<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/databricks-bets-on-ltap-postgresql-19-beta-ships-sql-graph-queries-oqdztqft7" -->

---
title: Databricks bets on LTAP, PostgreSQL 19 Beta ships SQL...
description: A curated roundup of database and data platform news. Databricks unveiled LTAP (Lake Transactional/Analytical Processing) at its Data + AI Summit, combining...
canonical: https://daily.dev/posts/databricks-bets-on-ltap-postgresql-19-beta-ships-sql-graph-queries-oqdztqft7
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: Databricks bets on LTAP, PostgreSQL 19 Beta ships SQL graph queries | daily.dev
og:description: A curated roundup of database and data platform news. Databricks unveiled LTAP (Lake Transactional/Analytical Processing) at its Data + AI Summit, combining...
og:url: https://daily.dev/posts/databricks-bets-on-ltap-postgresql-19-beta-ships-sql-graph-queries-oqdztqft7
og:image: https://api.daily.dev/og/posts/OqdzTQft7.png
og:image:alt: Databricks bets on LTAP, PostgreSQL 19 Beta ships SQL graph queries
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.

# Databricks bets on LTAP, PostgreSQL 19 Beta ships SQL graph queries

**[Databases Digest](https://daily.dev/sources/databases_digest)** · 5 min read · 0 upvotes · 0 comments

## Summary

A curated roundup of database and data platform news. Databricks unveiled LTAP (Lake Transactional/Analytical Processing) at its Data + AI Summit, combining OLTP and OLAP on a single lakehouse storage layer with Postgres-based Lakebase and sub-100ms analytics. PostgreSQL 19 Beta 1 ships SQL Property Graph Queries for graph traversals over relational tables, REPACK CONCURRENTLY for zero-downtime table rebuilds, and up to 2x faster FK inserts. Similarweb's migration of 255TB and 7 billion records from HBase to DynamoDB achieved a 20x cost reduction through careful composite key design. POSETTE 2026 talks covered random_page_cost tuning for SSDs, MCP server security patterns for Postgres, and timezone data corruption risks. Notable releases include pgstream v1.1.0 fixing bigint silent corruption, pg_kpart v1.0 for partition key enforcement, Debezium 3.6.0.Beta2, and a Coinbase postmortem on a Raft quorum loss from a single-AZ cooling failure.

## Content

**TLDR:** Databricks dominated the day with its Data + AI Summit announcements, centering on LTAP — an architecture that stores data once in a lakehouse while running separate OLTP and OLAP compute engines on top. PostgreSQL 19 Beta 1 is out with SQL Property Graph Queries and REPACK CONCURRENTLY for online table rebuilds. On the Postgres tooling front, POSETTE 2026 produced a dense set of talks worth working through, from random_page_cost tuning to MCP server construction. Similarweb's HBase-to-DynamoDB migration is a useful case study in composite key design at 255TB scale.

---

## Databricks LTAP: one storage layer for OLTP and OLAP, built for agents

Databricks introduced LTAP (Lake Transactional/Analytical Processing) at its Data + AI Summit — the core idea being that AI agents need simultaneous access to live operational data and historical context, which strains any architecture that requires ETL between transactional and analytical stores. LTAP stores data once in open formats on cloud object storage, with Lakebase (their Postgres-based operational database) handling writes and Lakehouse//RT handling analytics at sub-100ms latency at 12,000 QPS. Git-style database branching via the Neon acquisition lets agents fork petabyte-scale databases in seconds for safe experimentation. Analysts are cautiously optimistic but note real-world latency under mixed load hasn't been demonstrated yet — HTAP made similar promises and mostly failed. [Read more](https://app.daily.dev/feed-by-ids?id=MivQDovba&id=EiWC37ccu&id=VJFeEQlmw&id=UQTG9RCz2&id=HBlZ5JW8h)

## PostgreSQL 19 Beta 1: SQL graph queries, REPACK CONCURRENTLY, 2x faster FK inserts

PostgreSQL 19 Beta 1 is out, with GA expected in September. The headline features are SQL/PGQ (Property Graph Queries) for running graph traversals over relational tables without a separate graph database, and REPACK CONCURRENTLY for online table rebuilds with no downtime. Performance improvements include up to 2x faster inserts when foreign key checks are involved, plus anti-join optimizations and incremental sorts. Logical replication now syncs sequences automatically and can be enabled without a server restart. JIT compilation is disabled by default. AWS has the beta available in the RDS Preview Environment now. [Read more](https://app.daily.dev/posts/2adopzBmZ)

## Similarweb migrates 255TB, 7 billion records from HBase to DynamoDB at 20x cost reduction

Similarweb's HBase setup was collapsing under its own weight — RegionServer instability, painful disk upgrades, and always-on peak-sized clusters burning money. The DynamoDB migration is worth reading for the data modeling decisions alone: composite partition+sort keys using site_country/date for range queries, a synthetic 'worldwide' country code to avoid fan-out, and a per-month table strategy for time-partitioned datasets. The 20x cost reduction over flat key designs came from getting the key structure right, not from DynamoDB being inherently cheap. For bulk backfills, they used DynamoDB's Import from S3 feature for up to 90% cost reduction. [Read more](https://app.daily.dev/posts/Fwr1GW4K9)

