AWS

In short: “Amazon Web Services” — the largest cloud computing platform, offering hundreds of services (servers, storage, databases, AI) billed by usage instead of owning hardware.

In more detail: Instead of buying physical servers, you rent computing power (EC2), storage (S3), databases (RDS) and many other services from AWS based on actual consumption. AWS was one of the first major cloud providers in 2006 and remains market leader today, with Microsoft Azure and Google Cloud as its biggest competitors.

In Depth

Core services

AWS’s size is both its strength and its biggest hurdle for newcomers: with hundreds of services for practically every use case, almost any architecture can be built. The best-known core services:

• EC2 (“Elastic Compute Cloud”): virtual servers, billed by computing power/memory/hour.
• S3 (“Simple Storage Service”): object storage for any kind of file, highly scalable, often used for backups, images, videos.
• RDS (“Relational Database Service”): managed databases (PostgreSQL, MySQL and others), with backups/scaling handled by AWS.
• Lambda: serverless functions — code only runs when actually invoked, no permanently running server instance needed.
• IAM (“Identity and Access Management”): fine-grained access rights, who’s allowed to use which service and how.

A simple EC2 server (virtual machine) can be started via the command line, for example:

aws ec2 run-instances --image-id ami-12345 --instance-type t3.micro

Cost complexity as its own discipline

The sheer selection and the fine-grained, usage-based billing (often by the second/minute, data transfer volume, number of API calls) turn cost control and choosing the right services into their own discipline (“cloud architecture” or “FinOps” for the cost side). Unexpectedly high AWS bills from misconfigured or forgotten resources are a well-known, often-cited problem in the industry — a forgotten, running server or a misconfigured data transfer path can unnoticeably rack up significant costs.

Historical significance

AWS was one of the very first major cloud providers in 2006 (arising from Amazon’s internal infrastructure needs — Amazon itself had to run massively scalable infrastructure for its own retail platform and later marketed this capability as its own business), and thereby decisively shaped how cloud computing works today at all. To this day, AWS remains market leader by a clear margin over Microsoft Azure and Google Cloud, even though both competitors continuously gain market share.

Distinction from more specialised platforms

Unlike more specialised, more developer-friendly platforms like Vercel (which focus on a particular use case — frontend/Next.js deployments — and automatically configure a lot, e.g. CDN, SSL certificates, scaling), AWS requires considerably more manual configuration, but offers maximum control and flexibility in return — from individual virtual machines to global, highly available multi-region architectures for large enterprises. Many modern platforms like Vercel even run internally on AWS infrastructure themselves, but abstract away the complexity for their own users.

Availability zones and regions

AWS structures its global infrastructure into “regions” (geographically separate locations, e.g. Frankfurt, Ireland, Ohio) and, within those, “availability zones” (physically separate data centres within a region, with their own power supply/network connection). Applications running spread across several availability zones survive the failure of a single data centre without interruption — a central principle for highly available cloud architectures that would be barely economically replicable on classic owned hardware in this form.

See also: Vercel