Grilled Cheese

ExploreLog inSign up
Terms of UsePrivacy PolicyCommunity StandardsHelpGet the app

Grilled Cheese is a product of Village Compute

Version devBuilt at: 2026-10-11 02:37:10 EDT

Explore

PostsPeople
LatestRanked
@duckdb.orgSep 22, 2026, 4:00 PM

Attending the Rows & Columns Summit in San Francisco today?

*THE* @hannes.muehleisen.org, co-creator of #DuckDB, will give a talk on “Nobody Knows What OLTP Is: DuckDB Moves to the Middle” at 1:30 p.m. (today, Sept 22nd)

We hope to see you there! 🦆 🦆

Sassy abstract and more info here:

@hofundsoftware.bsky.socialSep 21, 2026, 4:16 PM

Stop writing throwaway scripts to inspect Parquet files. Rowist is a native macOS columnar studio: embedded DuckDB, zero-copy mmap for multi-GB files, and 100% offline. Zero telemetry. Perpetual licenses. Private beta waitlist is open: https://rowist.app #macOS #DataEngineering #DuckDB

@hrbrmstr.mastodon.social.ap.brid.gySep 18, 2026, 4:58 PM

The #webR and #DuckDB OPFS stuff I found today was too good not to poke at, especially since they have absolutely nothing to with "AI", "LLMs" or "agents". Plus, I rly needed to get back some @webawesome muscle memory.

This is a git repo with two thin web […]

[Original post on mastodon.social]

Landing page headed "OPFS from the browser: DuckDB and R", explaining that the Origin Private File System is a per-origin sandboxed filesystem with random-access reads and writes. Two cards follow. The DuckDB-Wasm card opens a database at opfs://analytics.duckdb, inserts rows, checkpoints, caches an aggregate as Parquet, and downloads the .duckdb file, with no SharedArrayBuffer needed. The webR card mounts /opfs as an Emscripten filesystem and uses write.csv and read.csv against it, and needs cross-origin isolation. Each card has a button to open its demo.The DuckDB-Wasm demo page after a page reload. Buttons run the steps: open database, insert a row, count rows, load remote orders, write and read a Parquet cache, download the .duckdb file, clear OPFS. The log below shows the reopened database already holding 13 rows in the observations table and 15,000 rows in orders, so both survived the reload. Two more inserts follow, each checkpointed. The orders table loads from OPFS in 96 milliseconds with no HTTP fetch, the Parquet cache is written to opfs://cache/monthly_totals.parquet and read back as five priority totals, and the database file downloads at 1,060,864 bytes.
@junaidwaseem474.bsky.socialSep 18, 2026, 2:30 PM

PostGIS isn't the only way to query spatial data. DuckDB offers a lightning-fast, serverless alternative for running heavy geospatial SQL queries directly on your local machine.

https://rsandgis.me/articles/serverless-spatial-database-duckdb.html

#DuckDB #Database #Geospatial

@hrbrmstr.mastodon.social.ap.brid.gySep 18, 2026, 2:02 PM

whoa.

#DuckDB works with OPFS (Origin Private File System) — https://duckdb.org/2026/09/18/opfs-wasm

and, so can #WebR! — https://github.com/seanbirchall/webr-opfs/blob/main/INTEGRATION.md

i.e., load a remote file once and persist to sandboxed private storage accessible by the browser

So […]

@villagesql.comSep 17, 2026, 5:15 PM

Very excited to @duckdb.org running in #MySQL as an extension! Query your parquet data in S3 and join with your operational data.

villagesql.com/blog/duckdb/

#DuckDB #MySQL #S3 #Parquet

@jelte.mastodon.nl.ap.brid.gySep 15, 2026, 8:19 AM

I finally looked into #duckdb for some of my analytics, and wow. HOW.

@hrbrmstr.mastodon.social.ap.brid.gySep 14, 2026, 8:49 PM

Speaking of AI cgit crawlers and whack-a-mole, ran some #duckdb over the logs and pplx, claude, and others took some turns too. Added them to the blocker config for the git dirs.