## POSETTE 2026: random_page_cost, MCP servers on Postgres, and timezone data corruption

POSETTE produced several talks worth your time. The random_page_cost deep dive is the most practically useful: raw SSD benchmarks show random-to-sequential I/O ratios of 27–140x, yet the default of 4.0 has held for 25 years because the cost model ignores cache pollution from sequential scans. Lowering to ~2.0 on SSDs is reasonable, but tune from whole-system monitoring, not raw I/O numbers. The MCP server talk covers defense-in-depth for Postgres: SQL parsing for read-only enforcement, database-level roles, and query timeouts as layered controls. Separately, British Columbia's move to permanent UTC-7 starting March 2026 is a real timestamptz corruption risk — a dual-column pattern storing wall-clock time, IANA timezone name, and a trigger-computed UTC timestamp is the recommended fix. [Read more](https://app.daily.dev/feed-by-ids?id=c2Wj76DXv&id=7SJS0XkGq&id=UiEbXsyyj&id=ejaEDxDgl&id=yNVn32mUG&id=PFIHO13UE)

---

## Also notable

- **pgstream v1.1.0 fixes bigint silent data corruption from wal2json float64 decoding:** pgstream v1.1.0 patches a critical correctness bug where wal2json decoded large bigint values as float64, silently corrupting data; the release also adds ltree, cube, and pgvector type support, JSON logging for Loki/Datadog, and HTTP health endpoints for Kubernetes probes — drop-in upgrade from v1.0.x. [Read more](https://app.daily.dev/posts/cEvLqJnnQ)
- **pg_kpart v1.0: enforce partition key predicates at the database level:** pg_kpart is a new Postgres extension that rejects queries missing a usable predicate on the partition key, turning a fragile developer convention into a hard database guarantee, with audit mode for gradual rollout and a dedicated SQLSTATE for application-level error handling. [Read more](https://app.daily.dev/posts/iiA1yDCZk)
- **Debezium 3.6.0.Beta2: MySQL GTID tags, unified CLI, CockroachDB per-table changefeeds:** Debezium 3.6.0.Beta2 resolves 51 issues including MySQL 8.3+ GTID tag support, a new unified CLI for managing CDC pipelines, Spanner UUID support, and a credential leakage fix in logs — plus CockroachDB gets per-table changefeed splitting for performance isolation. [Read more](https://app.daily.dev/posts/Jy8IRoHNV)
- **Coinbase postmortem: single-AZ cooling failure took down Raft quorum and Kafka, causing multi-hour trading halt:** A May 7, 2026 cooling failure in one AWS US-East-1 AZ knocked three of five Raft nodes offline in Coinbase's matching engine Cluster Placement Group, losing quorum and requiring emergency code changes and manual cluster reconstruction — a direct consequence of collocating nodes for low latency. [Read more](https://app.daily.dev/posts/fn8qhkDtj)
- **Aurora DSQL EXPLAIN: 81% DPU reduction by adding a covering index:** Amazon's Aurora DSQL EXPLAIN guide introduces a three-layer filter model (Index Condition, Storage Filter, Query Processor Filter) specific to its compute/storage-separated architecture, with a worked example showing 81% DPU reduction by adding an index with INCLUDE columns. [Read more](https://app.daily.dev/posts/n9NU4ezMA)

## Similar posts on daily.dev

- [From monolith to Lakebase to LTAP: rethinking the database from storage up](https://daily.dev/posts/from-monolith-to-lakebase-to-ltap-rethinking-the-database-from-storage-up-gsup5t8nr) · databricks · 6 upvotes · 0 comments
- [Databricks wants to merge the two databases every company runs](https://daily.dev/posts/databricks-wants-to-merge-the-two-databases-every-company-runs-mivqdovba) · The New Stack · 2 upvotes · 0 comments

---

Tags: [#backend](https://daily.dev/tags/backend), [#postgresql](https://daily.dev/tags/postgresql), [#databricks](https://daily.dev/tags/databricks), [#change-data-capture](https://daily.dev/tags/change-data-capture)

[View this post on daily.dev](https://daily.dev/posts/databricks-bets-on-ltap-postgresql-19-beta-ships-sql-graph-queries-oqdztqft7)

```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/databricks-bets-on-ltap-postgresql-19-beta-ships-sql-graph-queries-oqdztqft7","headline":"Databricks bets on LTAP, PostgreSQL 19 Beta ships SQL graph queries","text":"A curated roundup of database and data platform news. Databricks unveiled LTAP (Lake Transactional/Analytical Processing) at its Data + AI Summit, combining OLTP and OLAP on a single lakehouse storage layer with Postgres-based Lakebase and sub-100ms analytics. PostgreSQL 19 Beta 1 ships SQL Property Graph Queries for graph traversals over relational tables, REPACK CONCURRENTLY for zero-downtime table rebuilds, and up to 2x faster FK inserts. Similarweb's migration of 255TB and 7 billion records from HBase to DynamoDB achieved a 20x cost reduction through careful composite key design. POSETTE 2026 talks covered random_page_cost tuning for SSDs, MCP server security patterns for Postgres, and timezone data corruption risks. Notable releases include pgstream v1.1.0 fixing bigint silent corruption, pg_kpart v1.0 for partition key enforcement, Debezium 3.6.0.Beta2, and a Coinbase postmortem on a Raft quorum loss from a single-AZ cooling failure.","url":"https://daily.dev/posts/databricks-bets-on-ltap-postgresql-19-beta-ships-sql-graph-queries-oqdztqft7","datePublished":"2026-06-17T04:18:56.333Z","dateModified":"2026-06-17T04:19:25.068Z","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":"Databricks bets on LTAP, PostgreSQL 19 Beta ships SQL graph queries"}]}
```

