diff --git a/Antalya.md b/Antalya.md index be8b15a41beef..b05e99e625889 100644 --- a/Antalya.md +++ b/Antalya.md @@ -1,111 +1,137 @@ | Feature | Altinity PR | First Altinity Release | upstream PR | First upstream Release | |----------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------:|:----------------------:| :---: | :---: | |:arrow_forward: SECURITY| | || +|Map OAuth groups to settings profiles with `profiles_mapping`|https://github.com/Altinity/ClickHouse/pull/2664||| +|Forward authentication tokens to data lake catalogs|https://github.com/Altinity/ClickHouse/pull/2329|26.6.8|| +|Google identity provider for token-based authentication|https://github.com/Altinity/ClickHouse/pull/2499|26.6.8|| +|External JWT acquisition with `clickhouse-client --jwt-command`|https://github.com/Altinity/ClickHouse/pull/1809|26.3.10|| +|Microsoft Entra ID authentication through OIDC and local JWKS validation|https://github.com/Altinity/ClickHouse/pull/1784|26.3.10|| |Add OAuth2 login flow to `clickhouse-client`|https://github.com/Altinity/ClickHouse/pull/1606|26.1.11|| |Add support for EC JWT validation|https://github.com/Altinity/ClickHouse/pull/1596|26.1.6|| -|Add setting `allow_local_data_lakes`, turned off by default|https://github.com/Altinity/ClickHouse/pull/1531|25.8.16|| -| Token-based authentication and authorization |https://github.com/Altinity/ClickHouse/pull/1078| 25.8.12 || +| Token-based authentication and authorization |https://github.com/Altinity/ClickHouse/pull/1078| 25.8.12 |https://github.com/ClickHouse/ClickHouse/pull/89978, https://github.com/ClickHouse/ClickHouse/pull/89995| | --- | |:arrow_forward: PERFORMANCE| | || -|Parallelize object storage output|https://github.com/Altinity/ClickHouse/pull/1580|26.1.6|https://github.com/ClickHouse/ClickHouse/pull/99548|26.3.1| -|Introduce async prefetch and staleness for Iceberg metadata|https://github.com/Altinity/ClickHouse/pull/1575|26.1.6|https://github.com/ClickHouse/ClickHouse/pull/96191|26.3.1| -|Parquet metadata cache v2|https://github.com/Altinity/ClickHouse/pull/1574|26.1.6|https://github.com/ClickHouse/ClickHouse/pull/98140|26.3.1| +|Reuse AES-GCM cipher contexts across rows; up to 27x faster multithreaded encryption and decryption|https://github.com/Altinity/ClickHouse/pull/2898, https://github.com/Altinity/ClickHouse/pull/2900||https://github.com/ClickHouse/ClickHouse/pull/131001| +|Reuse Puffin object metadata across Iceberg deletion-vector reads to reduce HEAD requests|https://github.com/Altinity/ClickHouse/pull/2419|26.6.8|https://github.com/ClickHouse/ClickHouse/pull/121786| +|Parquet v3 read-ahead and decoding with separate memory and thread budgets|https://github.com/Altinity/ClickHouse/pull/2235|26.6.2|| +|Reduce partition-export lock contention with multiversion local manifests|https://github.com/Altinity/ClickHouse/pull/1912|26.3.13|| +|Avoid holding export-manifest locks during slow commits|https://github.com/Altinity/ClickHouse/pull/1813|26.3.10|| +|Skip remote `DESCRIBE TABLE` when the data lake schema is already known|https://github.com/Altinity/ClickHouse/pull/1845|26.3.10|| +|Reduce lock contention in the Iceberg cluster iterator|https://github.com/Altinity/ClickHouse/pull/1436, https://github.com/Altinity/ClickHouse/pull/1497|25.8.16|| +|Read optimization using Iceberg metadata|https://github.com/Altinity/ClickHouse/pull/1019|25.8.9|https://github.com/ClickHouse/ClickHouse/pull/90001| +| Iceberg table pruning in cluster requests |https://github.com/Altinity/ClickHouse/pull/770 |25.2.2 || | ListObjectsV2 cache |https://github.com/Altinity/ClickHouse/pull/743 | 25.2.2 || -| Lazy load metadata for metadata for DataLake |https://github.com/Altinity/ClickHouse/pull/742 | 25.2.2 || +| Lazy loading of data lake metadata |https://github.com/Altinity/ClickHouse/pull/742 | 25.2.2 || | Parquet file metadata caching: clear cache |https://github.com/Altinity/ClickHouse/pull/713 | 25.2.2 || | Parquet file metadata caching |https://github.com/Altinity/ClickHouse/pull/586 | 24.12.2 || -| Parquet file metadata caching: use cache for parquetmetadata format |https://github.com/Altinity/ClickHouse/pull/636 | 24.12.2 || -| Parquet file metadata caching: turn cache on by default |https://github.com/Altinity/ClickHouse/pull/669, https://github.com/Altinity/ClickHouse/pull/674| 24.12.2 || +| Use the Parquet metadata cache for the ParquetMetadata format |https://github.com/Altinity/ClickHouse/pull/636 | 24.12.2 || | --- | |:arrow_forward: SWARMS| | || -|Use `object_storage_remote_initiator` without `object_storage_cluster` on initial node|https://github.com/Altinity/ClickHouse/pull/1608|26.1.11|| +|Initiator-local JOIN and IN with `object_storage_cluster_join_mode=allow`, including the old analyzer|https://github.com/Altinity/ClickHouse/pull/2228|26.6.8|| +|Binary Antalya protocol for Iceberg metadata and task-distribution settings|https://github.com/Altinity/ClickHouse/pull/2378, https://github.com/Altinity/ClickHouse/pull/2482, https://github.com/Altinity/ClickHouse/pull/2489|26.6.8|| +|Allow empty object-storage clusters|https://github.com/Altinity/ClickHouse/pull/2221|26.6.8|| +|Reload cluster discovery settings from `remote_servers` without restart|https://github.com/Altinity/ClickHouse/pull/2197|26.6.4|https://github.com/ClickHouse/ClickHouse/pull/113669| +|Use `object_storage_remote_initiator` without `object_storage_cluster` on the initial node|https://github.com/Altinity/ClickHouse/pull/1608|26.1.11|| |Cluster Joins part 2 - global mode|https://github.com/Altinity/ClickHouse/pull/1527|26.1.11|| -|Improvement for locking Iceberg iterator|https://github.com/Altinity/ClickHouse/pull/1436|25.8.16|| +|Remote initiator credential and query-setting propagation|https://github.com/Altinity/ClickHouse/pull/1577|26.1.6|| |Profile events for task distribution in ObjectStorageCluster requests|https://github.com/Altinity/ClickHouse/pull/1172|25.8.12|| -| SYSTEM STOP SWARM MODE command for graceful shutdown of swarm node|https://github.com/Altinity/ClickHouse/pull/1014|25.6.5|| +| SYSTEM STOP SWARM MODE command for graceful shutdown of a swarm node|https://github.com/Altinity/ClickHouse/pull/1014|25.6.5|| | JOIN with *Cluster table functions and swarm queries|https://github.com/Altinity/ClickHouse/pull/972|25.6.5|| -| Restart cluster tasks on connection lost|https://github.com/Altinity/ClickHouse/pull/780|25.6.5|| -| Add `iceberg_metadata_file_path` to query when sent to swarm nodes|https://github.com/Altinity/ClickHouse/pull/898|25.3.3|| -| Add setting `lock_object_storage_task_distribution_ms` to improve cache locality with swarm cluster|https://github.com/Altinity/ClickHouse/pull/866|25.3.3|| -| Add setting `object_storage_max_nodes` | https://github.com/Altinity/ClickHouse/pull/677 | 25.2.2 || -| Convert functions with `object_storage_cluster` setting to cluster functions | https://github.com/Altinity/ClickHouse/pull/712 | 25.2.2 || -| Fix remote call of s3Cluster function | https://github.com/Altinity/ClickHouse/pull/583 | 24.12.2 || -| Limit parsing threads for distributed case | https://github.com/Altinity/ClickHouse/pull/648 | 24.12.2 || -| Distributed request to tables with Object Storage Engines | https://github.com/Altinity/ClickHouse/pull/615 | 24.12.2 || +| Restart cluster tasks on connection loss|https://github.com/Altinity/ClickHouse/pull/780|25.6.5|| +| Send `iceberg_metadata_file_path` to swarm nodes|https://github.com/Altinity/ClickHouse/pull/898|25.3.3|| +| Improve swarm cache locality with `lock_object_storage_task_distribution_ms`|https://github.com/Altinity/ClickHouse/pull/866|25.3.3|| +|Remote initiator selection with `object_storage_remote_initiator`|https://github.com/Altinity/ClickHouse/pull/756|25.3.3|| +| Limit participating nodes with `object_storage_max_nodes` | https://github.com/Altinity/ClickHouse/pull/677 | 25.2.2 |https://github.com/ClickHouse/ClickHouse/pull/74525| +| Convert functions with `object_storage_cluster` to cluster functions | https://github.com/Altinity/ClickHouse/pull/712 | 25.2.2 || +| Distributed requests to tables with Object Storage Engines | https://github.com/Altinity/ClickHouse/pull/615 | 24.12.2 |https://github.com/ClickHouse/ClickHouse/pull/76516| +| Alternative syntax for object-storage cluster functions |https://github.com/Altinity/ClickHouse/pull/592 |24.12.2 |https://github.com/ClickHouse/ClickHouse/pull/74615| | --- | |:arrow_forward: CATALOGS| | || -|Enable PREWHERE optimization for Iceberg tables|https://github.com/Altinity/ClickHouse/pull/1581|26.1.6|https://github.com/ClickHouse/ClickHouse/pull/95476|26.2.1| -|Google biglake catalog integration|https://github.com/Altinity/ClickHouse/pull/1528|26.1.6|https://github.com/ClickHouse/ClickHouse/pull/97104|26.2.1| -|Enabled experimental datalake catalogues by-default|https://github.com/Altinity/ClickHouse/pull/1504|26.1.6|| -|Add role-based access to Glue catalog|https://github.com/Altinity/ClickHouse/pull/1427|26.1.6|https://github.com/ClickHouse/ClickHouse/pull/90825|26.2.1| -|Add role-based access to Glue catalog|https://github.com/Altinity/ClickHouse/pull/1428|25.8.16|https://github.com/ClickHouse/ClickHouse/pull/90825|26.2.1| -|Add setting `iceberg_partition_timezone`|https://github.com/Altinity/ClickHouse/pull/1349|25.8.16|| +|Iceberg DROP PARTITION for catalog tables, data purge, mixed manifests, equality deletes and partition evolution|https://github.com/Altinity/ClickHouse/pull/2613||| +|Apache DataSketches-compatible `uniqApacheHLL` aggregate function and sketch import/export helpers|https://github.com/Altinity/ClickHouse/pull/2398, https://github.com/Altinity/ClickHouse/pull/2544|26.6.8|https://github.com/ClickHouse/ClickHouse/pull/124667| +|Read Iceberg v3 multi-column transforms with `source-ids`|https://github.com/Altinity/ClickHouse/pull/2318|26.6.8|| +|Read and write the Iceberg v3 `unknown` data type|https://github.com/Altinity/ClickHouse/pull/2363|26.6.8|| +|Native CREATE TABLE, CREATE TABLE AS and DROP TABLE in DataLakeCatalog, with optional data purge|https://github.com/Altinity/ClickHouse/pull/2305|26.6.8|https://github.com/ClickHouse/ClickHouse/pull/98670| +|FIRST and AFTER clauses for Iceberg ADD COLUMN and MODIFY COLUMN|https://github.com/Altinity/ClickHouse/pull/2289|26.6.8|| +|Support for Databricks deletion vectors|https://github.com/Altinity/ClickHouse/pull/2271|26.6.4|| +|Data lake catalog authentication-token profile events|https://github.com/Altinity/ClickHouse/pull/2222|26.6.4|| +|Iceberg deletion vectors support|https://github.com/Altinity/ClickHouse/pull/2183|26.6.2|https://github.com/ClickHouse/ClickHouse/pull/113561| +|Cache REST catalog vended credentials with `vended_credentials_cache_ttl`|https://github.com/Altinity/ClickHouse/pull/1923|26.3.13|https://github.com/ClickHouse/ClickHouse/pull/107960| +|Profile events for Iceberg REST catalogs|https://github.com/Altinity/ClickHouse/pull/1868|26.3.13|https://github.com/ClickHouse/ClickHouse/pull/107194| +|Read and write the Iceberg `time` type|https://github.com/Altinity/ClickHouse/pull/1761|26.3.10|https://github.com/ClickHouse/ClickHouse/pull/105540| |DataLakeCatalog namespace filter|https://github.com/Altinity/ClickHouse/pull/1337|25.8.16|https://github.com/ClickHouse/ClickHouse/pull/95388| -|Add metrics for Iceberg, S3, and Azure|https://github.com/Altinity/ClickHouse/pull/1123|25.8.9|https://github.com/ClickHouse/ClickHouse/pull/93332| -|Add setting `iceberg_timezone_for_timestamptz` for Iceberg TimestampTZ type|https://github.com/Altinity/ClickHouse/pull/1103|25.8.9|| -|Allow to read Iceberg data from any location|https://github.com/Altinity/ClickHouse/pull/1092|25.8.9|| -|Read optimization using Iceberg metadata|https://github.com/Altinity/ClickHouse/pull/1019|25.8.9|| -|Lazy load metadata for metadata for DataLake|https://github.com/Altinity/ClickHouse/pull/742|25.8.9|| -| Expose IcebergS3 table metainformation in system.tables|https://github.com/Altinity/ClickHouse/pull/959|25.6.5|| -| Support different warehouses behind Iceberg REST catalog|https://github.com/Altinity/ClickHouse/pull/860|25.3.3|| -| General engine definition for Iceberg tables | https://github.com/Altinity/ClickHouse/pull/675 | 25.2.2 || -| RBAC for S3 | https://github.com/Altinity/ClickHouse/pull/688 | 25.2.2 || +|Set the Iceberg partition time zone with `iceberg_partition_timezone`|https://github.com/Altinity/ClickHouse/pull/1349|25.8.16|https://github.com/ClickHouse/ClickHouse/pull/95659| +|Metrics for Iceberg, S3 and Azure|https://github.com/Altinity/ClickHouse/pull/1123|25.8.9|https://github.com/ClickHouse/ClickHouse/pull/93332| +|Set the Iceberg TimestampTZ time zone with `iceberg_timezone_for_timestamptz`|https://github.com/Altinity/ClickHouse/pull/1103|25.8.9|https://github.com/ClickHouse/ClickHouse/pull/94025| +|Read Iceberg data from external locations and different object stores|https://github.com/Altinity/ClickHouse/pull/1092|25.8.9|https://github.com/ClickHouse/ClickHouse/pull/90740| +| Expose IcebergS3 partition and sorting keys in system.tables|https://github.com/Altinity/ClickHouse/pull/959|25.6.5|https://github.com/ClickHouse/ClickHouse/pull/85571| +| Support different warehouses behind the Iceberg REST catalog|https://github.com/Altinity/ClickHouse/pull/860|25.3.3|| +|Iceberg database alias for DataLakeCatalog with a REST catalog|https://github.com/Altinity/ClickHouse/pull/822|25.3.3|| +| General engine definition for Iceberg tables | https://github.com/Altinity/ClickHouse/pull/675 | 25.2.2 |https://github.com/ClickHouse/ClickHouse/pull/77125| +| Role-based access for S3 | https://github.com/Altinity/ClickHouse/pull/688 | 25.2.2 |https://github.com/ClickHouse/ClickHouse/pull/77747| | --- | |:arrow_forward: TIERED STORAGE| | || -|Add support for moving Hybrid table watermarks|https://github.com/Altinity/ClickHouse/pull/1659|26.1.11|| -|Make changes to the export partition background engine and support experimental exports to Apache Iceberg|https://github.com/Altinity/ClickHouse/pull/1618|26.1.11|| -|Add TRUNCATE TABLE support for the Iceberg database engine|https://github.com/Altinity/ClickHouse/pull/1529|26.1.6|| -|Add setting to define filename pattern for part exports|https://github.com/Altinity/ClickHouse/pull/1490|26.1.6|| -|Add setting to define filename pattern for part exports|https://github.com/Altinity/ClickHouse/pull/1512|25.8.16|| -|Add query id to `system.part_log`, `system_exports` and `system.replicated_partition_exports`|https://github.com/Altinity/ClickHouse/pull/1330|25.8.14|| -|Allow merge tree materialized / alias columns to be exported through part export|https://github.com/Altinity/ClickHouse/pull/1324|25.8.14|| -|Accept table function