Posts - Page 24
All the articles I've posted, page 24.
- 13 MIN READ•Jun 22, 2026
Iceberg v4 Performance: Root Manifests and Calls
Apache Iceberg v4 discussion should focus on planning cost, metadata layout, and object storage round trips, not vague claims about faster tables.
root manifestsmetadata round tripsobject storage planning - 13 MIN READ•Jun 22, 2026
What Is LTAP in the Lakehouse?
Lakehouse transactional analytical processing is useful only when teams define freshness, isolation, and workload boundaries clearly.
lakehouse transactional analytical processingIceberg updatesanalytical freshness - 14 MIN READ•Jun 22, 2026
The Model Is Not the Moat
Enterprise AI advantage increasingly comes from governed context, semantic models, and operational data contracts, not only from model choice.
enterprise AI moatsemantic layergoverned lakehouse AI - 13 MIN READ•Jun 22, 2026
PyIceberg at Scale Without Apache Spark
Where PyIceberg works without Spark, such as serverless functions appending small batches, where it does not, and what to measure before scaling it up.
Spark-free IcebergPython Iceberg pipelinesREST catalog concurrency - 14 MIN READ•Jun 22, 2026
The Real-Time Lakehouse with Streaming and Iceberg
The real-time lakehouse is not one engine. It is a contract between streams, table commits, query paths, and freshness expectations.
streaming SQLIceberg storageevent-driven lakehouse - 13 MIN READ•Jun 22, 2026
REST Catalog V2 LoadTable and Client Capability
REST Catalog V2 LoadTable work matters because clients and catalogs need explicit contracts, not optimistic assumptions.
client capabilitiesIceberg REST catalogopen catalog contracts - 14 MIN READ•Jun 22, 2026
Rust vs C++ in Native Iceberg Scan Operators
The Rust versus C++ discussion is really about table-layer execution safety, interoperability, and performance envelopes.
Rust data systemsC++ query enginesvectorized scans - 14 MIN READ•Jun 22, 2026
Server-Side Commit Deconflicting in REST Catalogs
Why per-client retry loops break down with many writers, and how server-side commit deconflicting in Iceberg REST catalogs moves concurrency control there.
Iceberg commitsREST catalog concurrencyoptimistic concurrency control - 14 MIN READ•Jun 22, 2026
Snowflake Interoperable Lakehouse Lessons
Interoperable lakehouse announcements matter when they change production contracts, not just import and export narratives.
Iceberg interoperabilityopen tablesmulti-engine lakehouse - 15 MIN READ•Jun 22, 2026
Unity AI Gateway vs Apache Polaris Control Planes
Unity AI Gateway and Apache Polaris govern different parts of the analytics path. Compare a closed AI gateway with an open catalog control plane, and their fit.
governance control planeApache PolarisAI gateway - 14 MIN READ•Jun 8, 2026
Agentic Lakehouse Concurrency and Isolation
Agentic writes need isolation contracts, not just write permissions.
lakehouseAI agents - 14 MIN READ•Jun 8, 2026
Anatomy of an Agentic Lakehouse
An Agentic Lakehouse is storage, catalog governance, semantic context, and agents working as one operating model.
lakehouseAI agents