Yannick Welsch

@cwi_da graduate & @duckdblabs CTO @mraasveldt gave a cool @duckdb lecture in @andy_pavlo's Advanced DB course: youtube.com/watch?v=bZOvAK… Details relevant @motherduck e.g.,: pluggable catalog+storage, WASM + (min44) pausable pipelines (for hybrid execution between cloud & local)

I've experimented with a pipe syntax in #DuckDB SQL to simplify data exploration. Give it a try: github.com/ywelsch/duckdb…

@max_sixty Happy to chat, @max_sixty. Will reach out via DM.

@ywelsch let me know if you want to chat re github.com/ywelsch/duckdb…! We're huge fans of this and would like to be helpful if we can...

Run #PRQL queries directly within #DuckDB using this new DuckDB extension: github.com/ywelsch/duckdb…

Elasticsearch and Lucene often take inspiration from academic work. This post shares strategies for implementing research papers, drawing on our experiences: elastic.co/blog/implement…

@bluxte @pmanvi @jpountz maybe hits().docs()?

@pmanvi @jpountz Indeed. This is because the Elasticsearch search response has two nested "hits" properties. Would you have a suggestion on how to rename one of these in the client (not in the API, obviously)? #search-search-api-example" target="_blank" rel="nofollow noopener">elastic.co/guide/en/elast…

🎉 The new Elasticsearch Java Client I've been working on for the last 6 months is now public. Still experimental but already usable! The API code is generated from a formal spec of the API, which is no small feat considering its size and complexity! 🥳 github.com/elastic/elasti…

Add icing to your data cake! Search petabytes of data stored on object stores like S3 within minutes using the frozen tier, now generally available. Learn more → go.es.io/3ulhTvj

The new frozen data tier decouples compute from storage, making it easier and cheaper to manage data at scale. Learn how search performance on the new frozen tier stacks up against other data tiers: go.es.io/3enoAYY

You never outgrow good data. Make object stores like Amazon S3 fully searchable with the new frozen tier in Elasticsearch 7.12 → go.es.io/397jnBu

How about querying a petabyte-scale data set on S3 using a single #Elasticsearch node? The upcoming experimental release of searchable snapshots for the frozen tier in #Elasticsearch 7.12 has you covered → github.com/elastic/elasti… + go.es.io/3eVrzGF

Coding meme based upon @Nasser_Junior comic

Since version 7.0, Elasticsearch successively moved points, stored fields and terms off-heap. Memory usage for segments will be 7x smaller with Elasticsearch 7.7 than it was in 7.0 with the geonames dataset. #tracks/geonames/nightly/default/90d" target="_blank" rel="nofollow noopener">elasticsearch-benchmarks.elastic.co/index.html#tra…

After 20 years, Apache #Lucene is still an important #opensource project with a thriving community. Take a look at some of the highlights of 2019 – including new committers, new PMC members, new features, and new improvements → go.es.io/2QaPJ5y

#tlaplus conf 2019: "Using TLA+ for fun and profit in the development of Elasticsearch" by @ywelsch youtube.com/watch?v=qYDcbc… conf.tlapl.us/program/yannic… /cc @strangeloop_stl

Yannick gives an overview of how #tlaplus helps an #elasticsearch cluster reach consensus.

#tlaplus conf 2019 preview: "Using TLA+ for fun and profit in the development of Elasticsearch" by @ywelsch conf.tlapl.us/program/yannic… #pluscal #Elasticsearch /cc @strangeloop_stl

Goodbye Zen Discovery! Meet the new cluster coordination subsystem in Elasticsearch 7.0: simpler configuration, faster master elections, effortless scaling, improved resilience elastic.co/blog/a-new-era…

Super nice presentation by @bleskes, @ywelsch, @jasontedor and @DaveCTurner at #Elasticon. Highly recommended to watch it when it will be online. (And great slides BTW)

And now @ywelsch and @DaveCTurner about reliable by design #Elasticon