Before you answer, think of the different formats held by your enterprise. Then consider that 80%+ of data stored by enterprises is unstructured and unprotected.
What is XDP (Extended Data Privacy)
Why Choose XDP?
Develop confidently
Mask consistently
Improve efficiency
Mask what you have. Generate what you need.
Production data rarely covers every case you need to test or train on. XDP masks your real data so it's safe to use, and can also synthetically generate realistic data to complement it: edge cases, new products, rare scenarios, or volumes your production data doesn't have. It works across structured, unstructured and file-based data, so documents and files get the same treatment as database records.
Fill coverage gaps
Create the scenarios and edge cases your production data doesn't include.
Scale up
Generate the volume you need for performance testing and AI training.
Your choice of method
Use AI-based generation (GAN) for statistically realistic data, or rules-based generation for fast, predictable values.
Zero exposure
Synthetic data contains no real customer information, so it's safe to share and use anywhere.
Dynamic masking: the right view for every user
Dynamic Data Masking is part of XDP, not a separate product. Instead of creating masked copies, XDP can mask sensitive data on the fly as it is requested, showing each user only what their permissions allow while the original data stays unchanged.
Masked on the fly
Sensitive values are masked at the moment they are viewed, based on each user's permissions, so people only see what they are authorized to see.
No permanent changes
The original stays intact and keeps its look and feel. Store one file instead of several versions with different levels of redaction, and save storage along the way.
Rules you control
Use pre-defined or custom rules to set what each user or role can see, across data on-premise or in the cloud.
Works with the data you already have
XDP connects to relational and NoSQL databases, files and documents, streaming platforms and SaaS applications, with the same masking rules applied consistently across all of them.
Supported data sources
Relational
- DB2 (LUW and z/OS)
- Oracle
- SQL Server
- PostgreSQL
- Sybase
- Snowflake
- BLOBs, CLOBs, XML, JSON and more
NoSQL
- MongoDB
- Cassandra
- AWS DynamoDB
- Cosmos DB
- HBase
- and more
Semi-structured and unstructured
- Flat files: XML, JSON, CSV, system files, Jira, log files and more
- Microsoft Office documents
- Images
- Parquet
- EDI and HL7
Supported platforms
- Linux: Ubuntu, RHEL, Debian, Fedora and more
- Windows 10, 11 and Server 2016+
- z/OS
- IBM i (AS/400) 6.1+
Integrations
- Databricks: Delta Lake, Unity Catalog, Spark
- SaaS via REST APIs: Salesforce, SAP, Workday, Epic
- Streaming: Kafka, Kinesis, Pub/Sub
- Legacy adapters: DB2 z/OS, VSAM, QSAM, iSeries
- Full API access for IBM Optim Data Privacy, Watson Discovery and in-house tools
Architecture and governance
- Web-based, fully API-callable for CI/CD and ETL
- Containerized or container-less (ECS), Kubernetes; no database install
- LDAP/AD and multi-factor authentication
- Conditional (if-then-else) masking rules
- Lineage-aware before/after reporting
- Optional AI-based PII/PHI detection; synthetic data generation with your choice of AI-based (GAN) or rules-based methods
Solutions - FAQ’s
What is XDP (Extended Data Privacy)?
How does XDP differ from traditional data-masking tools?
Which Privacy Regulations do XDP address?
Is XDP AI powered?
What platforms can XDP be run on?
FAQ’s
What is the difference between structured, unstructured, and semi-structured data?
Structured data is typically identified as coming from a relational database, where the data is arranged in specific rows and columns inside a table.
Semi-structured data will usually be in a loose arrangement. The data might be arranged in rows, but could have different column lengths. It might contain tags or elements to identify the fields.
Unstructured data is just that — data with no defined grouping. This is sometimes referred to as ‘free-form text’. Common file types include DOCX, JPG, and PDF.
Why should I protect my data?
