HNHacker News
TopNewBestAskShowJobs

dkgs

61 karma · joined July 21, 2025

submissionscomments

In Defense of Curiosity

davidbau.com·2 pts·dkgs·
0

Show HN: Pivot – fast analytics on Iceberg, 2x+ faster than ClickHouse / DuckDB

pivotlake.io·14 pts·dkgs·
1

Does ClickHouse need its own storage format?

pivotlake.io·2 pts·dkgs·
0

Boring Iceberg Catalog

github.com·2 pts·dkgs·
0

Why vector databases are a scam

simon-frey.com·2 pts·dkgs·
0

Pivot – Real-time analytics on Iceberg and Delta, faster than ClickHouse

github.com·1 pts·dkgs·
0

Spanner Omni: A distributed, multi-model database that you can deploy anywhere

cloud.google.com·4 pts·dkgs·
0

WarpSpeed rewrites Pandas to run 38.4x faster on PDS-H

doubleai.com·1 pts·dkgs·
0

Agentic Coding at ClickHouse

clickhouse.com·1 pts·dkgs·
1

Are open-table-formats and lakehouses the future of observability?

clickhouse.com·16 pts·dkgs·
1

I love UUID, I hate UUID

blog.epsiolabs.com·5 pts·dkgs·
4

Stream Processing Without Duplications

blog.epsiolabs.com·2 pts·dkgs·
0

Real-Time Mode in Apache Spark Structured Streaming

databricks.com·1 pts·dkgs·
0

The Bidirectional Stream Processor: Why Pull Beats Push for Crash Recovery

blog.epsiolabs.com·1 pts·dkgs·
0

How we made our streaming joins 50% faster

blog.epsiolabs.com·2 pts·dkgs·
0

The equality delete problem in Apache Iceberg

blog.dataengineerthings.org·65 pts·dkgs·
45

Stream processing doesn't need to suck

blog.epsiolabs.com·2 pts·dkgs·
0

Not duplicating messages: a surprisingly hard problem

blog.epsiolabs.com·4 pts·dkgs·
0

Amazon to buy AI company Bee that makes wearable listening device

cnbc.com·2 pts·dkgs·
1