Andrey Abramov

CTO & Co-founder @ SereneDB

[Andrey Abramov @mrgnusi]

Glad you like it! You might also like Searchbench, a ClickBench-style search benchmark: 92 queries, 100M/1B/100B OTEL logs. Quickwit included! We'd love for you to tune configs & open PRs if needed.
We want to develop the same great spirit the Tantivy team built around their game!

Dashboard:
https://t.co/2fkhgwLMdM

Repo:
https://t.co/KYs6lSru7l

[François Massot 🦀 @FrancoisMassot]

Just discovered @serenedata did a cool version of the `search benchmark, the game` created a while ago by @fulmicoton

Their search engine (IResearch) shows impressive performance

https://t.co/Yl08jP3s83

[Andrey Abramov @mrgnusi]

"Just put a text index on @ClickHouseDB ."

So we measured. 92 queries over 10B OpenTelemetry logs on one box, finished inside 60s:

SereneDB 92
ClickHouse sorted 60
ClickHouse 48

https://t.co/2RevneFJfD

[Andrey Abramov @mrgnusi]

"Why build another search engine at all?"

So we measured. 92 queries, 1B logs, median latency:

@serenedata 32.5 ms
@elastic 98.4 ms
@OpenSearchProj 195.2 ms
@cratedb 543.5 ms

https://t.co/1cqaCqsxqV

[Andrey Abramov @mrgnusi]

"Why not just install a Postgres extension?"

So we measured. 92 queries, 1B logs, median latency:

@serenedata 35.5 ms
@paradedb 413 ms
@TimescaleDB > 60 s
@PostgreSQL > 60 s

https://t.co/DVMZpPBVM2

[Andrey Abramov @mrgnusi]

Excited to speak at #WeAreDevelopers 2026 in Berlin on how Search-OLAP helps AI agents access the best context!

Fri July 10 | 10:20 | Stage 2 https://t.co/51sOLeiRrZ

[Andrey Abramov @mrgnusi]

That's how modern search and analytics should look like. Don't move your data around, let your agents access and analyze it!

Check out GitHub-like code search over 11.5M solutions with zero ETL built on @serenedata.

https://t.co/DLamAMqxKj
https://t.co/eEdxrTQU3z

[Andrey Abramov @mrgnusi]

@jpountz Thanks for mentioning us, the implementation was highly inspired by your work.

[Adrien Grand @jpountz]

I was curious how SereneDB manages to perform well, on long queries in particular. Answer: it implemented the same block-at-a-time vectorized MAXSCORE algorithm as Lucene: https://t.co/LZRj6gVxYj! To my knowledge, this is the third implementation of this algorithm, very exciting.

[Paul Masurel 🦀 @fulmicoton]

Here comes a new challenger.
iresearch an engine implemented by SereneDB enters the contest with a bang!
https://t.co/AVE24oP1th