Last reviewed: October 2026. Broadcom statements checked against Broadcom TechDocs for Test Data Manager 5.0 and Broadcom's May 2026 solution brief.
The short answer. Broadcom Test Data Manager (Broadcom TDM) and K2view both mask, subset and generate test data. They differ in what a unit of test data is. Broadcom TDM works from a data model of tables and relationships. K2view delivers each business entity, such as a customer or an account, complete across every system it lives in. The difference shows up when one test spans several systems. Broadcom TDM keeps referential integrity across applications, but it cannot automate that across them. K2view masks and provisions the whole entity in one automated run, from the mainframe (through its partnership with Rocket Software) to the cloud and SaaS.
Naming: Broadcom TDM was formerly CA Test Data Manager, and its product line traces back to Grid-Tools. You will also see its modules named: the TDM Portal, Fast Data Masker and Datamaker.
If you run Broadcom TDM 4.x: Broadcom deprecated vTDM from release 4.11 and replaced it with DAVE in 5.0. Broadcom has ended support for older releases about a year after announcing it. Check your version's end-of-service date before you renew.
Whichever tool you pick, three things decide whether test data works. Does the same customer still match across every system after masking? Can a tester get exactly the data a test needs without waiting for a specialist? Can a pipeline request that data without a ticket?
The two platforms start from different units of test data. Broadcom TDM starts from a data model of tables and relationships, organized into projects and environments. K2view starts from the business entity. The first three rows are where that difference matters most.
| Broadcom TDM 5.0 | K2view | |
|---|---|---|
| Unit of test data | A data model of tables and relationships, in projects and environments | A business entity (customer, order, account) across all its systems |
| Consistency across systems | Keeps referential integrity across applications. Cannot automate it across them | Built in: each entity is assembled and masked as one unit across every connected system, with no table stitching |
| Deployment | On-premises only | SaaS deployment option |
| Product scope | Several modules: TDM Portal, Fast Data Masker, Datamaker, and optional DAVE | One integrated product from sensitive data discovery through masking, subsetting, synthetic data, reservation, versioning and CI/CD APIs |
| Data sources | Many relational databases certified. Snowflake, BigQuery, MongoDB and Cassandra listed as Supported. No SaaS application connectors | Structured, semi-structured and unstructured sources through one entity model: relational, NoSQL, APIs, files, mainframe, cloud and SaaS, including Salesforce, Workday and Snowflake |
| Masking | In place on a staged copy (Fast Data Masker) or in flight. Portal, Fast Data Masker and Datamaker each have their own source list. Scales on Kubernetes | Static, dynamic and in-flight, entity-aware. The same value masks the same way across systems, environments and runs |
| Core technique | Subsetting, masking and rule-based generation from the data model. Optional virtual clones through DAVE | Entity-based subsetting from per-entity Micro-Databases. Select entities by business attributes |
| Synthetic data | Rule-based generation for tables, files, images and NoSQL, with Data Assistant templates | Rule-based, AI-trained (GenAI), entity cloning, and masked production data in one workflow. Entity-aware, cross-system, with statistical validation reports |
| AI and test-case-driven provisioning | Optional AI Assistant from 5.0.1000.0: an in-portal chatbot and an MCP server | From test case to test data: the Test Case Data Agent in Agentic TDM plans and runs provisioning from test cases |
| Database virtualization | DAVE, new in 5.0: Oracle, MongoDB, Db2 (including z/OS), PostgreSQL, SQL Server and SAP HANA. Replaces vTDM, deprecated in 4.11 | Delivers entity subsets instead of virtual database copies |
| Mainframe | Native masking engines on z/OS for Db2, IMS and VSAM | With Rocket Software: real-time access to mainframe data, masked, subset and provisioned in K2view, with integrity kept from mainframe to cloud |
| Tester controls | Find and reserve, test data warehouse, self-service catalog and flows | Self-service portal for dev and QA. Versioning, entity reservation, rollback and aging. Same dataset to any environment in minutes |
| Automation | REST APIs, Javelin workflows, MCP server | REST APIs, scheduler, Agentic TDM |
Broadcom column: Broadcom TechDocs for Test Data Manager 5.0, Broadcom's Test Data Manager solution brief (6 May 2026), the Broadcom Software Academy blog (28 July 2025), and the Broadcom community site. Broadcom calls a data source Certified when it has been fully tested, and Supported when it is expected to work but has not been through a full testing cycle.
