Elasticity vs scalability

1. Vertical vs horizontal scalability #

Understanding how to increase processing power.

Scale-up vs scale-out #

Vertical scalability (scale-up) means adding RAM or CPU to a server. It has physical limits. Horizontal scalability (scale-out) means adding new identical servers, offering virtually unlimited growth.

2. Elasticity: autoscaling in action #

Elasticity is a system's ability to automatically add or remove resources based on precise metrics (CPU > 70% or network latency). It is what lets you save money at night and handle the load during sales.

3. Designing for high availability #

To be elastic, an application must be 'stateless' (no local state). Sessions must be stored in external databases or caches (Redis) so that servers can appear and disappear without cutting users' connections.

Hands-on: scaling #

Scalability commands.

Diagram #

Examples #

# Scalabilité horizontale manuelle (Kubernetes)
kubectl scale deployment web --replicas=6
# Élasticité : ajustement automatique selon le CPU
kubectl autoscale deployment web --cpu-percent=70 --min=2 --max=20

# AWS Auto Scaling : politique de suivi de cible
aws autoscaling put-scaling-policy --auto-scaling-group-name web-asg --policy-name cpu70 --policy-type TargetTrackingScaling \
  --target-tracking-configuration '{"PredefinedMetricSpecification":{"PredefinedMetricType":"ASGAverageCPUUtilization"},"TargetValue":70}'