as destination for part export|https://github.com/Altinity/ClickHouse/pull/1320|25.8.14|| -|Add settings to control the behavior of pending mutations / patch parts on the export part feature|https://github.com/Altinity/ClickHouse/pull/1294|25.8.14|| -|Add support for ALIAS columns in segments for `engine=Hybrid`|https://github.com/Altinity/ClickHouse/pull/1272|25.8.14|| -|Add ability to split large parquet files on part export|https://github.com/Altinity/ClickHouse/pull/1229|25.8.12|| -|Add experimental support to automatically reconcile column-type mismatches across segments in Hybrid table engine |https://github.com/Altinity/ClickHouse/pull/1156|25.8.9|| -|Preserve the entire format settings object in export part manifest|https://github.com/Altinity/ClickHouse/pull/1144|25.8.9|| -|Export partition support for ReplicatedMergeTree engine|https://github.com/Altinity/ClickHouse/pull/1124|25.8.9|| -|Allow any partition strategy to accept part export|https://github.com/Altinity/ClickHouse/pull/1083|25.8.9|| -|`Engine=hybrid` implementation|https://github.com/Altinity/ClickHouse/pull/1071|25.8.9|| -|Add observability for EXPORT PART|https://github.com/Altinity/ClickHouse/pull/1017|25.6.5|| -| Simple MergeTree part export to object storage|https://github.com/Altinity/ClickHouse/pull/1009|25.6.5|| -| s3Cluster hive partitioning for old analyzer | https://github.com/Altinity/ClickHouse/pull/703 | 25.2.2 |https://github.com/ClickHouse/ClickHouse/pull/93284|26.4.1| +|Iceberg data-file compaction with bin-packing RewriteFiles|https://github.com/Altinity/ClickHouse/pull/2420|26.6.8|| +|Export partitions from non-replicated MergeTree tables|https://github.com/Altinity/ClickHouse/pull/2290|26.6.8|| +|Safe unsigned-to-signed type promotion for part and partition exports|https://github.com/Altinity/ClickHouse/pull/2559|26.6.8|| +|Export across compatible, non-matching partition expressions|https://github.com/Altinity/ClickHouse/pull/2074|26.6.2|| +|Match export columns by name or position and optionally ignore extra source columns|https://github.com/Altinity/ClickHouse/pull/2111, https://github.com/Altinity/ClickHouse/pull/2229, https://github.com/Altinity/ClickHouse/pull/2220|26.6.2|| +|Expose Iceberg metadata and data-file paths in system.replicated_partition_exports|https://github.com/Altinity/ClickHouse/pull/1832|26.6.2|| +|Automatic type conversion and optional lossy casts for part and partition exports|https://github.com/Altinity/ClickHouse/pull/1779|26.3.13|| +|EXPORT PARTITION ALL|https://github.com/Altinity/ClickHouse/pull/1741|26.3.10|| +|Prune unreachable Hybrid table segments using query predicates|https://github.com/Altinity/ClickHouse/pull/1788|26.3.10|| +|Inspect Hybrid table watermarks in system.hybrid_watermarks|https://github.com/Altinity/ClickHouse/pull/1789|26.3.10|| +|Move Hybrid table watermarks with ALTER TABLE MODIFY SETTING|https://github.com/Altinity/ClickHouse/pull/1659|26.1.11|| +|Background partition export to Apache Iceberg|https://github.com/Altinity/ClickHouse/pull/1618|26.1.11|| +|Improve partition-export scheduling and system.replicated_partition_exports performance|https://github.com/Altinity/ClickHouse/pull/1402|26.1.6|| +|TRUNCATE TABLE for the Iceberg database engine with REST catalogs|https://github.com/Altinity/ClickHouse/pull/1529|26.1.6|https://github.com/ClickHouse/ClickHouse/pull/102065| +|Define the filename pattern for part exports|https://github.com/Altinity/ClickHouse/pull/1490, https://github.com/Altinity/ClickHouse/pull/1512|25.8.16|| +|Add query IDs to system.part_log, system_exports and system.replicated_partition_exports|https://github.com/Altinity/ClickHouse/pull/1330|25.8.14|| +|Export