Project Antalya Roadmap 2026

[svb-alt]

Here's the current Project Antalya roadmap for 2026. This year the focus is on direct writes to Iceberg, improved usability of the swarms and Hybrid table engine, and interoperability with regard to cloud providers and catalog types. All features are open source--there are no hold-backs.

Please suggest additional features and ideas in the comments to this issue. We also welcome contributions.

Make Iceberg a first-class citizen in ClickHouse

• CREATE TABLE for Iceberg/DataLakeCatalog database engine using 'AS <other_table>' syntax. It should respect table structure, partitioning and ordering of MergeTree table if given
• DROP TABLE for Iceberg/DataLakeCatalog database engine
• TRUNCATE TABLE for Iceberg/DataLakeCatalog database engine
• Map specific catalog namespaces as Iceberg databases DataLakeCatalog namespace filter #1337
• ALTER TABLE EXPORT PARTITION ALL to Iceberg/DataLakeCatalog

Swarms:

• Fully supported joins
• Standalone swarms
• Swarm for writes
• Swarm for merges/optimize
• Observability

Tiered Storage:

• Write MergeTree parts to Parquet
• export part - skip or throw on pending mutations and patch parts #1294
• Accept table function as destination for part export #1320
• Allow merge tree materialized / alias columns to be exported through part export #1324
• Write MergeTree partitions to Parquet
• Export replicated merge tree partition to object storage #1097
• Yet another export replicated partition pr #1124
• Improvements to partition export #1177
• Write MergeTree parts/partitions directly to Iceberg
• Improve observability
• Add query id to system.part_log, system_exports and system.replicated_partition_exports #1330

Hybrid table engine

• Support for ALIAS columns Add support for ALIAS columns in segments for engine=Hybrid #1272
• ALTER TABLE for manual watermark management
• Automatic watermarks
• Direct exports to Iceberg
• TTL MOVE to Iceberg
• TTL DELETE for Iceberg data

Catalogs & Interoperability:

• ARN access to Glue #90825
• GCP support
• Azure support
• Interoperability with Spark
• Iceberg v3

Performance:

• Parquet RowGroup adaptive size
• Lazy metadata loading