Auto-Scaling

Kind: capability

Layer: Correctness Core

Aliases: Demand-Based Capacity

Record: architecture:auto-scaling

Severity: contextual

Scope: deployment, service, infrastructure

Canonical: Ontology

The ability to change the number of running instances automatically, between set bounds, from a load metric.

Listed in Architecture principles, after Replication and before Auto-Remediation.

Requires

• Horizontal Scalability
• Metrics

Reinforces

• Elasticity

In tension with

• Cost/Cold Start
• Fixed Capacity

Tensions

• Auto-Scaling / Cost/Cold Start
• Auto-Scaling / Fixed Capacity

Violated by

• Fixed-Capacity Design

Refactored by

• Autoscaling Policy
• Externalize Session State

Severity

• contextual

Category

• Self-Healing / Recovery / Deployment Safety

Enabled by

• Statelessness

Linked from

• Scalability / Performance / Optimization
• Core Vocabulary
• Scalability / Performance / Optimization
• Self-Healing / Recovery / Deployment Safety
• Severity levels
• The resolutions