Tootfinder

What happens when you bring columnar storage and vectorized execution to search? Saurabh Singh and Ankit Jain (OpenSearch) are at #bbuzz26 to walk through what that architectural shift looks like in practice.
Learn more: https://2026.berlinbuzzwords.de/session/from-inverted-index-to-columnar-vectorized-execution-search/

Aggregation queries don't have to slow everything down. Sandesh Kumar and Shailesh Kumar Singh are at #bbuzz26 to show how Star-Tree indexing in OpenSearch enables constant-time aggregations regardless of dataset size.
Learn more: https://2026.berlinbuzzwords.de/session/constant-time-aggregations-with-star-tree-in-opensearch/
Get your ticket: https://2026.berlinbuzzwords.de/tickets/

So, jetzt mal das Mediawiki mit OpenSearch aufgesetzt.
Schön ist dann, dass Suchvorschläge Namensräume berücksichtigen _und_ case insensitive sind.
Bei den Ergebnissen selber, gefällt mir die Vanilla-Suche spontan noch besser. Aber vielleicht kann man OpenSearch bzw. die CirrusSearch -Extension noch optimieren 🤔 (am besten einfach :))
Naja, genug für heute :)
#mediawiki

Kris Freedain and Carlos Juzarte Rolo will be taking the stage at #bbuzz26 to talk through what transition has looked like for the OpenSearch Software Foundation, what they have learned, and where the project is headed next.
Learn more: https://2026.berlinbuzzwords.de/session/opensearch-software-foundation-1-year-of-open-governance/

Improving search relevance without complex pipelines, recommendation engines or ML models? This session by Hajer Bouafif at #bbuzz26 shows how lightweight agents collaborate through OpenSearch's agentic memory to enrich queries in real time.
Learn more: https://2026.berlinbuzzwords.de/session/personalize-search-results-with-opensearch-agentic-memory/

Berlin Buzzwords 2026 has come to an end, and it wouldn't have been possible without the support of our amazing partners. We're incredibly grateful to each of them for helping make it happen.
A big thank you to our Gold Partners OpenSearch Project, Qdrant, and Vespa.ai, our Recording Partner Bloomberg, our Silver Partners Hornet.dev, Linkup, NetApp Instaclustr, Thomson Reuters Lab, and Search Guard, our Bronze Partner OpenSource Connections

Mediawiki und SearchSuggestions sind mir irgendwie suspekt
1. Vanilla-Variante
** Ist Case-sensitive (nach Buchstabe 1). "open lib" schlägt nicht "Open Library" vor
2. TitleKey-Extension
** Nicht Case sensitive, aber dafür keine Namensräume. "Ausw" findet nicht "OUS:Ausweis"
3. CirrusSearch (OpenSearch)
** Case-insensitive Namensräume. Dafür werden Seitentitel erstaunlich gering bewertet
Subpages vorzuschlagen s…

Looking for the latest in open-source search, analytics, observability, and AI-powered search experiences? Make sure to swing by the OpenSearch booth! 🔍✨ #bbuzz26

Open Search has joined Berlin Buzzwords again, this time as a Gold Partner!
Learn more about Open Search here: https://opensearch.org/
Join us online and in-person at Kulturbrauerei, Berlin.
https://2026.berlinbuzzwords.de/