Examkin

Question

Difficulty: HardDesign Compute High Availability

You are designing the compute infrastructure for a critical real-time logistics routing application in Azure.

The solution must meet the following requirements:
- The compute tier must achieve an uptime Service Level Agreement (SLA) of 99.99%99.99\%.
- The tier must support a mix of standard CPU-optimized Virtual Machines (VMs) for routing calculations and GPU-enabled VMs for machine learning models within the same logical group.
- The compute instances must support auto-scaling and rolling upgrades.
- VMs that collaborate on a specific regional route calculation must have network latency of less than 1 ms1\text{ ms} between each other.

Which compute configuration should you recommend?

  1. A Virtual Machine Scale Set in Flexible orchestration mode distributed across multiple Availability Zones, with a Proximity Placement Group created and assigned for each individual Availability Zone.Answer
  2. B
    A Virtual Machine Scale Set in Flexible orchestration mode using Spot Virtual Machines distributed across multiple Availability Zones, with a Proximity Placement Group created and assigned for each individual Availability Zone.
  3. C
    A Virtual Machine Scale Set in Flexible orchestration mode deployed within a single Availability Zone, with all instances associated with a single Proximity Placement Group.
  4. D
    An Availability Set containing the virtual machines, with all instances associated with a single Proximity Placement Group and deployed across multiple Availability Zones.

Answer

A Virtual Machine Scale Set in Flexible orchestration mode distributed across multiple Availability Zones, with a Proximity Placement Group created and assigned for each individual Availability Zone.
To achieve a 99.99%99.99\% uptime SLA, virtual machines must be deployed across multiple Availability Zones. To support heterogeneous VM sizes (mixing CPU-optimized and GPU-enabled instances) along with auto-scaling and rolling upgrades, a Virtual Machine Scale Set in Flexible orchestration mode is required. Finally, to ensure low-latency communication (less than 1 ms1\text{ ms}) within each zone without causing deployment failures or routing latency across zones, you must create and assign a Proximity Placement Group (PPG) for each individual Availability Zone.

Step-by-Step Solution

1
Evaluate the uptime SLA requirement.
To meet the 99.99%99.99\% SLA requirement, the Virtual Machines must be deployed across at least two Availability Zones in an Azure region.
Azure offers a 99.99%99.99\% SLA for VM uptime only when two or more instances are deployed across two or more Availability Zones.
2
Evaluate the VM grouping and scale requirements.
A Virtual Machine Scale Set in Flexible orchestration mode must be chosen.
Flexible orchestration mode allows mixing multiple VM families (heterogeneous sizes like CPU-optimized and GPU-enabled) in the same scale set while supporting auto-scaling and rolling upgrades.
3
Address the low-latency placement requirement across multiple zones.
Create a separate Proximity Placement Group (PPG) for each Availability Zone and associate them with the VMs in those zones.
A single PPG cannot span multiple Availability Zones without either causing deployment allocation failures or failing the low-latency requirement because zones are physically separated. A PPG per zone ensures intra-zone low latency while preserving cross-zone high availability.

Key Concept

Designing compute high availability using Virtual Machine Scale Sets in Flexible orchestration mode across Availability Zones combined with Proximity Placement Groups.
Estimated Time:2m 30s
Rate this question

Topics

Design Compute High Availability

More questions from this topic