Fundamental
Prepare and test your skills
Fundamental
Prepare and test your skills
Using Autoscale with High Availability Replicas for Peak Demand
Free sample question, answered and explained below. The AZ-900 Microsoft Azure Fundamentals Exam practice bank has 688 more like it. Try 10 free: instant score, no account.
An online retail organization is preparing its cloud architecture for annual holiday sales events. During these events, customer traffic fluctuates unpredictably with massive, short-term spikes, while normal day-to-day operations experience moderate, steady traffic. The company must ensure business continuity with minimal downtime and maintain cost efficiency without manually provisioning capacity ahead of every spike.
Which cloud architecture strategy should the organization adopt to meet these operational and business continuity requirements?
Explanations
Statically provision compute capacity locked permanently at maximum estimated peak demand
Statically provision compute capacity locked permanently at maximum estimated peak demand
What Static Peak Provisioning Is
Static provisioning at peak capacity involves allocating and running the maximum compute and throughput resources required for the highest anticipated traffic spike continuously, 24 hours a day.
Why It Doesn't Meet the Requirement
- Severe Cost Inefficiency: The organization pays full price for peak capacity even during typical, moderate-traffic periods, wasting substantial budget.
- Lack of True Elasticity: It fails to leverage cloud scalability benefits, behaving like traditional, rigid on-premises infrastructure.
- Forecasting Risk: If actual traffic exceeds the estimated peak, the application will still face throttling or outages because capacity cannot dynamically scale beyond the fixed allocation.
Key Characteristics and Limitations
- Continuous billing for unutilized hardware resources during off-peak hours.
- Inflexible capacity boundaries that cannot adapt to unexpected spikes.
- High administrative overhead required to manually re-estimate and resize infrastructure.
This Answer Would Be Correct If
The application had a completely flat, predictable workload with continuous 24/7 high-volume transactions that never varied.
Implement provisioned throughput in autoscale mode paired with high availability replicas
Implement provisioned throughput in autoscale mode paired with high availability replicas
What Provisioned Autoscale with High Availability Is
Autoscale mode with provisioned throughput dynamically adjusts resource capacity in response to real-time workload demand, while high availability (HA) replicas ensure redundancy by maintaining secondary copies of data and compute instances.
How It Meets the Requirements
- Variable Traffic Management: Autoscale automatically scales compute and throughput up instantly during unexpected promotional spikes and scales back down when traffic subsides, optimizing operational costs.
- Business Continuity: High availability replicas provide fault tolerance by automatically failing over if a primary node or zone encounters an outage, eliminating service downtime.
- Operational Efficiency: Eliminates the need for administrators to perform manual capacity adjustments or guess peak traffic thresholds.
Key Advantages and Capabilities
- Dynamic Elasticity: Ensures resources are instantly available during sudden customer demand surges.
- Cost Governance: Charges reflect actual dynamic demand rather than continuously paying for maximum peak headroom.
- Built-in Redundancy: Protects against unplanned hardware or localized data center failures.
- Performance Predictability: Mitigates latency risks and prevents request throttling under heavy transactional loads.
Why This Is Optimal
Combining autoscale with high availability replication delivers the ideal balance of elasticity, reliability, and cost-efficiency. It avoids cold-start latency while maintaining active operational readiness and resilience against downtime during critical revenue events.
Deploy a serverless tier configured with a single replica to minimize baseline costs
Deploy a serverless tier configured with a single replica to minimize baseline costs
What Serverless with Single Replica Is
Serverless database compute runs on-demand and charges strictly per request, scaling down to zero during idle periods, while a single replica configuration provides no redundant failover nodes.
Why It Doesn't Meet the Requirement
- Cold-Start Latency: Serverless tiers often introduce cold-start latency after periods of inactivity, causing response delays that degrade customer experience during promotional spikes.
- Lack of High Availability: A single replica represents a single point of failure; if the underlying host or zone fails, the service experiences significant downtime.
- Production Suitability: It is optimized for intermittent or low-traffic development workloads rather than mission-critical retail applications.
Key Characteristics and Limitations
- Cold-start delays can disrupt real-time transactions during traffic surges.
- Absence of redundancy creates substantial business continuity risk.
- Strict throughput ceilings may throttle large simultaneous user bursts.
This Answer Would Be Correct If
The workload were an internal testing database or an intermittent, low-priority background task where cost minimization is the primary goal and occasional latency or downtime is acceptable.
Maintain a single-instance database and rely on disk IOPS bursting without redundancy
Maintain a single-instance database and rely on disk IOPS bursting without redundancy
What Single-Instance with IOPS Bursting Is
This configuration relies on a single database instance using storage disk bursting to temporarily handle I/O spikes without deploying redundant nodes or scalable compute.
Why It Doesn't Meet the Requirement
- High Risk of Throttling: When application demand surpasses disk limits, the system enters an I/O wait state and can become completely unresponsive.
- No Fault Tolerance: Relying on a single instance without high availability replicas ensures that any host hardware disruption leads to total downtime.
- Failure Propagation: In intense traffic scenarios, disk I/O exhaustion can cause cascading failures across interconnected service components.
Key Characteristics and Limitations
- Disk IOPS limits trigger throttling when transaction volume exceeds disk thresholds.
- Single point of failure prevents meeting basic business continuity standards.
- Unresponsive states occur when storage buffers and queues become backlogged under heavy load.
This Answer Would Be Correct If
The deployment were a non-critical development environment where hardware faults do not impact business revenue and temporary unresponsive states are tolerated.
Try 10 interactive questions like this one, free
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
- 688 practice questions across all 3 exam domains
- Every answer explained: why it is right, and why the others are not
- Scout, an AI tutor that answers your follow-up questions
- Readiness tracking by domain, so you know when to book the exam
Related questions
Keep the momentum going with these hand-picked practice scenarios
FundamentalAZ-900 Microsoft Azure Fundamentals Exam
Azure Autoscaling and High Availability
Describe the benefits of high availability and scalability in the cloud
FundamentalAZ-900 Microsoft Azure Fundamentals Exam
Azure Autoscale
Describe the benefits of high availability and scalability in the cloud
FundamentalAZ-900 Microsoft Azure Fundamentals Exam
Azure Availability Zones
Describe the benefits of high availability and scalability in the cloud
FundamentalAZ-900 Microsoft Azure Fundamentals Exam
Virtual Machine Scale Sets
Describe the benefits of high availability and scalability in the cloud
FundamentalAZ-900 Microsoft Azure Fundamentals Exam
Hybrid Cloud
Define cloud models, including public, private, and hybrid
IntermediateAZ-204 Developing Solutions for Microsoft Azure Exam
Azure Blob System Properties vs User-Defined Metadata
Set and retrieve properties and metadata
Last updated Validated