¿El tamaño importa?
Al inicio de mi carrera todo lo resolvía con un VPS. Unos años después estaba convencido de que cualquier proyecto serio necesitaba microservicios, multi-tenancy y un orquestador. Hoy, con más cicatrices encima, volví al VPS. Y no, no es una regresión: es la curva completa. El meme de la portada resume esa curva mejor que cualquier libro de arquitectura: he sido los tres personajes. Y como…
The journey of choosing the right system size is depicted through three stages. In the first stage, the author built an IoT measurement system using a single VPS, storing, processing, and displaying data all on one machine. This was despite the system's need for constant incoming data, potential traffic growth, and the risk of losing data. This stage represents an uninformed decision made only to ensure functionality, not through careful consideration of the best tool for the job.
The second stage introduces the author to web-scale concepts like microservices, multi-tenancy, containers, and orchestration. Inspired by these practices, the author designed a multi-tenant ERP system with microservices and orchestrated it using ECS. The result was a more complex architecture that, while impressive in its design, ultimately served only a single active company with predictable loads and gradual growth.
In the third and final stage, the author revisits the IoT project with a more practical approach. They chose a VPS, this time containerizing the application behind an nginx reverse proxy. This decision proved to be a more balanced choice, allowing for future scalability without the overcomplication of a premature microservices architecture.
The key takeaway from this story is that the right system size depends on the specific needs and growth trajectory of the project, and it's crucial to strike a balance between simplicity and future scalability.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
