Examkin

Question

Difficulty: Very hardCompute Virtualization Solutions

A financial services company is designing a new payment processing application on Azure. The architecture consists of a transaction processing API tier and a stateful database tier. The application design must satisfy the following requirements:

* The database tier requires at least 256 GB of memory and high temporary storage throughput. To minimize SQL Server core-based licensing costs, the operating system must only see 8 vCPUs, while the database continues to utilize the memory and I/O performance of a 32-core virtual machine size.
* The transaction processing API tier handles highly sensitive financial data in transit. It must run in a secure, hardware-isolated execution environment that protects data in memory from external access, including the host operating system, hypervisor, and cloud administrators.
* The database tier must remain highly available and survive a primary datacenter outage without unexpected interruptions or evictions.

Which two compute virtualization solutions or configurations should you recommend to meet these requirements?

  1. Deploy the database tier on E32-8ds_v5 constrained vCPU virtual machinesAnswer
  2. Deploy the transaction processing API tier on DC-series or EC-series confidential virtual machinesAnswer
  3. C
    Deploy the database tier on E32ds_v5 Azure Spot virtual machines
  4. D
    Deploy all database tier virtual machines within a single Availability Zone

Answer

Deploy the database tier on E32-8ds_v5 constrained vCPU virtual machines and deploy the transaction processing API tier on DC-series or EC-series confidential virtual machines.
Deploying the database tier on E32-8ds_v5 constrained vCPU virtual machines provides the required 256 GB of memory and storage throughput of the E32ds_v5 size while limiting active vCPUs to 8, which minimizes SQL Server licensing costs. Deploying the transaction processing API tier on DC-series or EC-series confidential virtual machines meets the security requirement by encrypting memory in use at the hardware level using AMD SEV-SNP or Intel SGX, preventing access from the hypervisor and cloud operators.

Step-by-Step Solution

1
Analyze the database core licensing constraint.
Determine that a constrained vCPU virtual machine size is required to provide high memory (256 GB) and high I/O throughput while limiting the OS-visible vCPUs to 8 to minimize licensing costs.
Azure offers constrained vCPU VM sizes (such as E32-8ds_v5) specifically for database workloads where memory and storage performance are needed but licensing costs are bound to core counts.
2
Analyze the secure execution environment requirement for the API tier.
Determine that confidential virtual machines (such as DC-series or EC-series) are required to protect data in memory from external access, including the hypervisor and administrators.
Confidential computing uses hardware-based Trusted Execution Environments (TEEs) to encrypt memory in use.
3
Evaluate the database tier availability requirements.
Rule out Azure Spot VMs due to eviction risks, and rule out single-zone deployments due to lack of resilience against datacenter outages.
Production database workloads must run on standard VM tiers across multiple availability zones to satisfy high availability SLAs.

Key Concept

Azure compute virtualization selection, including constrained vCPUs for cost optimization, confidential computing for security in use, and high availability constraints.
Rate this question

Topics

Compute Virtualization Solutions

More questions from this topic