Examkin

Question

Difficulty: MediumCompute Virtualization Solutions

A digital media company is designing the Azure compute infrastructure for a new video rendering and encoding platform. The platform consists of two main components:

1. An orchestration portal that manages rendering jobs, user profiles, and subscription details. This portal represents a steady-state production workload and requires a 99.9%99.9\% availability SLA.
2. A batch processing layer containing multiple worker instances that execute the CPU-intensive rendering tasks. These tasks write progress checkpoints to Azure Blob Storage, allowing interrupted jobs to resume from the last checkpoint.

You need to minimize compute costs for both layers while meeting the availability requirements.

Which two Azure compute configurations should you recommend?

  1. Azure Spot Virtual Machines to host the batch processing worker instancesAnswer
  2. Standard Azure Virtual Machines combined with Azure Reserved Virtual Machine Instances to host the orchestration portalAnswer
  3. C
    Azure Spot Virtual Machines to host the orchestration portal to minimize steady-state compute costs
  4. D
    Multiple standard Azure Virtual Machines deployed within a single Availability Zone to host the orchestration portal

Answer

Azure Spot Virtual Machines to host the batch processing worker instances, and Standard Azure Virtual Machines combined with Azure Reserved Virtual Machine Instances to host the orchestration portal.
The correct configurations align the workload requirements with the appropriate Azure compute features. The rendering workers write progress checkpoints and can tolerate interruption, making them ideal candidates for the cost-effective Azure Spot VMs. Conversely, the orchestration portal represents a steady-state production workload with a 99.9%99.9\% SLA constraint, requiring standard VM instances. Applying Azure Reserved Instances to these standard VMs is the correct way to reduce costs without risking eviction.

Step-by-Step Solution

1
Analyze the workload characteristics and SLA requirements for the orchestration portal.
The orchestration portal requires a steady-state production environment with a 99.9%99.9\% availability SLA, meaning it cannot tolerate evictions.
This rules out Spot VMs for this layer and necessitates standard VMs. Since it is a steady-state, predictable workload, Azure Reserved Virtual Machine Instances should be used to minimize costs.
2
Analyze the workload characteristics and fault tolerance of the batch processing layer.
The rendering workers perform CPU-intensive tasks that checkpoint progress and can tolerate interruptions.
Because the workload is interruptible, Azure Spot VMs are the most cost-effective option, offering up to a 90%90\% discount compared to pay-as-you-go rates.
3
Evaluate high availability design for the orchestration portal.
Deploying VMs across multiple Availability Zones or using zone-redundant VMSS configurations is required for resiliency against zone outages.
Placing all VMs in a single Availability Zone exposes the application to zone-level failures, which does not align with best practices for high availability.

Key Concept

Selecting cost-effective and resilient Azure compute solutions based on workload characteristics, SLA requirements, and fault tolerance.
Rate this question

Topics

Compute Virtualization Solutions

More questions from this topic