What is Confluent?
Confluent is the company founded in 2014 by Jay Kreps, Neha Narkhede, and Jun Rao, who created Apache Kafka at LinkedIn. It sells two main products: Confluent Cloud, a fully managed Kafka service, and Confluent Platform, a self-managed distribution. Around Kafka it adds Schema Registry, managed connectors, stream governance, and managed Apache Flink for stream processing.
IBM announced an agreement to acquire Confluent in December 2025 and completed the acquisition on March 17, 2026, for an enterprise value of about $11 billion. Confluent continues to sell its products, now as part of IBM.
Tableflow: Kafka Topics as Lakehouse Tables
The lakehouse-relevant product is Tableflow, a Confluent Cloud feature that materializes Kafka topics as open tables in object storage without a separate pipeline. Support for Apache Iceberg became generally available in March 2025, and Delta Lake support followed in October 2025. Tableflow uses the topic’s schema from Schema Registry, writes Parquet data files, and can publish the tables to external catalogs, including the AWS Glue Data Catalog, Snowflake Open Catalog, and Databricks Unity Catalog, so engines such as Dremio, Trino, Spark, and Snowflake can query them.
Ways to Land Kafka Data in Iceberg
| Path | Who runs it | Strengths | Trade-offs |
|---|---|---|---|
| Confluent Tableflow | Confluent Cloud | No pipeline code; schema from Schema Registry; catalog sync | Tied to Confluent Cloud; usage-based pricing |
| Iceberg Kafka Connect sink | You, on any Kafka | Open source, part of the Apache Iceberg project; exactly-once commits | You operate Connect and tune commit intervals |
| Apache Flink job | You or a managed Flink service | Transform, join, and aggregate before writing | More engineering; commits happen on checkpoints |
| Redpanda Iceberg Topics | Redpanda | Built into the broker | Redpanda only |
All four produce the same thing at the end: Iceberg snapshots that any engine can read. The choice depends on whether you need transformation before landing, which Kafka distribution you run, and who you want operating the pipeline. Whichever path you choose, frequent small commits create small files, so plan for compaction.
To go deeper, read Kafka Connect to Iceberg: How the Commit Actually Works, The State of Streaming to Apache Iceberg in July 2026, and Kafka 4.0 Changes Streaming Platform Operations.
Frequently Asked Questions
Is Confluent the same as Apache Kafka?
No. Kafka is an Apache Software Foundation project anyone can run. Confluent is a company that sells managed and self-managed Kafka plus additional tools, some under its own licenses.
Does Tableflow replace ETL into the lakehouse?
For landing raw topics as tables, often yes. Cleaning, joining, and modeling still happen downstream, for example with dbt or SQL in the silver and gold layers.
What changed with the IBM acquisition?
Confluent is now an IBM company. IBM has announced integrations with products such as watsonx.data; check Confluent’s documentation for current product and pricing details.
Learn More
To learn more about the Data Lakehouse, read The Lakehouse Built for Everyone by Alex Merced. You can find this and other books by Alex Merced at books.alexmerced.com.