What does "elasticity" in cloud computing refer to?

Concept

Elasticity in cloud computing is the system's ability to automatically add or remove compute resources so that the supplied capacity tracks the current workload in real time. The NIST definition (SP 800-145) calls this "rapid elasticity": capabilities are provisioned and released to scale outward and inward with demand.

Applying it here

When demand rises, such as during a traffic spike, the platform provisions extra instances; when demand falls, it releases them. The defining trait is this two-directional, demand-driven, automatic adjustment of capacity, that is, scaling up and scaling down. So elasticity means the ability to scale up and scale down compute resources.

Why the other choices differ

• Parallel execution of jobs is concurrency or throughput, running many tasks at once on existing resources, not changing how much capacity exists.

• High availability of systems is a reliability property, keeping services running with minimal downtime through redundancy and failover, not capacity adjustment.

• Minimum response time is a latency or performance target about how fast a single request is served, not the amount of provisioned resources.