In general, there are two main reasons to mask and protect your enterprise data:
1. Regulations and compliance: Heavy fines and penalties may be applied if you fail an audit or don’t comply with a regulation. A security breach or incident does not need to occur if it is discovered that you are out of compliance. Apart from financial costs, brand and company reputation can also be negatively affected. Laws, rules, and regulations are only going to tighten, so protecting your data will not only help to prevent data loss but potentially mitigate non-compliance to any applicable regulatory requirements.
2. Data breach or incident: A data breach could occur when someone either deliberately or inadvertently steals or releases data outside an organization. An example is when a contractor has their laptop stolen, which could contain sensitive production data loaded into a spreadsheet for analysis. This information can have immense value and is often sold for value or shared to help negatively affect an organization's reputation.
What data sources does XDP support?
Although XDP’s primary focus is to mask sensitive data contained within unstructured and semi-structured files and repositories — it can also mask data in structured databases as well.
Structured: DB2, Oracle, and SQL Server fields can be masked. These can be BLOB, CLOB, XML, or free form text fields. Other databases that leverage JDBC/ODBC connectivity can also be masked. And for databases not directly supported, XDP can even leverage the database systems’ unload/export utilities.
Semi-structured and Unstructured: XDP works with a multitude of types of files. Any text-based files, such as JSON, XML, and CSV, can be masked. It can also mask Microsoft Office documents, PDF, and image files (e.g. JPG).
What platforms does XDP support?
XDP supports Windows and Linux operating systems. Versions 2.8.0.3 and earlier are native to Windows as they leverage ‘Dot Net’ technology. Newer versions support Linux and Mainframe USS systems.
How does XDP scale?
XDP can easily scale up and out. By adding more memory and/or CPU power, the overall performance can be enhanced. Another solution is to simply add another server device to provide better scalability.
Can XDP do consistent masking?
Yes. Depending upon the masking rule applied, XDP can replace like-for-like values. For example, it can replace a valid VISA card number with another card number, and a valid U.S. address with another valid U.S. address. Applying the same masking rule will ensure the new masked value is consistent regardless of where it resides.
How long does it take to implement? Where do we start?
XDP takes minutes to install. Depending on the requirements, implementation takes between a few days to several weeks. Each project is delivered in phases which typically include: discovery and analysis, connectivity setup, building masking rules, and final execution.
How does XDP know what to mask?
XDP is a rules-driven application. Rules are created to identify where the data is and what the new value would be. XDP leverages advanced regular expression pattern matching alongside customized routines to discover and mask the data.
How does XDP discover data?
By implementing pre-defined rules, XDP searches for patterns which match those defined by the rules. Once the pattern is identified, the applicable data masking rule will be executed. XDP can also remediate sensitive data discovered by third-party data discovery tools.
Can XDP access cloud storage?
Yes. XDP has been tested and certified for numerous cloud storage solutions including: S3 Cloud storage, Google Drive, Microsoft One Drive and Box.
Can XDP integrate with my existing structured data masking process?
If you’re already using products like IBM InfoSphere Optim Data Privacy, XDP can perform the same data masking routines. This helps to ensure a consistent data masking value for all types of data.
What’s the difference between redaction, masking, and encryption?
Redaction is simply the replacement of the data pattern found with X’s or the word ‘REDACTED’. Different redaction values can also be used.
Masking is considered an intelligent replacement of data. Using this type of process replaces sensitive data with realistic looking alternatives, which satisfy the regulatory requirements.
Encryption is a process where specialized algorithms are executed on data, transforming it to a series of numbers. While this is very secure, it sometimes renders the data unusable in the original application.
What is obfuscation, anonymization, de-identification?
These three terms are often used interchangeably. They essentially mean the process of altering a source data element with another value in order to render the data non-sensitive.
Can XDP do dynamic masking?
Yes. Dynamic Data Masking is built into XDP. Sensitive data is masked on the fly based on each user's permissions, without permanently changing the original.
Interested in XDP (Extended Data Privacy)?
