--- schema_version: "1" id: 0c5ac2f4f5b7d0556ea8d4c505c31325fa5b0cc8d0524a1f5d1d1a76bf587d71 title: Data Processing Engine collection: admin version: current locale: en-us portal_url: https://docs.salesforce.com/admin/current/en-us/release-notes/262-0-0/rn_data_processing_engine.md canonical_url: https://help.salesforce.com/s/articleView?id=release-notes.rn_data_processing_engine.htm&release=262.0.0&type=5 source_site: help.salesforce.com content_type: documentation content_hash: 688dcc475bf9eb3b1cbfaca2ede0e871e678531127ab463d6512cdb8c67dafca description: Transform context data with intricate product hierarchies and groupings at scale for business processes such as document generation. Debug DPE definitions directly in the builder for CRM Analytics runtime to pinpoint and resolve logic issues faster before writing back transformed data. Validate configurations during the design phase by using data preview to trace data flow and identify errors before saving Data 360 runtime definitions. products: - Cross Cloud release: "262" --- # Data Processing Engine Transform context data with intricate product hierarchies and groupings at scale for business processes such as document generation. Debug DPE definitions directly in the builder for CRM Analytics runtime to pinpoint and resolve logic issues faster before writing back transformed data. Validate configurations during the design phase by using data preview to trace data flow and identify errors before saving Data 360 runtime definitions. - **[Process Small Datasets Faster with On-Demand Processing](https://help.salesforce.com/s/articleView?id=release-notes.rn_dpe_on_demand_processing.htm&release=262.0.0&type=5)** Transform small to medium datasets and write back results in near real time without waiting for scheduled windows. Configure definitions to run in on-demand mode to perform ad hoc data validations and processing from Salesforce objects, context definitions, or CSV files. - **[Transform Context Data and Write Data to JSON with Data Processing Engine](https://help.salesforce.com/s/articleView?id=release-notes.rn_dpe_context_service_integration.htm&release=262.0.0&type=5)** Use Data Processing Engine (DPE) to read data from context definitions and write it to JSON to support high-volume data transformations for processes such as document generation. Use DPE in place of Apex plug-ins to scale transformations for complex product hierarchies and groupings without hitting governor limits. - **[Troubleshoot Data Processing Engine Definitions with Debug Mode](https://help.salesforce.com/s/articleView?id=release-notes.rn_dpe_debug_mode.htm&release=262.0.0&type=5)** Pinpoint and resolve issues in Data Processing Engine (DPE) definitions faster by debugging them directly in the DPE builder for CRM Analytics runtime. Instead of waiting for full writeback runs or creating temporary nodes, reduce troubleshooting time by inspecting filtered datasets and intermediate transformation stages before the final writeback. - **[Validate Logic and Debug Data 360 Runtime Definitions Faster with Data Preview](https://help.salesforce.com/s/articleView?id=release-notes.rn_dpe_data_preview.htm&release=262.0.0&type=5)** Validate definitions during the design phase by previewing node-level data transformations. Confirm that configurations meet business needs and identify logic errors early. Data preview is available for Data 360 and CRM Analytics definitions. Debug configurations in a controlled environment before saving or running definitions. - **[Aggregate Data Efficiently Across Related Records with Hierarchical Aggregation](https://help.salesforce.com/s/articleView?id=release-notes.rn_dpe_dmo_hierarchical_aggregation.htm&release=262.0.0&type=5)** Roll up values across multi-level parent-child hierarchies, such as from child accounts to a parent account, for definitions with CRM Analytics runtime. Aggregate data, such as total assets or revenue, from child records at any level of a hierarchy to the parent record. - **[Create Targeted Datasets with Single-Value and Multi-Value Lookup Joins](https://help.salesforce.com/s/articleView?id=release-notes.rn_dpe_debug_lookup_joins.htm&release=262.0.0&type=5)** Create a unique dataset by joining a primary and secondary data source with the Lookup join type. Preserve all records from the primary source and add matching data from the secondary source based on a common criteria. For example, match opportunity records (primary) with their corresponding account details (secondary) by using an account ID. If the secondary source has multiple matches, include data from only the first matching record or from all matching records in the resulting dataset for downstream nodes. Multi-value lookups aren't supported for Data 360 runtime definitions. - **[Select Writeback Operation Types for Data 360 Objects](https://help.salesforce.com/s/articleView?id=release-notes.rn_dpe_datacloud_writeback_operations.htm&release=262.0.0&type=5)** Choose the writeback operation type that fits data processing needs for data model objects (DMOs) and data lake objects (DLOs). Select Insert, Upsert, or Overwrite as the writeback operation type for the target in the Writeback Object node. These options cater to more business requirements, such as appending new engagement data or updating existing profile records without replacing the entire dataset. - **[Assign Data Model Object Categories for Data 360 Compatibility](https://help.salesforce.com/s/articleView?id=release-notes.rn_dpe_dmo_category_assignment.htm&release=262.0.0&type=5)** Categorize data model objects (DMOs) during writeback to make sure they work seamlessly with downstream Data 360 features. Data Processing Engine (DPE) previously assigned the Other category to all DMOs it created. Now, select the category that matches the object's purpose, such as Profile for customer details or Engagement for interactions. Accurate categorizing is essential for Data 360 functionalities, such as identity resolution and segmentation.