Building a Serverless ClickHouse
Comments on “Building a Serverless ClickHouse”
Claim-checked. Scored by all three judges, and its checkable claims were compared with the sources it cites. How far stories are checked
Jordi Villar describes ClickHamlet, a serverless ClickHouse service built on open source plain_rewritable S3 disks. A C++ router speaks the native protocol and sends queries to a single writer or a reader, both started on demand and stopped after five idle minutes. Villar reports that routing benchmarks rose from 451 to 2,713 queries a second, mostly through controller changes, though the figures are his own and come from one machine. Readers lag the writer by up to five seconds, and multiple writers and schema changes remain unsupported.
Why it ranked
No. 11 of 60 with a score of 0.66. The panel rated it highest on technical depth (0.95) and new information (0.93), lowest on audience breadth (0.50). Placed No. 11, No. 18 by score alone: the placement terms moved other stories past it.
- AI judgeJudge 2claude-haiku-5-5
- Detailed, well-evidenced engineering writeup on building serverless ClickHouse with compute/storage separation, routing, and cost measurements; a product launch with real technical depth, of interest to data and infra builders but narrow in audience.
- AI judgeJudge 3gpt-5.6-luna
- A detailed, well-supported engineering account of a serverless ClickHouse architecture, covering routing, storage semantics, Kubernetes coordination, performance bottlenecks, cost optimization, and operational trade-offs. It is highly actionable but still early and limited in scale.
Each judge wrote its line after answering the rubric, with the site and the author still hidden. Every answer behind this is in the table under Score.
Score
0.66total under rubric v2.3
- Significance0.54
- Reader interest0.82
- Audience breadth0.50
- Practical relevance0.70
- Substantiated0.87
- Technical depth0.95
- New information0.93
- Counts against: Promotional0.15
- Counts against: Overclaim0.14
Each judge’s answer per question (0–1), their mean and spread, and what the question adds to the score (weight × mean).
| Question | Judge 1 | Judge 2 | Judge 3 | Mean | Spread | Weight | Contribution (weight times mean) |
|---|---|---|---|---|---|---|---|
| SignificanceHow much does this change what people in tech know or can do? | 0.50 | 0.45 | 0.66 | 0.54 | 0.21 | +0.24 | +0.129 |
| Reader interestWould a well-informed senior engineer or tech founder want to read this today, and likely pass it on? | 0.77 | 0.80 | 0.90 | 0.82 | 0.13 | +0.26 | +0.214 |
| Audience breadthHow many people who build technology does it concern? | 0.49 | 0.50 | 0.50 | 0.50 | 0.01 | +0.13 | +0.065 |
| Practical relevanceHow useful is it to someone who builds software or tech products? | 0.78 | 0.55 | 0.78 | 0.70 | 0.23 | +0.22 | +0.155 |
| SubstantiatedHow well is it backed? Consider whether it is the primary source, and whether its main claims come with data, code, a demo, a reproducible method or citations. | 0.92 | 0.80 | 0.90 | 0.87 | 0.12 | +0.05 | +0.044 |
| Technical depthHow technically substantive is it? | 0.99 | 0.90 | 0.95 | 0.95 | 0.09 | +0.05 | +0.047 |
| New informationDoes it contain information absent from the stories covered in the last 7 days (list provided)? | 0.87 | 0.95 | 0.97 | 0.93 | 0.10 | +0.05 | +0.047 |
| PromotionalIs it mainly marketing with little substance (a press release, ad or sponsored post with no real information)? A launch post that carries real news is not promotional. | 0.06 | 0.20 | 0.18 | 0.15 | 0.14 | −0.19 | −0.028 |
| OverclaimDo the headline or conclusions claim more than the body supports? | 0.18 | 0.15 | 0.08 | 0.14 | 0.10 | −0.11 | −0.015 |
| Sum of contributions | +0.657 | ||||||
| Claim-check penalty | 0.000 | ||||||
| Total, as published | 0.657 | ||||||
Spread is the highest minus the lowest judge answer. marks a spread above 0.40: the judges disagree, and the story is marked contested.
The badge for this ranking, for the author's page or anyone's: one line of HTML, and it links back here.
<a href="https://rubric.hrih.si/editions/2026-11-16/11/"><img src="https://rubric.hrih.si/badge/2026-11-16/11.svg" alt="Ranked #11 on Rubric, 16 Nov 2026" width="245" height="28"></a>Claim check
Claim check: 1 matches source · 7 no source given. Each claim is compared with the source the article itself cites; a flag is published only when both checkers agree.
No source givenGeneralization
SharedMergeTree is a ClickHouse table engine that stores data in S3 and metadata in ClickHouse Keeper, and it is only available in ClickHouse Cloud.
The article gives no link or citation for it.
The article says “SharedMergeTree, a table engine that stores data in S3 and keeps its metadata in ClickHouse Keeper, which gives the compute and storage separation many users were asking for. It’s only…”
The source says No source to compare with.
No source givenGeneralization
In the open source version, zero-copy replication never became production ready and was later deprecated and removed.
The article gives no link or citation for it.
The article says “the closest thing was zero-copy replication, which never became production ready and was later deprecated and removed”
The source says No source to compare with.
Matches sourceVersion
The ClickHouse 26.8 release post introduces a JSON AST dialect.
The cited source says the same thing.
The article says “the 26.8 release post introduces a JSON AST dialect”
The source says “ClickHouse 26.8 also adds an experimental dialect, clickhouse_json, that lets you send a JSON AST instead of SQL”
Source cited by the article: clickhouse.com/blog/clickhouse-release-26-08
No source givenNumber
A query to an already running backend costs one controller call of about 0.2 ms.
The article gives no link or citation for it.
The article says “A query to a running backend costs one call to the controller, about 0.2 ms”
The source says No source to compare with.
No source givenBenchmark result
The benchmark of 16 clickhouse client sessions running 300 SELECT count() each went from 451 queries a second to 2713 queries a second, 6x the baseline.
The article gives no link or citation for it.
The article says “After a handful of changes, the same benchmark does 2713 queries a second, 6x the baseline”
The source says No source to compare with.
No source givenNumber
Controller CPU per query dropped from 2.34 ms to 0.30 ms.
The article gives no link or citation for it.
The article says “the controller went from 2.34 ms to 0.30 ms”
The source says No source to compare with.
No source givenCausal conclusion
Packing small parts into a single object with min_bytes_for_full_part_storage cut PUT-class requests by 69%.
The article gives no link or citation for it.
The article says “Packing small parts into a single object (min_bytes_for_full_part_storage, the same approach ClickHouse Cloud uses) cut that by 69%”
The source says No source to compare with.
No source givenNumber
A Graviton r7g.xlarge costs $0.25 an hour on demand and around $0.10 on spot.
The article gives no link or citation for it.
The article says “A Graviton r7g.xlarge (4 vCPU, 32 GiB) costs $0.25 an hour on demand and around $0.10 on spot”
The source says No source to compare with.
Also covered by
No other coverage was clustered with this story.
Elsewhere
Elsewhere: HN 4 pts, not on front page
Recorded for context only, checked 16 Nov 2026, 05:47 UTC. It played no part in the ranking.