A financial analytics company is designing the compute architecture on Google Compute Engine for two distinct workloads: Web Backend Service: An application with predictable resource demand that is licensed on a strict per-core basis. It requires an exact ratio of 6 vCPUs and 32 GB of RAM per VM instance to avoid software licensing overcharges while preventing out-of-memory errors. Batch Data Processing Pipeline: A stateless, highly distributed data analytics workload that is completely fault-tolerant and can resume processing automatically if worker instances are stopped unexpectedly. Management wants to minimize infrastructure and software licensing costs while maintaining operational flexibility without locking the batch processing pipeline into rigid long-term commitments. Which compute configuration and provisioning strategy should you recommend?

Configure custom machine types on the N2 series for the web backend service, and deploy Spot VMs within a Managed Instance Group (MIG) for the batch data processing pipeline.: ### What Custom Machine Types and Spot VMs Are **Custom machine types** allow administrators to configure exact vCPU and memory allocations on supported Compute Engine machine series (such as N1, N2, N2D, and N4) rather than being constrained to rigid predefined ratios. **Spot VMs** are virtual machine instances provisioned from surplus Google Cloud compute capacity offered at significant discounts (60%–91%) with the caveat that Compute Engine can preempt them if capacity is reclaimed. ### How This Solution Meets the Requirements - **Licensing Cost Optimization**: Custom machine types allow the web backend to be tailored exactly to **6 vCPUs and 32 GB RAM**, eliminating the need to select an oversized predefined instance (such as an 8-vCPU shape) that would incur unnecessary per-core third-party software licensing fees. - **Cost Efficiency for Fault-Tolerant Workloads**: The batch data processing workload is stateless and fault-tolerant, making it an ideal candidate for **Spot VMs**, which drastically reduce compute spend without compromising data integrity. - **Operational Flexibility**: Spot VMs require no upfront commitments or long-term contracts, providing complete flexibility to scale up or pivot infrastructure based on changing project demands. ### Key Advantages and Capabilities - **Precise Hardware Tailoring**: Configures vCPU count to match exact software licensing entitlements independently of memory. - **Steep Cost Reductions**: Saves up to 91% on compute expenses for non-critical and interruptible batch workloads. - **Automated Lifecycle Management**: Managed Instance Groups (MIGs) automatically recreate Spot instances when capacity becomes available after preemption. - **No Long-Term Lock-in**: Retains the ability to decommission, modify, or relocate batch workloads without unamortized contract penalties. ### Why This Is Optimal This architecture pairs precise resource sizing for core-licensed steady workloads with deeply discounted, pay-as-you-go preemptible resources for interruptible batch analytics, achieving maximum cost efficiency while keeping cloud infrastructure agile.

StudyCerts

Compute Resource Optimization Provisioning Models | Intrigued by the art of cloud architecture? Discover how to design, develop, and manage robust, secure, scalable, and dynamic solutions on Google Cloud as you prepare for the Professional Cloud Architect exam! | StudyCerts

Professional Cloud Architect

Intrigued by the art of cloud architecture? Discover how to design, develop, and manage robust, secure, scalable, and dynamic solutions on Google Cloud as you prepare for the Professional Cloud Architect exam!

Intermediate

Prepare and test your skills

Track Readiness

Set Exam Goal

Intermediate

Prepare and test your skills

Track Readiness

Set Exam Goal

Compute Resource Optimization Provisioning Models

Section: Flexibility of cloud resources

Worked example. The correct answer is already marked and every option is explained below, so there is nothing to select here. To answer questions yourself, start the free trial.

A financial analytics company is designing the compute architecture on Google Compute Engine for two distinct workloads:

Web Backend Service: An application with predictable resource demand that is licensed on a strict per-core basis. It requires an exact ratio of 6 vCPUs and 32 GB of RAM per VM instance to avoid software licensing overcharges while preventing out-of-memory errors.
Batch Data Processing Pipeline: A stateless, highly distributed data analytics workload that is completely fault-tolerant and can resume processing automatically if worker instances are stopped unexpectedly.

Management wants to minimize infrastructure and software licensing costs while maintaining operational flexibility without locking the batch processing pipeline into rigid long-term commitments.

Which compute configuration and provisioning strategy should you recommend?

Configure predefined memory-optimized machine types with 3-year resource-based Committed Use Discounts (CUDs) for the web backend, and use calendar-mode future reservations for the batch processing pipeline.

Configure custom machine types on the N2 series for the web backend service, and deploy Spot VMs within a Managed Instance Group (MIG) for the batch data processing pipeline.

Configure Spot VMs with extended memory for the web backend service, and deploy on-demand standard VMs for the batch data processing pipeline.

Configure predefined general-purpose machine types attached to spend-based Committed Use Discounts (CUDs) for both the web backend and the batch processing pipeline.

Professional Cloud Architect

Compute Resource Optimization Provisioning Models

Related questions