What's new in Striim Cloud 5.4.0

Kafka Writer has been significantly enhanced with SmartMap capabilities, allowing each source table, object, or collection to be mapped to a single, dedicated Kafka topic on the target, simplifying downstream processing and analysis by preserving source-level separation of data changes.

The upgraded Kafka writer can write both snapshot data (for initial load) and ongoing change data—including CDC and other incremental changes—from a wide range of sources supported by Striim, including OLTP databases, data warehouses, message buses, files in multiple formats (Avro, CSV, JSON, Parquet, and XML), and SaaS applications.

Additional Kafka Writer enhancements include:

Sherlock and Sentinel, Striim’s AI agents for sensitive data governance, now detect sensitive data embedded in free-form text fields such as doctor’s notes or customer comments. Both agents can identify multiple types of sensitive data within a single text field, and Sentinel AI applies protection at the text-span level—safeguarding only the sensitive portions of text while leaving surrounding content unchanged. See Sherlock AI: sensitive data discovery in sources and related topics.

Striim now supports UPSERT operations for Database Writer when writing to MySQL, Oracle, PostgreSQL, and SQL Server, as well as for Spanner PG Dialect Writer and Spanner Writer (GoogleSQL dialect) when writing to Google Spanner. UPSERT operations update an existing target row or insert a new row if none exists, making them particularly useful for bulk data loads where some records may already be present, and for scenarios in which multiple processes write concurrently to the same table. See Upsert support in Database Writer, Upsert support in Spanner PG Dialect Writer, and Upsert support in Spanner Writer.

Oracle Reader and MSJet support TransactionCache Intelligence to monitor long-running and high-volume transactions in the transaction cache of source Oracle and SQL Server databases. When such conditions are detected, Striim can alert users to help proactively identify and address potential performance or data capture issues.

Striim Vault has been extended to support CyberArk, in addition to its existing support for Azure Key Vault, Google Secrets Manager, and HashiCorp Vault. You can securely store sensitive information—such as passwords, tokens, and keys—in your preferred enterprise credential store and access them seamlessly through Striim Vault. For details, see Using vaults.

The Flow Designer has been refreshed with a new icon library and a redesigned sidebar, improving visual clarity and navigation.

The wizard library now includes lite wizards, enabling faster and simpler application creation across most supported source and target combinations.

Wizard progress is now saved as a draft. If you exit a wizard before completion and later attempt to create an application with the same topology, Striim detects the unfinished draft and prompts you to resume where you left off.

Mutual TLS (mTLS) authentication is now supported for Kafka Writer and for accessing Confluent Schema Registry, enabling secure, certificate-based authentication.

Fabric Data Warehouse Writer now supports end-to-end private connectivity using Azure Private Link and Fabric workspace identity, enabling secure data writes to Microsoft Fabric Data Warehouse without exposing traffic to the public internet.

You can use the new HubSpot Writer to write to Hubspot.

This release includes multiple enhancements to MSJet that improve security, performance, and operational robustness when capturing data from SQL Server:

This release introduces several improvements to OJet that enhance security, flexibility, and enterprise compliance when capturing data from Oracle databases (see Oracle Database):

Oracle Reader includes the following performance improvements (see Oracle Database):

Salesforce Reader supports Salesforce Bulk Load API 2.0 to optimize large initial loads Configuring Bulk API for large data volumes).

Striim introduces enhanced security configuration for Kafka persistent streams and Schema Registry connections, with native support for Confluent Cloud.

Striim Cloud also supports external access to its internal Kafka cluster and Karapace Schema Registry. External systems can write data to Striim Cloud-hosted Kafka topics using mTLS authentication, enabling integration with third-party change data capture tools and other external data sources. See Accepting data from external sources using Kafka.

You can use Database Writer to write to Databricks Lakebase (compatible with PostgreSQL).

S3 Reader (see Amazon S3) and S3 Writer include minor functional changes resulting from Striim’s adoption of AWS SDK for Java version 2.31.

Striim Cloud now provides a REST API to programmatically upload .tql files and optionally execute them, enabling seamless integration with CI/CD workflows (see Using the Striim Application Management REST API).