Article URL: https://arnoldgalovics.com/microservices-in-production/ Comments URL: https://news.ycombinator.com/item?id=29576352 Points: 44
Comments: 14
Utworzony
4y
|
16 gru 2021, 10:20:03
Zaloguj się, aby dodać komentarz
Inne posty w tej grupie
Article URL: https://davidgomes.com/async-queue-interview-ai/

Chrome now includes a native on-device LLM (Gemini Nano) starting in version 138. I've been building with it since it was in origin trials, it's powerful but the official Prompt API is still a bit

I managed to reverse engineer the encryption (refered to as “Obfuscation” in the framework) responsible for managing the safety filters of Apple Intelligence models. I have extracted them into a r

Article URL: https://github.com/MrLesk/Backlog.md
Comments URL: https://news.ycomb

Article URL: https://www.dwarkesh.com/p/timelines-june-2025
4y | 16 gru 2021, 12:35:46
Can be said, that each time start with monolith, because it easier. If you grow, you can set up required microservices, which is highly optimized for current task.
Problem with microservices, when you are starting is communication, separation and you dont clearly see, which function should that microservice really do. Because usually each service must have access to database layer, then you dont know if databases should be separated or not.
Problem with monolith is basically harder horizontal scaling, but you can, in these days, scale vertically, server configurations are pretty powerful these days. Then if your website will work, you can start thinking of microservices.
So starts small, if project grows, just try to separate services.