--- title: Postgres 19 switches to LZ4 by default, Dolt expands to... description: PostgreSQL 19 Beta 2 is out, and the headline storage change is the switch from pglz to LZ4 as the default TOAST compression — roughly 8x faster compression... canonical: https://daily.dev/posts/postgres-19-switches-to-lz4-by-default-dolt-expands-to-four-database-flavors-8im0oxv0g twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Postgres 19 switches to LZ4 by default, Dolt expands to four database flavors | daily.dev og:description: PostgreSQL 19 Beta 2 is out, and the headline storage change is the switch from pglz to LZ4 as the default TOAST compression — roughly 8x faster compression... og:url: https://daily.dev/posts/postgres-19-switches-to-lz4-by-default-dolt-expands-to-four-database-flavors-8im0oxv0g og:image: https://api.daily.dev/og/posts/8iM0oXV0G.png og:image:alt: Postgres 19 switches to LZ4 by default, Dolt expands to four database flavors og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Postgres 19 switches to LZ4 by default, Dolt expands to four database flavors **[Databases Digest](https://daily.dev/sources/databases_digest)** · 5 min read · 0 upvotes · 0 comments ## Summary PostgreSQL 19 Beta 2 is out, and the headline storage change is the switch from pglz to LZ4 as the default TOAST compression — roughly 8x faster compression with better ratios on modern hardware. Dolt now ships in four flavors: MySQL-compatible, Postgres-compatible, SQLite-embedded, and MongoDB-style, with Doltgres targeting 1.0 in August. A real incident post walks through manually recovering a status page from a half-finished schema migration that was OOMKilled mid-run, leaving the database in a dirty state for 38 minutes. Apache Spark 4.2 lands in Databricks Runtime 19 Beta with vector similarity search, native geometry types, and Auto CDC support. ## Content **TLDR:** PostgreSQL 19 Beta 2 is out, and the headline storage change is the switch from pglz to LZ4 as the default TOAST compression — roughly 8x faster compression with better ratios on modern hardware. Dolt now ships in four flavors: MySQL-compatible, Postgres-compatible, SQLite-embedded, and MongoDB-style, with Doltgres targeting 1.0 in August. A real incident post walks through manually recovering a status page from a half-finished schema migration that was OOMKilled mid-run, leaving the database in a dirty state for 38 minutes. Apache Spark 4.2 lands in Databricks Runtime 19 Beta with vector similarity search, native geometry types, and Auto CDC support. --- ## PostgreSQL 19 Beta 2 released, switches default TOAST compression from pglz to LZ4 PostgreSQL 19 Beta 2 is available now, with general availability expected around September/October 2026. The most consequential storage change is the default TOAST compression algorithm switching from pglz to LZ4 — about 8x faster compression and better ratios thanks to a larger 64 kB sliding window, with no external dependencies added. Beta 2 also fixes vacuumdb --analyze-in-stages for partitioned tables, pg_createsubscriber, FOR PORTION OF temporal table syntax, and postgres_fdw handling of imported foreign-table statistics. Worth testing against your workloads now, especially if you store large variable-length columns. [Read more](https://daily.dev/feed-by-ids?id=BUODRKcA3&id=p8YrK9V3C) ## Dolt expands to four database flavors, Doltgres 1.0 targeting August Dolt now comes in four variants: the original MySQL-compatible Dolt, Postgres-compatible Doltgres, SQLite-embedded DoltLite, and MongoDB-document-style Dumbo. Doltgres is the most mature of the new additions and is targeting a 1.0 release in August. DoltLite and Dumbo were built almost entirely by coding agents, which is either impressive or unsettling depending on your tolerance for that. Full ecosystem integration across all four flavors — DoltHub, DoltLab, Hosted Dolt, Dolt Workbench — is expected within 6 to 12 months. [Read more](https://daily.dev/posts/4NeBYaTpB) ## Status page down 38 minutes after OOMKilled migration left Postgres in a dirty state A Helm upgrade triggered a Postgres schema migration that was OOMKilled mid-run, leaving the database at version 23 marked applied but with no DDL committed. Both the new binary and the rolled-back binary refused to start. Rather than restoring from a 6-hour-old backup, the team manually ran the three DDL statements from migration 0023 inside a single transaction with IF NOT EXISTS guards, then cleared the dirty flag in schema_migrations — restoring the page in 51 minutes with all data intact. Post-incident fixes: a pre-upgrade Helm hook that snapshots the schema to object storage, and higher memory limits on migration jobs to prevent future OOMKills. [Read more](https://daily.dev/posts/AzumdRs2Q) ## Apache Spark 4.2 ships vector similarity search, native geometry types, and Auto CDC Spark 4.2 is available in Databricks Runtime 19 Beta with over 1,900 commits from 260+ contributors. The SQL additions most relevant to data engineers: NEAREST BY for vector similarity search, built-in GEOMETRY/GEOGRAPHY types with ST_* functions, QUALIFY for window filtering, and time-series