northdan.
Vezi pagina în română

IT Glossary

What is a Microservices Architecture?

An architecture that splits an application into small, independent services — each with its own job, team and release cycle.

There are two ways to build a large application: as a single building — the monolith: all the code together, deployed together, scaled together — or as a neighborhood: microservices — small independent pieces (an orders service, a billing service, a notifications service), each with its own database, talking over APIs, shippable and scalable separately. The pattern was popularized by giants like Amazon and Netflix because it solves the problems of enormous scale: hundreds of developers no longer treading on each other's code, the busiest component scaled on its own, an isolated failure that doesn't topple everything, freedom to use different technologies per service. What the conference talks mention less: every one of those benefits is paid for in operational complexity — dozens of services mean orchestration (Kubernetes and its kin), distributed monitoring, transactions that span services (hard) and debugging that spans services (harder) — a fixed tax carried regardless of your size. Hence the honest advice for a smaller company, against the marketing current: a well-structured monolith (modular, with clean internal boundaries) is the rational choice for most business applications — cheaper to build, operate and understand; microservices become the answer when the pain of scale actually arrives — many teams, components with radically different scaling needs — and a modular monolith then splits elegantly along the boundaries already drawn. The test phrase for elaborate proposals: “why microservices at our size, and who operates them?” — a concrete answer separates architects from fashion salesmen.

Let’s talk about your project

Message us on WhatsApp or send an email — you talk directly to a developer.

office@northdan.com · +40 752 070 247

Why it matters for your business

Surgical scaling under heavy load

Only the overwhelmed component receives more resources — the whole application isn't multiplied for one hot spot.

Teams that don't block each other

Separate services, separate releases: the billing team ships today without waiting for, or breaking, the orders team.

Failures with a limited blast radius

A downed notifications service doesn't stop selling — fault isolation is designed into the architecture, not hoped for.

Frequently asked questions

We're commissioning an application — do we need microservices?

Almost certainly not at the start: below multiple teams and serious volume, a modular monolith delivers the same features faster, cheaper and with simple operations — and if success brings the pain of scale, it splits later along its internal boundaries. Warning signal: a proposal pitching microservices for a 20-user application is optimizing the team's CVs, not your cost.

What is a modular monolith — the compromise architects mention?

A single application at deployment time, organized internally into modules with strict boundaries (orders, billing, customers — each with its own code and rules, communicating through clear interfaces). It provides the order, and the readiness for a future split, without the operational tax of distributed services — for most companies, the best discipline-to-cost ratio available.

What are the hidden costs of microservices?

Operations: orchestration (Kubernetes or equivalent), distributed monitoring and logging, a delivery pipeline per service, managing the network between them — plus the engineers who master all that, an expensive and scarce species. On top comes logical complexity: transactions and consistency across services, tracing one flow through five systems. All solvable — and all billable, monthly, whether or not your scale justifies them.