Broadcom is actively developing TDM. Release 5.0 added database virtualization (DAVE), which now covers Oracle, MongoDB, Db2 including z/OS, PostgreSQL, SQL Server and SAP HANA, Data Assistant templates, masking for CSV, JSON and Parquet files, a self-service catalog, and, from 5.0.1000.0, an optional AI Assistant. Everything is still built from projects, data models and tables. DAVE replaces vTDM, which Broadcom deprecated from release 4.11. The model behind Data Assistant Plus is heuristic-based.
Reviewers praise Broadcom TDM's masking, synthetic data and test data reservation. The criticisms that recur are about effort and fit. Examples from Gartner Peer Insights, with dates:
Most of these reviews were written before release 5.0. The proof-of-concept plan below turns each one into a check you can run on any vendor, K2view included.
Results from K2view case studies (2023 for the bank and insurer).
A real customer does not live in one database. The profile sits in CRM, invoices in billing, history on a mainframe. K2view stores each customer as one entity in its own Micro-Database, with data from every connected system, structured, semi-structured or unstructured, and replaces shared keys consistently across systems so referential integrity holds after masking. Complex joins and table stitching disappear from test data workflows.
K2view supports Snowflake, Workday, mainframe, Oracle, Salesforce, MongoDB and SQL Server data masking, as well as SAP test data management. Broadcom TDM has no SaaS application connectors. K2view was named a Visionary in the 2025 Gartner Magic Quadrant for Data Integration Tools.
A reviewer on Gartner Peer Insights describes using this approach to supply test data with referential integrity across a large number of applications, spanning mainframe, relational and NoSQL sources.
K2view is a single integrated product: sensitive data discovery, masking, subsetting, synthetic data generation, reservation, versioning and CI/CD APIs, with less scripting and fewer handoffs. Masking is entity-aware and available as static, dynamic and in-flight. It is deterministic: the same source value produces the same masked value across systems, environments and runs, and policies are defined once in a central catalog. Broadcom TDM spreads these jobs across the TDM Portal, Fast Data Masker and Datamaker, and each has its own supported-source list. Before release 4.11.3000.0, for example, flat-file masking had to be configured in Fast Data Masker, not the Portal.
As AI speeds up test creation, the data request becomes the slowest step. K2view Agentic TDM takes over that step. Its Test Case Data Agent reads test cases from your test management tool, works out what data each one needs, builds a provisioning plan, and runs the K2view tasks to deliver it. For each test case it can blend masked production data, cloned data, rules-based synthetic data and AI-based synthetic data. Masking and access rules apply automatically. Broadcom's AI Assistant lets users operate TDM through chat and MCP clients. K2view Agentic TDM starts from the test case itself.
Most tests need a few hundred well-chosen customers, not a copy of production. Testers use a self-service portal to select entities by business attributes, such as customers on a given plan with an overdue balance, and provision only those, to any environment, in minutes. They can save versions of a dataset and reload them, reserve entities so no one overwrites them mid-test, and roll back after a destructive run. Testers unblock themselves, which means fewer tickets and faster cycles.
No single method fits every scenario. K2view supports several inside one TDM workflow:
Generation is entity-aware and works across systems. Synthetic sets are validated statistically, with reports for accuracy and compliance, and datasets can be aged. Teams can mix methods in the same environment for edge-case, negative, performance and regression testing, and all of them follow the same entity model and masking rules.
Every K2view TDM task can run from the self-service portal, on a schedule, or through a REST API. A pipeline can start a task with its own parameters, track the execution, and extract or load a specific data version. Permissions are enforced per environment, so automation never bypasses access rules.
