Describe cloud concepts
Free Microsoft Certified: Azure Fundamentals practice — 6 questions on Describe cloud concepts, with explanations. No sign-up. Full 12-question mixed test →
A healthcare company is migrating its patient record system to Azure. The company must ensure that they maintain full control over encryption key management and can prove to auditors that Microsoft cannot access patient data. Under the shared responsibility model, which statement is correct?
In the shared responsibility model, Microsoft always handles physical infrastructure (datacenters, cooling, power, physical security), while customers retain control over their data, including encryption key management, data classification, and identity/access management. This allows the customer to prove Microsoft cannot access their data by using customer-managed keys (CMK) or bring-your-own-key (BYOK) solutions.
A startup with unpredictable traffic patterns is choosing between a consumption-based pricing model and a reserved capacity model for their Azure App Service. Their traffic can range from 100 users during off-hours to 50,000 users during flash sales (occurring 3-4 times per year). Which pricing approach BEST optimizes cost?
Hybrid approach optimizes cost by using reserved instances (1-year or 3-year commitments with up to 72% discount) for predictable baseline load, while leveraging consumption-based pricing's elasticity for unpredictable spikes. You only pay for peak capacity when actually needed (3-4 times/year), avoiding wasted reserved capacity costs for 99% of the year.
An enterprise runs a legacy application that requires specific Windows Server 2012 patches and custom registry modifications. They want to migrate to Azure but must maintain these exact configurations for compliance reasons. After migration, they need Microsoft to handle physical hardware maintenance but want to retain full OS-level control. Which cloud service model meets these requirements?
IaaS (Azure Virtual Machines) provides full control over the OS, registry, and installed software while Microsoft handles physical infrastructure (compute, storage, networking hardware). You can upload custom VHD images with Windows Server 2012, specific patches, and registry modifications. Customer manages OS, applications, and runtime; Microsoft manages virtualization layer and below.
A financial trading platform requires 99.99% availability SLA and must process transactions within 50ms latency. During a regional Azure outage, the system must automatically failover to another region within 30 seconds with zero data loss. Which combination of cloud benefits directly addresses these requirements?
High availability (99.99% SLA achieved through Azure Availability Zones and multi-region deployments) combined with geo-distribution (deploying across multiple Azure regions) directly addresses uptime requirements and regional failover. Azure Traffic Manager, Azure Front Door, or geo-replicated databases enable automatic failover within seconds with geo-redundant storage (GRS) or zone-redundant storage (ZRS) ensuring zero data loss through synchronous replication.
A global manufacturing company operates factories in 15 countries with strict data residency laws requiring customer data to remain within national borders. They want to use a unified identity system and centralized management while ensuring data sovereignty. Which cloud deployment model satisfies these requirements?
Public cloud with Azure's 60+ regions worldwide allows deploying resources in specific geographic regions (e.g., Germany West for German data, Japan East for Japanese data). Azure Policy can enforce that specific customer datasets never leave designated regions, ensuring data residency compliance. Azure Active Directory provides unified identity across all regions while data stays geo-locked. This is more cost-effective and manageable than 15 separate private clouds.
A university research department has a data processing job that runs once per month, consumes 500 compute cores for exactly 8 hours, then requires zero resources for the remaining 29-30 days. The job uses custom Python libraries and requires root access to install system dependencies. Total monthly compute cost with always-on dedicated VMs is $12,000. Which approach reduces cost while meeting technical requirements?
Azure Batch (IaaS) provides job scheduling, automatic VM provisioning/deprovisioning, and supports low-priority VMs (up to 80% discount vs regular VMs). Scales to 500 cores for 8 hours, then auto-scales to zero - you pay only for 8 hours of compute monthly. Provides full OS control for custom libraries and system dependencies (root access). Estimated cost: ~$240-600/month (8 hours @ low-priority) vs $12,000 always-on.
Ready for the real thing?
The full course has two full-length practice tests, video lessons for every exam domain, hands-on labs and detailed answer explanations.
Start my full course on Udemy →