Scalability
Vertical vs horizontal scale, bottlenecks, capacity planning.
Theory
Scalability is the ability to handle increasing load by adding resources. Vertical scaling (bigger CPU/RAM/disk) is quick but hits hardware ceilings. Horizontal scaling (more nodes) enables web-scale but demands stateless services and data partitioning.
Find the real bottleneck before scaling: CPU (compute-bound), memory (dataset size), disk I/O (database), network bandwidth, or lock contention. Amdahl's law: speeding up one component has diminishing returns if another is the limit.
Capacity planning uses measured peak QPS, growth projections, and load tests at 2× expected traffic. Auto-scaling handles diurnal variance; right-sized baseline capacity controls cost. Plan stateless tiers first — stateful tiers (databases) scale harder.
Production rollouts require idempotent automation, peer review, staged apply, and documented rollback — treat changes as production code.
Interviewers want STAR stories linking Scalability to measurable outcomes: fewer outages, faster deploys, lower cost, or reduced toil.
Architecture Diagram
Users / clients
|
Scalability
|
Core services
|
Data + observabilityExamples
# Scalability
# Vertical vs horizontal scale, bottlenecks, capacity planning.
# Validate in staging before production rollout.Interview Questions
What problem does Scalability solve?
It addresses the core use case described in production architecture — map features to reliability, scale, or velocity outcomes.
Key components of Scalability?
Identify inputs, outputs, control plane, data plane, and failure domains — interviewers want structured decomposition.
Common production pitfalls?
Misconfiguration, missing observability, no rollback path, and scaling bottlenecks under peak load.
How do you test changes safely?
Staging parity, canary/gradual rollout, automated health checks, and documented rollback.
Metrics to prove success?
Error rate, latency percentiles, throughput, cost, and toil reduction — pick one primary SLO.
Beginner vs advanced concern?
Beginners focus on setup; advanced teams focus on blast radius, security boundaries, and operability at 10× scale.
Best Practices
- Treat Scalability config as code with review and CI validation.
- Define SLOs and dashboards before production cutover.
- Document rollback and ownership for on-call.
- Use least privilege for credentials.
Common Mistakes
- Adopting Scalability without measurable success criteria.
- No staging environment mirroring production constraints.
- Missing rollback path during incidents.
- Undocumented on-call expectations.
Trade-off Analysis
Scalability improves vertical vs horizontal scale, bottlenecks, capacity planning. but adds operational and cognitive complexity — justify with load and team size.
Favor simplicity until metrics (p99 latency, error rate, cost) prove the pattern necessary.
Every redundancy layer trades capital/operational cost for availability — align with explicit SLO targets.
Document accepted inconsistency windows and recovery behavior before production cutover.
Cheat Sheet
Practical Exercises
Stand up Scalability locally or in free tier; document commands and failure recovery.
Introduce misconfiguration; practice detection and rollback under time limit.