The bottom line. Broadcom TDM can manage test data. K2view removes test data as a bottleneck. If your roadmap includes CI/CD, cloud platforms or multi-system applications, an entity-based platform is the faster path. Calculate what that is worth with the K2view TDM ROI calculator.
These questions come from what buyers ask when a Broadcom agreement comes up. Put them to any vendor, including us.
Experience K2view Test Data Management first-hand in our interactive product tour, or request a demo.
Broadcom Test Data Manager builds test data from a data model of tables and relationships, with subsetting, masking, rule-based generation and, from release 5.0, database virtualization. K2view Test Data Management delivers business entities, such as a customer or an order, with their data from every system. Broadcom TDM suits mainframe and database-centric estates. K2view suits tests that span several systems.
No. Broadcom TDM keeps referential integrity across applications, but it cannot automate that across them. Ask any vendor to show cross-application masking in one automated run, with the same customer masked to the same values in every system.
On Gartner Peer Insights, reviewers praise Broadcom TDM's masking, synthetic data and test data reservation. Recurring criticisms are the learning curve, setup effort, integration with other tools and cost. Most reviews date from 2023 or earlier, before release 5.0, and each is an individual opinion.
Rule-based generation, AI-based generation from a model trained on your data, entity cloning, and masked production data. Generation is entity-aware and works across systems, with statistical validation reports. Teams choose a method per test, within one workflow.
Yes. Broadcom TDM is the product formerly sold as CA Test Data Manager, and Broadcom's documentation still uses the CA name. Its product line traces back to Grid-Tools. Modules you may see named include the TDM Portal, Fast Data Masker and Datamaker.
No. Broadcom TDM is on-premises only. K2view Test Data Management offers a SaaS deployment option.
Broadcom deprecated Virtual Test Data Management (vTDM) from release 4.11 and added the Database Virtualization Engine (DAVE) in release 5.0. DAVE supports Oracle, MongoDB, Db2 (including z/OS), PostgreSQL, SQL Server and SAP HANA.
Yes, as optional features. From release 5.0.1000.0, Broadcom TDM includes an AI Assistant: an in-portal chatbot and an MCP server that tools such as Cursor, Claude Code and ChatGPT can call. Its Data Assistant Plus model is heuristic-based.
Yes. Broadcom TDM supports Db2, IMS and VSAM on z/OS with native masking engines. K2view supports the mainframe through its partnership with Rocket Software: Rocket gives real-time access to mainframe data, and K2view masks, subsets and provisions it, keeping integrity from mainframe to cloud.
Broadcom TDM certifies many relational databases, including SQL Server 2012 to 2016 and Oracle 11g and 12c. It lists newer versions, Snowflake, BigQuery, MongoDB and Cassandra as Supported, meaning expected to work but not fully tested. It has no SaaS application connectors.
For some needs, yes. Teams look at K2view when tests span many systems, including SaaS and cloud sources, when they want one masking approach, or when they want test data provisioned from test cases. Teams with mostly mainframe workloads may keep Broadcom TDM. See also Informatica TDM vs K2view and IBM Optim vs K2view.
Enterprises including AT&T use K2view Test Data Management. K2view case studies describe a top-10 US bank that cut the staff provisioning test data from 50 to 10 FTEs, and a Fortune 500 insurer that cut test data provisioning cost by 70%, excluding hardware. Both replaced in-house or legacy tools.
Ask which version you run and when Broadcom ends support for it, what TDM costs inside your agreement, which masking path covers each source, whether you use vTDM and what the DAVE path is, and how referential integrity holds across systems after masking. Broadcom ended service for release 4.8 on 31 March 2023.
Each entity's data from all systems is stored together and masked as one unit. Shared identifiers are replaced consistently across systems, so joins keep working after masking.
It is test data provisioning driven by an AI agent, from test case to test data. In K2view Agentic TDM, the Test Case Data Agent reads test cases, plans the data each one needs, and runs the provisioning tasks under your existing masking and access rules.
Related: Delphix vs K2view, IBM Optim vs K2view, DATPROF vs K2view, Informatica TDM vs K2view, and K2view Test Data Management.