✳flâneur — a map of the web's best reading

Ankur Goyal on X: "Quick rant about RDBMS, multi-tenancy, and AI. Data warehouses are great if you control the shape of your data and hand-code schema definitions. And they are basically useless for multi-tenant products (like @datadoghq, @splunk, @braintrustdata, etc.) where you have to ingest…" / X

twitter.com · 316 words · saved by 1 readers

To view keyboard shortcuts, press question mark View keyboard shortcuts Messages Messages 17 Recipe Chat (Let Us Cook) · 19h ⃤ rissa.exe.𑁍 reacted to @samwisefornow’s photo with cindy @cindywu · Apr 3 You sent a link James Hill-Khurana @jtvhk · Apr 2 hi! 17 SF ball · Apr 2 ZEKE: Running a game today at Potrero @ 7:30 if anyone wants to join tazik @taziksh · Mar 30 https://alexkehayias.com Antonio Yong @AntonioYongSam · Mar 29 You shared a post Charles Liu @CharlesLiu9 · Mar 29 You shared a post Brian Chau @psychosort · Mar 28 Curious to hear your take on AI solving the scalable search problem 3 Brian Chau, tazik and 1 more · Mar 25 tazik: Yeah can we do in 5min? Neel Baronia is "Design as Art" @nbaronia1 · Mar 25 You reacted with : curiosity is basically adding creative entropy to your life, and i find images to be a faster way to get creative entropy than any other medium (video, audio, text) brian-machado-finetuned-7b (e/snack) @sincethestudy · Mar 24 Bet Andrew McLuhan @amicusada

Ankur Goyal @ankrgyl Quick rant about RDBMS, multi-tenancy, and AI. Data warehouses are great if you control the shape of your data and hand-code schema definitions. And they are basically useless for multi-tenant products (like @ datadoghq , @ splunk , @braintrustdata, etc.) where you have to ingest customer data. The recommended fixes -- "reshape your data", "add casts", "fix the schema", "create more tables", etc. -- simply do not work with multi-tenant data. In fact, companies that scale, like @ datadoghq end up building tech like datadoghq.com/blog/engineeri… , which overlap significantly

Explore this link on the map →

related reading