Distinguishing Elasticity, Scalability, and Agility
2 min read
Lesson objectives
- Explain selecting the cloud property that addresses demand change without confusing capacity growth with experimentation speed.
- Apply the decision rule to a new case and justify the alternatives.
Distinguishing Elasticity, Scalability, and Agility
Scalability changes attainable capacity, elasticity matches capacity to demand in time, and agility shortens experimentation and delivery cycles. This lesson applies the rule by selecting the cloud property that addresses demand change without confusing capacity growth with experimentation speed.
An authorized user issued a valid DELETE and accidentally removed RDS data; the database and infrastructure remain healthy. What response fits? Decision: The customer performs point-in-time recovery to a new instance and reconciles the data. Reason: Recovery to an earlier point addresses customer deletion, while reconciliation and cutover remain customer tasks.
Inspector reports a vulnerability in a team-owned deployed image. The team must remediate the current deployment without claiming that a clean new image automatically closes a finding tied to an old image that still exists. What should it do? Decision: Rebuild and deploy a fixed image, rescan that new image identity, then separately verify the finding state and any affected old resources. Reason: This repairs the new deployment, binds evidence to its image identity, and prevents overclaiming the old finding state.
A team wants AWS to operate the relational database infrastructure and host OS while the team retains schema, users, data, database configuration, and applicable maintenance-window and optional-update choices. Which choice fits? Decision: Amazon RDS with customer responsibilities and choices at the database layer. Reason: RDS manages the hosting infrastructure while the customer retains data, schema, users, settings, and applicable maintenance choices.
A GuardDuty finding exists, and an analyst needs linked entities, activity, and historical context to understand what happened. Which service fits? Decision: Amazon Detective. Reason: Detective links entities, activity, and history to investigate an existing finding; it does not generate the original finding.
An organization wants supported security data centralized in an account-owned data lake and normalized to OCSF for cross-source analytics. Which service fits? Decision: Amazon Security Lake. Reason: Security Lake centralizes supported security data in a customer-owned lake and normalizes it to OCSF.
Baseline load grows 25% annually and is reviewed quarterly, while a surprise campaign raises demand eightfold for two hours. Which capacity behavior is required? Compound growth changes the baseline and requires scalability planning; the short spike requires elasticity. Substituting one for the other either constrains sustained growth or leaves peak capacity idle.
Lesson summary
Scalability changes attainable capacity, elasticity matches capacity to demand in time, and agility shortens experimentation and delivery cycles. Auto Scaling does not prove elasticity unless capacity response is measured, and more copies in one Availability Zone do not prove high availability.