functions. Auto CDC in Spark Declarative Pipelines handles SCD Type 1 ingestion declaratively. Arrow-optimized UDFs are now on by default, and the Arrow C Data Interface adds interoperability with Polars and DuckDB. Real-Time Mode extends to PySpark for millisecond-latency stateless streaming. [Read more](https://daily.dev/posts/Mv67Nz1Si) --- ## Also notable - **Spock 6 Beta: pgEdge multi-master replication moves progress tracking to shared memory:** Spock 6 Beta for PostgreSQL 16–19 moves replication progress tracking from a catalog table to shared memory, reducing I/O contention and bloat, adds spill-to-disk for oversized replay queues, and introduces per-subscription conflict statistics via PG18's pgstat infrastructure — breaking changes include renamed lag_tracker columns and removed GUCs (use_spi, batch_inserts, feedback_frequency). [Read more](https://daily.dev/posts/aLOFpXgFN) - **ClickHouse self-hosted S3 restore costs ~$45,000 for 500 TB due to $0.09/GB egress:** The real problem is teams skip backup testing because restores are expensive; Tigris, an S3-compatible store with zero egress fees, can replace S3 as the backup destination with no ClickHouse config changes, and the post also warns against zero-copy replication in open-source ClickHouse due to metadata race conditions that have kept it disabled since version 22.8. [Read more](https://daily.dev/posts/6bt7S6pdB) - **Manticore Search 27.x adds built-in auth with five permission types, replacing nginx workarounds:** Manticore Search 27.1.5 ships SQL/MySQL username/password auth, HTTP Basic and Bearer token auth, and five permission types (read, write, schema, replication, admin) with explicit deny rules that override any matching allow — auth is disabled by default, so existing deployments need a staged rollout before enabling in production. [Read more](https://daily.dev/feed-by-ids?id=a3bX7sPfU&id=vLgkdviLf&id=ZxwohhAyt) - **pg_hardstorage open-sourced: PostgreSQL backup tool targeting pgBackRest/Barman/WAL-G as alternatives:** CYBERTEC released pg_hardstorage under Apache 2.0 after more than a year in customer deployments; key differentiators are no archive_command requirement, backpressure via replication slots, dual-stream WAL deduplication across primary and replica, and a migration shim for side-by-side cutover. [Read more](https://daily.dev/posts/CjPow1TkD) - **Databricks sub-50ms fraud detection benchmark: 27ms median with Lakebase (Postgres) and CatBoost:** A FastAPI pipeline combining Databricks Model Serving route optimization, OAuth-authenticated connection pooling to Lakebase (Postgres), and a CatBoost model via MLflow pyfunc hit 27ms median end-to-end with single-digit millisecond feature lookups — full source on GitHub. [Read more](https://daily.dev/posts/sjy8uNUCB) ## Similar posts on daily.dev - [Postgres 19 Compression: from pglz to LZ4](https://daily.dev/posts/postgres-19-compression-from-pglz-to-lz4-p8yrk9v3c) · Crunchy Data · 3 upvotes · 0 comments --- Tags: [#database](https://daily.dev/tags/database), [#backend](https://daily.dev/tags/backend), [#postgresql](https://daily.dev/tags/postgresql), [#apache-spark](https://daily.dev/tags/apache-spark) [View this post on daily.dev](https://daily.dev/posts/postgres-19-switches-to-lz4-by-default-dolt-expands-to-four-database-flavors-8im0oxv0g) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"DiscussionForumPosting","mainEntityOfPage":"https://daily.dev/posts/postgres-19-switches-to-lz4-by-default-dolt-expands-to-four-database-flavors-8im0oxv0g","headline":"Postgres 19 switches to LZ4 by default, Dolt expands to four database flavors","text":"PostgreSQL 19 Beta 2 is out, and the headline storage change is the switch from pglz to LZ4 as the default TOAST compression — roughly 8x faster compression with better ratios on modern hardware. Dolt now ships in four flavors: MySQL-compatible, Postgres-compatible, SQLite-embedded, and MongoDB-style, with Doltgres targeting 1.0 in August. A real incident post walks through manually recovering a status page from a half-finished schema migration that was OOMKilled mid-run, leaving the database in a dirty state for 38 minutes. Apache Spark 4.2 lands in Databricks Runtime 19 Beta with vector similarity search, native geometry types, and Auto CDC support.","url":"https://daily.dev/posts/postgres-19-switches-to-lz4-by-default-dolt-expands-to-four-database-flavors-8im0oxv0g","datePublished":"2026-07-17T04:18:00.548Z","dateModified":"2026-07-17T04:18:21.553Z","author":{"@type":"Organization","name":"Databases Digest","logo":"https://media.daily.dev/image/upload/s--ir1hdV8b--/f_auto,q_auto/v1773839400/logos/databases_digest?_a=BAMAMiiu0","url":"https://daily.dev/sources/databases_digest"},"interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"isPartOf":{"@type":"WebPage","url":"https://daily.dev/sources/databases_digest","name":"Databases Digest"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Databases Digest","item":"https://daily.dev/sources/databases_digest"},{"@type":"ListItem","position":3,"name":"Postgres 19 switches to LZ4 by default, Dolt expands to four database flavors"}]} ```