Data Analytics News: BigQuery Soars With Borderless Lakehouse

The data analytics news cycle heads into November 2026 with a packed week: Google just rewired how BigQuery talks to rival data platforms, DuckDB shipped the most ambitious version in its history, Meta bought an AI analytics startup founded by one of Silicon Valley’s most recognisable product leaders, and two significant M&A deals continued reshaping the BI consolidation landscape. Here is everything data engineers, analysts, and scientists need to know going into the new month.

Google’s BigQuery Soars With “Borderless Lakehouse” — Now Queries Databricks and Snowflake Natively

On October 27, 2026, Google Cloud announced that BigQuery can now federate queries directly across Databricks Unity Catalog, Snowflake Horizon Catalog, and AWS Glue Data Catalog without copying or moving a single byte of data. The feature, branded Borderless Lakehouse, uses the Apache Iceberg REST catalog standard to expose external tables as first-class BigQuery objects referenced with four-part paths. Query compute runs entirely on Google Cloud infrastructure, so no Databricks cluster, Snowflake virtual warehouse, or AWS instance is touched at query time. Google also announced a Conversational Analytics API that lets developers build natural-language agents capable of querying across the full federated set, and a Gemini Enterprise integration that brings the same cross-cloud querying capability to business users who never write SQL. For data teams already on BigQuery who need to join against datasets locked in a rival platform — the kind of cross-source filtering covered in our guide to filtering data with the SQL WHERE clause — the feature removes the last excuse for duplicating data across clouds. The catalog federation capability is currently in preview. (Google Cloud Blog, IT Brief)

DuckDB 2.0.0 LTS “Cinnamon Teal” Ships — and Becomes a Full Server

The biggest open-source data analytics news of the autumn: on October 21, 2026, DuckDB 2.0.0 LTS, codenamed “Cinnamon Teal,” reached stable release — and it is a fundamentally different product from v1.x. The headline addition is the Quack remote protocol, which turns DuckDB into a client-server system: one DuckDB instance can act as a server over HTTP while others connect as clients, a capability that has no parallel in any previous release. Version 2.0 also ships the VARIANT type for storing semi-structured JSON natively, NEAREST joins for top-k vector similarity search (important for AI embedding pipelines), asynchronous I/O for Parquet and CSV files from object stores, and a new default SQL parser. The release is designated LTS with support guaranteed through at least October 2027. For SQL practitioners who work with advanced filtering patterns like IN, BETWEEN, and LIKE, DuckDB 2.0’s improved performance on large analytical workloads makes it an increasingly credible choice for local and pipeline analytics. (DuckDB Official Blog)

Amazon Acquires DuckLabs — DuckDB Stays Open Source

Contextualising that release: on August 26, 2026, Amazon announced a definitive agreement to acquire DuckLabs, the Amsterdam-based company behind DuckDB. The DuckLabs team — including co-founders Hannes Mühleisen and Mark Raasveldt — will join AWS and continue leading the project’s technical direction from Amsterdam. Crucially, DuckDB itself remains free and open source under the independent DuckDB Foundation’s MIT licence, and Amazon explicitly said it is not acquiring the DuckDB project. The two organisations had been working together since 2024 on DuckDB support for Amazon S3 Tables and SageMaker Lakehouse. An October InfoQ report confirmed the first visible integration: Amazon Aurora PostgreSQL is now the first AWS service to ship a user-facing feature powered by DuckDB. In data analytics news for cloud infrastructure, the deal adds Amazon to the growing list of hyperscalers that have acquired or deeply embedded open-source analytics technology teams. (About Amazon, GeekWire)

Meta Acquires Sundial — AI Analytics Built by Former Meta Leaders

On October 23, 2026, Axios confirmed that Meta has acquired Sundial, an AI-powered data analytics platform co-founded by Julie Zhuo — Facebook’s former VP of Product Design — and Chandra Narayanan. Sundial is being folded into Meta’s newly formed DATA unit (Data Analytics Tooling & Agents), headed by Chief Data Officer Alex Schultz, appointed in July. Narayanan will lead Meta’s internal data analytics and agents work. Sundial had raised $23 million including a $16 million Series A in July 2025, with clients that included OpenAI. Those external customers will stop being served when the deal closes. A significant data analytics news story: the acquisition signals how seriously the largest platforms are investing in proprietary AI-native analytics infrastructure — and raises a pointed question for any data team that has standardised on a third-party analytics vendor: what happens when a hyperscaler decides to build the same capability in-house? (Axios)

Progress Software to Close $400M Domo Acquisition by November 30

One deal to watch as November begins: Progress Software’s $400 million acquisition of Domo’s AI and data platform business is expected to close by November 30, 2026 — the end of Progress’s fiscal year. Progress announced the all-cash asset purchase in July and is absorbing Domo’s operating business, technology, customer contracts, and intellectual property, bringing roughly 2,400 business customers and a network of cloud data warehouse partners. Domo will retain its public listing (under a new name and ticker after close) and keep its net operating loss carryforwards. Progress has committed to continuing to support and develop the platform. For organisations running Domo for BI, the message is continuity — but data teams should monitor the roadmap closely in the months following the close. (Techzine)

This Week in Data Analytics News: What It Means for Your Stack

That is today’s data analytics news wrap. The BigQuery Borderless Lakehouse story is the technical headline of the week: zero-copy cross-cloud federation via Apache Iceberg is exactly what multi-cloud data teams have been waiting for, and putting it into preview at this scale sends a clear signal that open table formats are now the connective tissue of the modern data stack. Meanwhile, the DuckDB 2.0 release and the Amazon/DuckLabs acquisition together represent the biggest shift in the in-process analytics market since DuckDB first appeared — and with Aurora PostgreSQL now running DuckDB under the hood, the database is quietly embedding itself into cloud infrastructure in ways that go well beyond a developer tool.