Cloud Computing
Definition
On-demand delivery of servers, storage, and databases over the internet with pay-as-you-go pricing, replacing upfront hardware with variable costs.
Use Cases
- Netflix: Run large-scale video streaming and supporting services with elastic capacity — Migrated major parts of its infrastructure to AWS and uses a mix of cloud services (compute, storage, databases, and content delivery) to scale services and improve resilience. (Can scale to meet variable demand and improve reliability by using cloud regions and managed services instead of operating all infrastructure in its own data centers.)
- Airbnb: Operate a global marketplace with traffic spikes and rapid feature development — Uses cloud infrastructure (notably AWS) to run application services, data stores, and analytics, enabling teams to provision resources on demand rather than buying hardware. (Faster provisioning and the ability to handle seasonal and event-driven demand without long hardware procurement cycles.)
- Spotify: Deliver music streaming and personalization at global scale — Moved significant workloads to Google Cloud, using cloud compute and data services to support streaming, data processing, and machine learning workflows. (Improved agility for data and platform teams and the ability to scale infrastructure to support growth and new features.)
Frequently Asked Questions
- What's the difference between cloud computing and on-premises infrastructure?
- On-premises means you buy and run your own servers in your own data center (or a colocation facility). Cloud computing means you rent computing resources over the internet from a provider. With cloud, you can provision resources in minutes and pay for what you use; with on-premises, you pay upfront and manage hardware, power, cooling, and maintenance yourself.
- When should I use cloud computing?
- Use cloud computing when you want fast setup, flexible scaling, global reach, or managed services (like databases) without running the underlying hardware. It’s especially useful for startups, new projects, variable or unpredictable traffic, disaster recovery, analytics, and short-term environments (dev/test). On-premises can make sense when you have strict data residency requirements, specialized hardware needs, or very steady workloads where you can fully utilize owned equipment.
- How much does cloud computing cost?
- Cloud costs vary based on what you use: compute time (VMs/containers/serverless), storage amount and type, data transfer (especially outbound), managed database usage, and support plans. Pricing is typically pay-as-you-go, with discounts for committed usage or reserved capacity. Common cost drivers include always-on servers, large databases, high outbound bandwidth, and overprovisioned resources.
Category: cloud
Difficulty: basic
Related Terms
See Also