Study Hub / CompTIA Cloud+ (CV0-004) / Domain 2.0
CompTIA Cloud+ (CV0-004) Domain 2.0: Deployment — Explained
Domain 2.0 is how cloud resources get built and released: provisioning compute and storage, defining infrastructure as code, orchestrating containers, and choosing deployment strategies that don't cause outages.
Provisioning compute and storage
- Compute options — VMs (full control, slower to start), containers (light, fast, orchestrated), serverless/functions (no servers at all — code runs on events, billed per execution).
- Storage types — block (disks attached to VMs — databases), object (flat blob storage via API — backups, media, data lakes), file (shared filesystems — NAS-style). Pick by access pattern, not habit.
- Storage tiers and lifecycle — hot (frequent access), cool/archive (rare access, cheaper, slower retrieval). Lifecycle policies move aging data down tiers automatically — storage cost management on autopilot.
- Right-sizing — match instance size to actual workload: oversized burns money, undersized burns performance. Monitor, then resize — provisioning is a starting guess, not a commitment.
Infrastructure as code and orchestration
- IaC — define infrastructure in versioned templates (Terraform, CloudFormation, ARM/Bicep): declarative ("here's the desired state") beats imperative ("run these steps") for repeatability. Code-reviewed infrastructure, reproducible environments, no snowflake servers.
- Configuration management — tools like Ansible/Puppet/Chef keep systems in desired state after provisioning: packages installed, services configured, drift corrected.
- Container orchestration — Kubernetes (and simpler orchestrators): schedules containers, heals failures, scales with demand, rolls out updates. Containers at scale without orchestration is manual chaos.
- CI/CD — the deployment pipeline: commit → build → test → deploy, automated. Artifacts flow from code to production with gates (tests, approvals) in between.
Deployment strategies
- Rolling — replace instances gradually: some old, some new, zero downtime, slow rollout. The default safe choice.
- Blue-green — run two full environments; switch traffic from blue (current) to green (new) at once. Instant rollback (switch back) at the cost of double resources during deploy.
- Canary — route a small slice of real traffic to the new version first; expand if metrics look good. Real-world validation with blast-radius control.
- Immutable infrastructure — never modify running systems; deploy new ones and replace. Pairs with IaC: every deploy is a fresh, known-good build.
- Scaling — horizontal (more instances — the cloud-native way) vs. vertical (bigger instance — simple, with a ceiling). Auto-scaling policies add/remove capacity on metrics.
The must-memorize table: deployment strategies
How each strategy ships the new version.
| Strategy | How it works | Strength |
|---|---|---|
| Rolling | Replace instances gradually | Zero downtime, no extra infra |
| Blue-green | Two environments, switch traffic at once | Instant rollback |
| Canary | Small traffic slice first, then expand | Real-user validation, tiny blast radius |
| Immutable | Deploy new, replace old — never modify | Known-good builds, no drift |
Try it: 4 sample questions
Sample questions to check your understanding of this domain. Answers and full reasoning are tucked under each one.
1. A team wants deployments with instant rollback and is willing to run double infrastructure briefly during each release. Which strategy?
- A. A. Rolling
- B. B. Blue-green — full parallel environment with traffic switched at once; rollback is switching back
- C. C. Canary
- D. D. Recreate (all at once in place)
Show answer & reasoning
B. Blue-green's defining trade: double resources during deploy in exchange for atomic cutover and one-step rollback. Rolling (A) is cheaper but rollback is another slow rollout; canary (C) validates gradually but isn't instant. "Instant rollback, cost no object" → blue-green.
2. A payment service needs real-user validation of a new version but a bug could cost millions per minute. Which strategy limits the blast radius?
- A. A. Blue-green — 100% cutover immediately
- B. B. Canary — route a tiny slice of traffic first and expand on good metrics
- C. C. Rolling to all instances at once
- D. D. Deploy Friday night and watch
Show answer & reasoning
B. Canary exists for exactly this: prove the new version on 1–5% of real traffic, with automated rollback on bad metrics, before exposing everyone. Blue-green (A) flips everyone at once — fast rollback, but the blast already happened. Blast-radius control → canary.
3. What is the core advantage of infrastructure as code over manual console provisioning?
- A. A. It's faster to click through the console
- B. B. Versioned, reviewable, repeatable environments — no snowflake servers, reproducible disaster recovery
- C. C. It eliminates the need for monitoring
- D. D. It's always cheaper
Show answer & reasoning
B. IaC's value isn't speed of first deploy — it's repeatability: the same template builds dev, staging, and prod identically; changes are code-reviewed; recovery is "re-apply the template." Manual consoles produce unique snowflakes nobody can reproduce. Reproducibility is the point.
4. A workload needs double the instances during business hours and minimal capacity overnight. Which approach fits?
- A. A. Vertical scaling — buy the biggest instance once
- B. B. Horizontal auto-scaling on schedule/metrics — add instances for the peak, remove them after
- C. C. Serverless for everything regardless of fit
- D. D. Manual provisioning each morning
Show answer & reasoning
B. Predictable daily variation is the textbook auto-scaling case: scheduled or metric-driven horizontal scaling matches capacity to demand — paying for peak only during peak. Vertical (A) pays for peak 24/7; manual (D) is toil with a 6 AM alarm. Elasticity exists to be used.
This was one domain. The simulator drills all 6.
This free preview covered Deployment (Domain 2.0, 19%). Not covered here: Operations (Domain 3.0, 17%); Troubleshooting (Domain 6.0, 12%). Cloud operations and scenario troubleshooting stay in the paid guide and practice bank. The ByteBar Practice Exam Simulator drills every domain with 1,000 exam-style questions — $9.99 one-time unlocks all 28 exams. Prefer to read? The full Cloud+ Study Guide is in the ByteBar shop on Etsy.
Practice Exam Simulator — $9.99 one-time, all 28 exams Full Study Guides on Etsy