Knowledge hub · 4 guides

Cloud

Renting computing power, storage and ready-made services from AWS, Microsoft Azure or Google Cloud instead of running your own data centre.

What it is

Cloud means renting infrastructure and managed services by the hour instead of buying servers. The three hyperscalers (AWS, Microsoft Azure and Google Cloud) run data centres in dozens of regions and sell hundreds of services on top: virtual machines, databases, analytics, AI and more.

Most enterprises now run a mix: some workloads in the cloud, some still on-premises, and often more than one provider. IT service firms help in three ways: planning and running migrations, building cloud-native platforms, and operating the estate day to day, including keeping the bill under control (FinOps).

Explainer

Who manages what

	On-premises	IaaS	PaaS	SaaS
Data & access	You manage	You manage	You manage	You manage
Applications	You manage	You manage	You manage	Provider manages
Runtime	You manage	You manage	Provider manages	Provider manages
Middleware	You manage	You manage	Provider manages	Provider manages
Operating system	You manage	You manage	Provider manages	Provider manages
Virtualisation	You manage	Provider manages	Provider manages	Provider manages
Servers & storage	You manage	Provider manages	Provider manages	Provider manages
Networking	You manage	Provider manages	Provider manages	Provider manages

You manage
Provider manages

The further right you go, the more the provider runs for you and the less you can tune. Your data, and who can access it, stays your job everywhere.

Explainer

Four ways to move an application

← Least change, fastestMost change, most benefit →

Rehost
“Lift and shift”: move servers as they are.
Example
Copy VMs to cloud VMs
Replatform
Small changes that cut running effort.
Example
Move to a managed database
Repurchase
Replace the application with SaaS.
Example
On-premises CRM to Salesforce
Refactor
Rebuild it cloud-native.
Example
Monolith to containers and serverless

Most migrations mix all four. Two more options sit outside the line: retire what nobody uses, and retain what has to stay put for now.

Guides

Guides in Cloud

Amazon Web Services (AWS)

The largest public cloud, with the broadest catalogue of services.

Cloud-native products that need many managed building blocks
Large migrations where partner capacity and AWS funding programmes help
Teams that want the deepest hiring pool

Typical project: 4–8 weeksSee vendors

Microsoft Azure

Microsoft's cloud: the natural home for Windows, SQL Server and Microsoft 365 estates.

Organisations standardised on Microsoft 365, Windows and SQL Server
Hybrid estates that keep some workloads on-premises
Teams building on Dynamics 365 and Power Platform

Typical project: 4–8 weeksSee vendors

Google Cloud

Google's cloud, strongest in data, analytics, AI and Kubernetes.

Analytics platforms built around BigQuery
Containerised apps on GKE or Cloud Run
AI projects using Gemini and Vertex AI

Typical project: 2–6 monthsSee vendors

Kubernetes & containers

Package an app once and run it anywhere: the standard way to run modern software at scale.

Microservice architectures with many teams shipping often
Running one platform across clouds or on-premises
Workloads with spiky demand

Typical project: 6–16 weeksSee vendors

At a glance

Guide	Best fit	Pricing	Typical project
AWS	Cloud-native builds and large migrations needing the widest choice	Pay as you go; Savings Plans and Reserved Instances for one- to three-year commitments	4–8 weeks
Azure	Microsoft-centred estates, hybrid setups and enterprise identity	Pay as you go; reservations and savings plans; Azure Hybrid Benefit reuses existing Windows Server and SQL Server licences	4–8 weeks
Google Cloud	Data- and AI-led workloads, BigQuery analytics and Kubernetes	Pay as you go; sustained-use and committed-use discounts	2–6 months
Kubernetes	Many services, frequent releases and portability across clouds	Free software; you pay for the machines it runs on, plus a per-cluster fee on some managed services	6–16 weeks

Glossary

Key terms

Region & availability zone: A region is a geographic location; zones are separate data centres inside it, used for resilience.
IaaS, PaaS, SaaS: Infrastructure, platform and software as a service: how much of the stack the provider runs for you.
Landing zone: The ready-made foundation (accounts, network, identity, guardrails) every workload lands in.
FinOps: Engineering and finance working together to track and reduce cloud spend.
Reservations & savings plans: Discounts in return for committing to one to three years of usage.
Hybrid & multi-cloud: Running across on-premises and cloud, or across several cloud providers.

Also worth reading

Other areas

Find Cloud partners

IT service companies with these skills on their profile, every fact labelled with its source.

See vendors Post a project