Arc steals the show as a high-throughput time-series warehouse, pairing embedded analytics with SQL queries. Arc ingests via a binary MessagePack API, stores data as Parquet with hourly partitions, and queries through DuckDB. Write throughput runs around 1.88 million records per second, with benchmark notes like ClickBench on AWS c6a.4xlarge: 35.18 s cold, ~0.81 s hot for 43 queries (all succeed). [1]
Embedded analytics: Arc’s stack - Ingest via binary MessagePack for a fast path. [1] - Store as Parquet with hourly partitions. [1] - Query through DuckDB’s SQL engine. [1]
ClickGems: Free Analytics for RubyGems — analytics powered by ClickHouse for the RubyGems ecosystem. [2]
Django audit trails with ClickHouse — structured JSON logs are captured and shipped to ClickHouse, with Grafana making them queryable and visual. [3]
Bottom line: Arc shows how embedded analytics with DuckDB can handle high-throughput time-series, while ClickHouse shines for scalable analytics like RubyGems and Django audit logs. The right pick depends on workload, formats, and operational goals.
References
Show HN: Arc – high-throughput time-series warehouse with DuckDB analytics
Show HN: Arc ingests time-series data via MessagePack, stores as Parquet, queries with DuckDB; benchmarks and comparisons discussed.
View sourceClickGems: Free Analytics for RubyGems
Announcement of ClickGems: Free analytics service for RubyGems powered by ClickHouse.
View sourceShow HN: Rethinking audit trails in Django (structured and database-free)
Replaces DB writes with structured JSON audit logs; ships to ClickHouse; Grafana queries; asks how others handle audit logs.
View source