How We Log LLM Requests at Sub-50ms Latency Using ClickHouse
We were logging 50,000 LLM requests per day to PostgreSQL. Query latency was fine. At 400,000 requests, cost aggregation queries started taking 3 seco…
Latest Programming news from Tech News
We were logging 50,000 LLM requests per day to PostgreSQL. Query latency was fine. At 400,000 requests, cost aggregation queries started taking 3 seco…
TL;DR ClickHouse has full native JSON support, and has since v25.3. The JSON type stores each path as a separate columnar subcolumn with native type p…
When working with location data, one problem shows up almost immediately: “How do I calculate the distance between two coordinates stored in my databa…
1. Вступление: какую проблему решает ClickHouse ClickHouse нужен там, где обычная OLTP-база начинает мешать аналитике: большие объемы событий, логов, …
Part 4 of a series on building a metrics pipeline into ClickHouse Read Part 3: Understanding Vector Transforms When Things Still Don’t Work At this po…
Part 3 of a series on building a metrics pipeline into ClickHouse Read Part 2: Understanding Vector Pipelines Where Things Got Real By this point, the…
Открываю ноутбук. Захожу в метрики своего интернет-магазина. 500 тысяч посетителей в день. Каждый клик — событие. Каждый просмотр товара — строка в ло…