From 35ba11ca4c4bfffd1c82bd3887ddef00d0711537 Mon Sep 17 00:00:00 2001 From: atovpeko Date: Fri, 14 Nov 2025 12:03:28 +0200 Subject: [PATCH 1/3] changelog --- about/changelog.md | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/about/changelog.md b/about/changelog.md index 189b01d973..0771367b7e 100644 --- a/about/changelog.md +++ b/about/changelog.md @@ -9,6 +9,29 @@ products: [cloud] All the latest features and updates to $CLOUD_LONG. +## TimescaleDB v2.23 – improved getting started, automatic columnstore, and faster continuous aggregate invalidation tracking + + +TimescaleDB v2.23 was released on October 29th and is now available on Tiger Cloud. + +### Highlighted features in TimescaleDB v2.23.0 + +- Simplified hypertable creation: starting with TimescaleDB is now even easier with less configuration and smarter defaults. [Create a hypertable](https://docs.tigerdata.com/api/latest/hypertable/create_table/#samples) in only one step with automatic selection of all parameters. This includes: + - Automatic selection of the partitioning column: TimescaleDB automatically selects a partitioning column so you no longer need to specify which column to use during creation. + - Automatic columnstore policy: TimescaleDB enables the columnstore by default and automatically creates a columnstore policy that runs after one chunk interval (defaults to 7 days). +- UUIDv7 compression enabled by default: the UUIDv7 vectorized compression and query acceleration introduced in 2.22 are now enabled by default. You automatically benefit from ~30% storage savings and up to 2x faster query performance. +- Direct-to-Columnstore (tech preview): this release adds `INSERT` support to Direct-to-Columnstore (Direct Compress), which previously supported only `COPY`. For more information, see our [documentation](https://docs.tigerdata.com/use-timescale/latest/write-data/insert/#direct-compress-on-insert) and [blog post](https://www.tigerdata.com/blog/introducing-direct-compress-up-to-40x-faster-leaner-data-ingestion-for-developers-tech-preview). +- Relaxed locking for chunk merging: concurrency during chunk merges has been improved to eliminate potential deadlocks, enabling safer, faster background merges. +- Unlogged hypertables: added the ability to set hypertables as `UNLOGGED`, improving insert and update performance for workloads where durability is not required. Ideal for large imports or transient datasets. +- Continuous aggregates improvements: continuous aggregates no longer use triggers for invalidation tracking. The approach has been refactored resulting in 10–20% faster DML performance. [Set-returning functions](https://www.postgresql.org/docs/current/functions-srf.html) are now supported in continuous aggregate materialization queries ([community request #1717](https://github.com/timescale/timescaledb/issues/1717)). + +### Deprecations + +- PostgreSQL 15 deprecation: TimescaleDB will continue supporting PostgreSQL 15 until June 2026, after which support will be removed. We recommend that you begin planning upgrades to PostgreSQL 16 or higher to ensure continued access to performance improvements, security updates, and new TimescaleDB features. See [Supported platforms](https://docs.tigerdata.com/about/latest/supported-platforms/#postgres-timescaledb-support-matrix) for currently supported versions. +- WAL-based invalidation: introduced as a tech preview in 2.22, WAL-based invalidation will be sunset in the upcoming releases. The approach was not the right architecture to address customers hitting IOPS limits during continuous aggregate invalidation tracking. This release already removes the trigger, and we are adding more improvements in upcoming releases to address IOPS. The first step is to gate the feature behind a GUC, and remove it in an upcoming release. + +For a comprehensive list of changes, refer to the [TimescaleDB 2.23 release notes](https://github.com/timescale/timescaledb/blob/main/CHANGELOG.md#2230-2025-10-29). + ## S3 source connector and crypto payments From 36e6f2f84f5ba173f81e269f288abf34cc8fb908 Mon Sep 17 00:00:00 2001 From: Anastasiia Tovpeko <114177030+atovpeko@users.noreply.github.com> Date: Fri, 14 Nov 2025 14:03:50 +0200 Subject: [PATCH 2/3] Update about/changelog.md Co-authored-by: Iain Cox Signed-off-by: Anastasiia Tovpeko <114177030+atovpeko@users.noreply.github.com> --- about/changelog.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/about/changelog.md b/about/changelog.md index 0771367b7e..ff8490bf5f 100644 --- a/about/changelog.md +++ b/about/changelog.md @@ -16,7 +16,7 @@ TimescaleDB v2.23 was released on October 29th and is now available on Tiger Clo ### Highlighted features in TimescaleDB v2.23.0 -- Simplified hypertable creation: starting with TimescaleDB is now even easier with less configuration and smarter defaults. [Create a hypertable](https://docs.tigerdata.com/api/latest/hypertable/create_table/#samples) in only one step with automatic selection of all parameters. This includes: +- Simplified hypertable creation: now even easier with less configuration and smarter defaults. [Create a hypertable](https://docs.tigerdata.com/api/latest/hypertable/create_table/#samples) in only one step with automatic selection of all parameters. This includes: - Automatic selection of the partitioning column: TimescaleDB automatically selects a partitioning column so you no longer need to specify which column to use during creation. - Automatic columnstore policy: TimescaleDB enables the columnstore by default and automatically creates a columnstore policy that runs after one chunk interval (defaults to 7 days). - UUIDv7 compression enabled by default: the UUIDv7 vectorized compression and query acceleration introduced in 2.22 are now enabled by default. You automatically benefit from ~30% storage savings and up to 2x faster query performance. From 2f12e2457eb44d0d3bd8289b3f7deddfb1a76d6e Mon Sep 17 00:00:00 2001 From: Anastasiia Tovpeko <114177030+atovpeko@users.noreply.github.com> Date: Fri, 14 Nov 2025 14:04:00 +0200 Subject: [PATCH 3/3] Update about/changelog.md Co-authored-by: Iain Cox Signed-off-by: Anastasiia Tovpeko <114177030+atovpeko@users.noreply.github.com> --- about/changelog.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/about/changelog.md b/about/changelog.md index ff8490bf5f..5ff0d26abf 100644 --- a/about/changelog.md +++ b/about/changelog.md @@ -27,7 +27,7 @@ TimescaleDB v2.23 was released on October 29th and is now available on Tiger Clo ### Deprecations -- PostgreSQL 15 deprecation: TimescaleDB will continue supporting PostgreSQL 15 until June 2026, after which support will be removed. We recommend that you begin planning upgrades to PostgreSQL 16 or higher to ensure continued access to performance improvements, security updates, and new TimescaleDB features. See [Supported platforms](https://docs.tigerdata.com/about/latest/supported-platforms/#postgres-timescaledb-support-matrix) for currently supported versions. +- Postgres 15 deprecation: TimescaleDB will continue supporting Postgres 15 until June 2026, after which support will be removed. We recommend that you begin planning upgrades to Postgres 16 or higher to ensure continued access to performance improvements, security updates, and new TimescaleDB features. See [Supported platforms](https://docs.tigerdata.com/about/latest/supported-platforms/#postgres-timescaledb-support-matrix) for currently supported versions. - WAL-based invalidation: introduced as a tech preview in 2.22, WAL-based invalidation will be sunset in the upcoming releases. The approach was not the right architecture to address customers hitting IOPS limits during continuous aggregate invalidation tracking. This release already removes the trigger, and we are adding more improvements in upcoming releases to address IOPS. The first step is to gate the feature behind a GUC, and remove it in an upcoming release. For a comprehensive list of changes, refer to the [TimescaleDB 2.23 release notes](https://github.com/timescale/timescaledb/blob/main/CHANGELOG.md#2230-2025-10-29).