MergeTree materialized and ALIAS columns|https://github.com/Altinity/ClickHouse/pull/1324|25.8.14|| +|Use a table function as the destination for part export|https://github.com/Altinity/ClickHouse/pull/1320|25.8.14|| +|Control export behavior for pending mutations and patch parts|https://github.com/Altinity/ClickHouse/pull/1294|25.8.14|| +|Support ALIAS columns in Hybrid table segments|https://github.com/Altinity/ClickHouse/pull/1272|25.8.14|| +|Split large Parquet files during part export|https://github.com/Altinity/ClickHouse/pull/1229|25.8.12|| +|Automatically reconcile column-type mismatches across Hybrid table segments |https://github.com/Altinity/ClickHouse/pull/1156|25.8.9|| +|Preserve format settings in the export-part manifest|https://github.com/Altinity/ClickHouse/pull/1144|25.8.9|| +|Partition export for ReplicatedMergeTree|https://github.com/Altinity/ClickHouse/pull/1124|25.8.9|| +|Allow any partition strategy for part export|https://github.com/Altinity/ClickHouse/pull/1083|25.8.9|| +|Hybrid table engine|https://github.com/Altinity/ClickHouse/pull/1071|25.8.9|| +|Observability for EXPORT PART|https://github.com/Altinity/ClickHouse/pull/1017|25.6.5|| +| MergeTree part export to object storage|https://github.com/Altinity/ClickHouse/pull/1009|25.6.5|https://github.com/ClickHouse/ClickHouse/pull/91228| +| --- | +|:arrow_forward: CAS| | || +|Reduce CAS write round trips: INSERT uses 14 object-store requests instead of 25|https://github.com/Altinity/ClickHouse/pull/2831||| +|Rename CAS parts with one ref-log owner Move instead of republishing manifests|https://github.com/Altinity/ClickHouse/pull/2647||| +|Packed parts on CAS disks|https://github.com/Altinity/ClickHouse/pull/2579||| +|Rejoin a decommissioned CAS server root as an empty pool member on restart|https://github.com/Altinity/ClickHouse/pull/2574|26.6.8|| +|Live mount-renewal progress and error columns in system.cas_mounts|https://github.com/Altinity/ClickHouse/pull/2513|26.6.8|| +|Parallel blob deletion during garbage collection with `cas_gc_io_concurrency`|https://github.com/Altinity/ClickHouse/pull/2351|26.6.8|| +|Answer in-part directory probes from manifests, avoiding per-file object-store LIST requests|https://github.com/Altinity/ClickHouse/pull/2440|26.6.8|| +|Remove a CAS part with one ref-log record|https://github.com/Altinity/ClickHouse/pull/2452|26.6.8|| +|Batch garbage-collection namespace cleanup under a soft time budget|https://github.com/Altinity/ClickHouse/pull/2483|26.6.8|| +|EXPORT PARTITION from CAS disks and major garbage-collection round-trip reductions|https://github.com/Altinity/ClickHouse/pull/2300|26.6.4|| +|Configure CAS mount lease TTL and renewal period|https://github.com/Altinity/ClickHouse/pull/2307|26.6.4|| +|Experimental content-addressed MergeTree storage with shared S3/GCS blobs, fetch-by-relink, garbage collection, FSCK and system tables|https://github.com/Altinity/ClickHouse/pull/2159|26.6.2|| | --- | |:left_right_arrow: Feature parity in the latest Antalya release :left_right_arrow:||| | --- | -|Datalakes catalogs database for distributed processing|https://github.com/Altinity/ClickHouse/pull/1458|25.8.16|https://github.com/ClickHouse/ClickHouse/pull/88273|25.10.1| -|Add `icebergLocalCluster` table function to allow cluster reads from shared disk|https://github.com/Altinity/ClickHouse/pull/1120|25.8.9|https://github.com/ClickHouse/ClickHouse/pull/93323|26.1.1| -|Google cloud storage support for data lakes|https://github.com/Altinity/ClickHouse/pull/1318|25.8.14|https://github.com/ClickHouse/ClickHouse/pull/93866|26.1.1|| -| Set max message size on parquet v3 reader |https://github.com/Altinity/ClickHouse/pull/1198 | 25.8.12 |https://github.com/ClickHouse/ClickHouse/pull/91737|25.12.1| -| Distributed execution: better split tasks by row groups IDs |https://github.com/Altinity/ClickHouse/pull/1237 | 25.8.12 |https://github.com/ClickHouse/ClickHouse/pull/87508|25.11.1| -| Enable parquet reader v3 by default |https://github.com/Altinity/ClickHouse/pull/1232 | 25.8.12 |https://github.com/ClickHouse/ClickHouse/pull/88827|25.11.1| -| Allow key-value arguments in s3/s3cluster engine |https://github.com/Altinity/ClickHouse/pull/1028 |25.6.5|https://github.com/ClickHouse/ClickHouse/pull/85134|25.8.1| -| AWS S3 authentication with an explicitly provided IAM role |https://github.com/Altinity/ClickHouse/pull/986 |25.6.5 |https://github.com/ClickHouse/ClickHouse/pull/84011|25.8.1| -| Support for hive partition style reads and writes |https://github.com/Altinity/ClickHouse/pull/934 |25.6.5|https://github.com/ClickHouse/ClickHouse/pull/76802|25.8.1| -| Support compressed metadata in Iceberg |https://github.com/Altinity/ClickHouse/pull/1005 |25.6.5|https://github.com/ClickHouse/ClickHouse/pull/81451|25.7.1| -| Support TimestampTZ in Glue catalog |https://github.com/Altinity/ClickHouse/pull/992 |25.6.5|https://github.com/ClickHouse/ClickHouse/issues/83132|25.7.1| -| Support writing parquet enum as byte array |https://github.com/Altinity/ClickHouse/pull/989 |25.6.5 |https://github.com/ClickHouse/ClickHouse/pull/81090|25.7.1| -| Iceberg table pruning in cluster requests |https://github.com/Altinity/ClickHouse/pull/770 |25.2.2 |https://github.com/ClickHouse/ClickHouse/pull/82131|25.7.1| -| Support for Iceberg partition pruning bucket transform |https://github.com/Altinity/ClickHouse/pull/786 |25.3.3|https://github.com/ClickHouse/ClickHouse/pull/79262|25.5.1| -| Improve performance of hive path parsing |https://github.com/Altinity/ClickHouse/pull/734 |25.2.2 |https://github.com/ClickHouse/ClickHouse/pull/79067|25.5.1| -| Iceberg metadata files cache |https://github.com/Altinity/ClickHouse/pull/733 |25.2.2 |https://github.com/ClickHouse/ClickHouse/pull/77156 |25.5.1| -| Rendezvous hashing filesystem cache |https://github.com/Altinity/ClickHouse/pull/709 |25.2.2 |https://github.com/ClickHouse/ClickHouse/pull/77326|25.5.1| -| Better S3 URL parsing for Hive partitioning |https://github.com/Altinity/ClickHouse/pull/700 |25.2.2 |https://github.com/ClickHouse/ClickHouse/pull/78185|25.5.1| -| Support MinMax index for Iceberg |https://github.com/Altinity/ClickHouse/pull/733 |25.2.2 |https://github.com/ClickHouse/ClickHouse/pull/78242|25.4.1| -| Support partition pruning in DeltaLake engine |https://github.com/Altinity/ClickHouse/pull/733 |25.2.2 |https://github.com/ClickHouse/ClickHouse/pull/78486|25.4.1| -| Iceberg time travel by snapshots |https://github.com/Altinity/ClickHouse/pull/733 |25.2.2 |https://github.com/ClickHouse/ClickHouse/pull/77439|25.4.1| -| Cluster auto discovery |https://github.com/Altinity/ClickHouse/pull/629 |24.12.2 |https://github.com/ClickHouse/ClickHouse/pull/76001|25.3.1| -| Alternative syntax for object storage cluster functions |https://github.com/Altinity/ClickHouse/pull/592 |24.12.2 |https://github.com/ClickHouse/ClickHouse/pull/70659|25.3.1| -| Unity catalog integration |same as upstream => |25.3.3 |https://github.com/ClickHouse/ClickHouse/pull/76988|25.3.1| -| Glue catalog integration |same as upstream => |25.3.3 |https://github.com/ClickHouse/ClickHouse/pull/77257|25.3.1| -| Parquet: merge bloom filter and min/max evaluation |https://github.com/Altinity/ClickHouse/pull/590 |24.12.2 |https://github.com/ClickHouse/ClickHouse/pull/71383|25.2.1| -| Parquet: Int logical type