5. What is spot instance? how it works? when it is used ?
Spot Instances are spare AWS server capacities offered at massive discounts of up to 90% off standard On-demand pricing. They run on a dynamic marketplace where costs fluctuate based on real-time infrastructure supply and demand. The critical trade-off is zero stability, as AWS can reclaim the hardware whenever standard customers need it. When AWS reclaims capacity, your system is given a strict 2-minute warning before abrupt termination.
- They are ideal for fault-tolerant workloads like big data analytics, batch processing, and machine learning training.
- They work perfectly with containerized microservices (Kubernetes/ECS) that automatically shift traffic between active servers.
- They are highly cost-effective for running automated software testing pipelines and CI/CD environments.
- They must never be used for production databases or monolithic applications requiring 24/7 uninterrupted uptime.