Dark-themed table of daily web request counts by user agent, Aug 22 through Sep 14. Columns cover GPT, ChatGPT, OAI-Srch, Claude, ClaudeUser, ByteSpider, Perplexity, Amazon, Petal, Google, Baidu, Sogou, Yandex, Dotbot, Meta, Cohere, You, MJ12, DataForSEO, Serpstat, other bots, and human traffic. Nearly every cell is a dash or a single digit, with a few large bursts: GPT hits 119,301 on Sep 5 and 62,282 on Sep 6; Claude hits 116,889 on Sep 7, 132,491 on Sep 8, and 101,612 on Sep 9; Perplexity hits 10,523 on Sep 10, 28,702 on Sep 11, and 5,051 on Sep 12. Human requests stay between roughly 150 and 1,900 per day the entire time, several orders of magnitude below the crawler spikes.
@sergiocuellar.bsky.socialSep 14, 2026, 4:39 AM

DuckDB, a powerful library-based analytics engine, has joined AWS to revolutionize structured data processing. 🦆⚙️ #DuckDB #AWS #DataAnalytics

@darryl-ruggles.cloudSep 13, 2026, 10:22 PM

https://lckhd.eu/PnG7TS

#DynamoDB #DuckDB #ApacheIceberg #Serverless

There are many different options for databases on AWS and it's great to be able to pick the one that best matches your use case. DynamoDB (DDB) is an example of a highly performant non-relational database which can provide

@javapedianet.bsky.socialSep 13, 2026, 7:19 PM

How do you troubleshoot schema evolution issues when querying a DuckLake table over time?

https://www.javapedia.net/duckdb-interview-questions/11097/how-do-you-troubleshoot-schema-evolution-issues-when-querying-a-ducklake-table-o

#javapedia #interviewquestions #programming #shorts #duckdb

@javapedianet.bsky.socialSep 13, 2026, 7:18 PM

What is the DuckDB Quack protocol, and how does it change DuckDB's deployment model?

https://www.javapedia.net/duckdb-interview-questions/11096/what-is-the-duckdb-quack-protocol-and-how-does-it-change-duckdbs-deployment-mode

#javapedia #interviewquestions #programming #shorts #duckdb

@javapedianet.bsky.socialSep 13, 2026, 7:18 PM

Why should you avoid treating DuckDB as a high-concurrency, multi-writer OLTP database?

https://www.javapedia.net/duckdb-interview-questions/11095/why-should-you-avoid-treating-duckdb-as-a-high-concurrency-multi-writer-oltp-dat

#javapedia #interviewquestions #programming #shorts #duckdb

@javapedianet.bsky.socialSep 13, 2026, 7:17 PM

How does DuckDB's cost-based optimizer decide on a query plan?

https://www.javapedia.net/duckdb-interview-questions/11094/how-does-duckdbs-cost-based-optimizer-decide-on-a-query-plan

#javapedia #interviewquestions #programming #shorts #duckdb

@javapedianet.bsky.socialSep 13, 2026, 7:17 PM

How do you troubleshoot memory issues when DuckDB processes a dataset larger than available RAM?

https://www.javapedia.net/duckdb-interview-questions/11092/how-do-you-troubleshoot-memory-issues-when-duckdb-processes-a-dataset-larger-tha

#javapedia #interviewquestions #programming #shorts #duckdb

@javapedianet.bsky.socialSep 13, 2026, 7:16 PM

Explain the lifecycle of a write operation in a DuckLake-backed table?

https://www.javapedia.net/duckdb-interview-questions/11091/explain-the-lifecycle-of-a-write-operation-in-a-ducklake-backed-table

#javapedia #interviewquestions #programming #shorts #duckdb

@javapedianet.bsky.socialSep 13, 2026, 7:16 PM

What is the difference between MotherDuck's hybrid execution and running DuckDB fully locally?

https://www.javapedia.net/duckdb-interview-questions/11090/what-is-the-difference-between-motherducks-hybrid-execution-and-running-duckdb-f

#javapedia #interviewquestions #programming #shorts #duckdb

@javapedianet.bsky.socialSep 13, 2026, 7:15 PM

When should you use DuckDB instead of a distributed system like Spark?

https://www.javapedia.net/duckdb-interview-questions/11078/when-should-you-use-duckdb-instead-of-a-distributed-system-like-spark

#javapedia #interviewquestions #programming #shorts #duckdb

@qiita-trending.bot.chrs.toSep 12, 2026, 8:20 PM

dbt ＋ DuckDB で学ぶ Spotify 式 Context Layer - 正確なKPI集計と自由なデータ探索を両立する Text-to-SQL 設計論

#duckdb #dbt #dwh #SQL #データエンジニアリング

Load more
The webR demo page. Buttons boot webR and mount /opfs, append a run, read the CSV, list /opfs, and clear OPFS. The log shows webR booting, the 12,355-byte filesystem plugin installing, and /opfs mounting. Five appends grow runs.csv from 21 to 25 rows, and the R data frame prints its last five rows with timestamp, host, and latency columns. The final directory listing shows analytics.duckdb at 1,060,864 bytes, its write-ahead log, the cache directory, and runs.csv at 1,123 bytes, so the R demo and the DuckDB demo share one origin's OPFS.