support on native reader |https://github.com/Altinity/ClickHouse/pull/589 |24.12.2 |https://github.com/ClickHouse/ClickHouse/pull/72105|25.1.1| -| Iceberg REST Catalog integration |same as upstream => |24.12.2 |https://github.com/ClickHouse/ClickHouse/pull/71542|24.12.1| -| Parquet: boolean support on native reader |same as upstream => |24.12.2 |https://github.com/ClickHouse/ClickHouse/pull/71055|24.11.1| -| Auxiliary autodiscovery |https://github.com/Altinity/ClickHouse/pull/531 |24.12.2 |https://github.com/ClickHouse/ClickHouse/pull/71911|24.11.1| -| Parquet: bloom filters support |same as upstream => |24.12.2 |https://github.com/ClickHouse/ClickHouse/pull/62966|24.10.1| -| Parquet: page header v2 support on native reader |same as upstream => |24.12.2 |https://github.com/ClickHouse/ClickHouse/pull/70807|24.10.1| +|Adaptive Parquet footer reads to reduce storage round trips|https://github.com/Altinity/ClickHouse/pull/2503|26.6.8|https://github.com/ClickHouse/ClickHouse/pull/115686|26.10.2| +|Puffin file format reader|https://github.com/Altinity/ClickHouse/pull/2179|26.6.2|https://github.com/ClickHouse/ClickHouse/pull/109990, https://github.com/ClickHouse/ClickHouse/pull/103936|26.8.1| +|Amazon S3 Tables catalog support|https://github.com/Altinity/ClickHouse/pull/1808|26.3.10|https://github.com/ClickHouse/ClickHouse/pull/103220, https://github.com/ClickHouse/ClickHouse/pull/113505|26.8.1| +|Share S3 client caches per bucket to avoid repeated region discovery|https://github.com/Altinity/ClickHouse/pull/1807|26.3.10|https://github.com/ClickHouse/ClickHouse/pull/96802|26.6.1| +|S3Queue auxiliary ZooKeeper support with keeper_path|https://github.com/Altinity/ClickHouse/pull/2115|25.8.33|https://github.com/ClickHouse/ClickHouse/pull/95203|26.2.1| +|EXECUTE AS for user impersonation|https://github.com/Altinity/ClickHouse/pull/2165|25.8.33|https://github.com/ClickHouse/ClickHouse/pull/70775|25.11.1| +|icebergLocalCluster table function for cluster reads from shared disks|https://github.com/Altinity/ClickHouse/pull/1120|25.8.9|https://github.com/ClickHouse/ClickHouse/pull/93323|26.1.1| +| Hive partition style reads and writes |https://github.com/Altinity/ClickHouse/pull/934 |25.6.5|https://github.com/ClickHouse/ClickHouse/pull/76802|25.8.1| +| Write Parquet enums as byte arrays |https://github.com/Altinity/ClickHouse/pull/989 |25.6.5 |https://github.com/ClickHouse/ClickHouse/pull/81090|25.7.1| +| Rendezvous hashing for object-storage filesystem-cache locality |https://github.com/Altinity/ClickHouse/pull/709 |25.2.2 |https://github.com/ClickHouse/ClickHouse/pull/82511|25.8.1| +| Faster Hive path parsing with extractKeyValuePairs |https://github.com/Altinity/ClickHouse/pull/734 |25.2.2 |https://github.com/ClickHouse/ClickHouse/pull/79067|25.5.1| +| Dynamic cluster autodiscovery |https://github.com/Altinity/ClickHouse/pull/629 |24.12.2 |https://github.com/ClickHouse/ClickHouse/pull/76001|25.3.1| +| Auxiliary ZooKeeper autodiscovery |https://github.com/Altinity/ClickHouse/pull/531 |24.12.2 |https://github.com/ClickHouse/ClickHouse/pull/71911|24.11.1| +| s3Cluster Hive partition pruning, including the old analyzer | https://github.com/Altinity/ClickHouse/pull/584, https://github.com/Altinity/ClickHouse/pull/703 | 24.12.2 |https://github.com/ClickHouse/ClickHouse/pull/93284|26.4.1| +| Merge Parquet bloom-filter and min/max evaluation |https://github.com/Altinity/ClickHouse/pull/590 |24.12.2 |https://github.com/ClickHouse/ClickHouse/pull/71383|25.2.1| +| Parquet integer logical types on the native reader |https://github.com/Altinity/ClickHouse/pull/589 |24.12.2 |https://github.com/ClickHouse/ClickHouse/pull/72105|25.1.1|