Scalability

A measurement of a system’s ability to grow to accommodate an increase in demand.
Both horizontally and vertically.
Not necessarily automatically.

Elasticity

The ability to acquire resources as you need them and release resources when you no longer need them.
Both horizontally and vertically.
Automatically.

An elastic resource is also scalable, the opposite is not always true.

High availability

The system will continue to function despite the complete failure of any component of the architecture. NAT GW, Elastricache, Redshift, RDS multiaz…

Fault tolerance

The system will continue to function without degradation in performance despite the complete failure of any component of the architecture. S3, DynamoDB, API GW, Cloudfront, Route53…

Resiliency

The ability of a workload to recover from infrastructure or service disruptions, dynamically acquire computing resources to meet demand, and mitigate disruptions, such as misconfigurations or transient network issues

Durability

The ability of a system to remain functional when faced with the challenges of normal operation over its lifetime

Availability

The percentage of time that a workload is available for use, where “available for use” means that it performs its agreed function when required.
It is a commonly used metric to quantitatively measure reliability.
Determined by percentage uptime, expressed in 9s.

Reliability

The ability of a workload to perform its intended function correctly and consistently when it’s expected to. This includes the ability to operate and test the workload through its total lifecycle.

Partition tolerance

A system’s ability to continue to operate correctly when it has been partial separated; often in reference to a distributed system.

Latency

A measurement of the amount of time between an action and the result, often between a request and a response.

Auto scaling

  • Unmanaged services (EC2, RDS…)
    When new resources are added, minimum reaction time is a few minutes.
  • Managed services (APIGW, S3, ALB, Lambda, ECS, DynamoDB…)
    Can scale instantly, or at least very fast.

Predictive scaling is based on ML and can be a good solution to scale faster

Infrastructure as Code (IaC)

  • Provision and manage cloud resources by writing human readable and machine consumable templates
  • Replicate, redeploy and repurpose your infrastructure
  • Rollback to the last good state on failures
  • Single source of truth
  • Version control infrastructure and app together
  • Build infrastructure and run it through your CI/CD pipeline

Continuous integration (CI)

The software development practice in which you continuously integrate (or check in) all code changes into a main branch of a central repository.
This practice enables you to verify your code changes early and often with an automated build and test process.

Continuous delivery (CD)

The software development practice in which all code changes are automatically prepared and always deployable (ready to go into production) at a single step.

Continuous delivery extends continuous integration to include testing production-like stages and running verification testing against those deployments.

Although continuous delivery can extend to a production deployment, it requires manual intervention between a code check-in and when that code is available for customer use.

Continuous deployment extends continuous delivery and is the automated release of software to customers, from check-in through production, without human intervention.

AWS Study Guide
New pages coming soon...