A global logistics provider is architecting a new tracking platform on Google Cloud. The system consists of two distinct components: (1) A stateless, event-driven gRPC web microservice that experiences unpredictable traffic spikes and requires rapid scaling down to zero to minimize idle costs, and (2) A legacy data transformation batch processing engine that requires custom OS kernel modules, specialized system drivers, and direct access to high-performance local NVMe storage. Which architectural compute platform choices should you recommend to satisfy these requirements while minimizing operational management overhead? (Select TWO.)
- Deploy the stateless gRPC microservice on Cloud Run to leverage managed auto-scaling to zero and native HTTP/2 support.Answer
- Deploy the legacy data transformation batch engine on Compute Engine Virtual Machines with attached Local SSDs.Answer
- CDeploy both the gRPC microservice and the legacy batch engine onto a single Google Kubernetes Engine (GKE) Autopilot cluster.
- DDeploy both application components onto App Engine Standard Environment using background threads and shared Memorystore cache.
- EDeploy the legacy data transformation engine on Cloud Run using mounted Cloud Storage buckets for high-throughput batch processing.
Answer
The optimal architecture combines Cloud Run for the stateless gRPC microservice and Compute Engine VMs with Local SSDs for the legacy data transformation engine requiring custom OS kernel modules.
Selecting Cloud Run for the stateless gRPC microservice satisfies the requirement for scale-to-zero, low maintenance overhead, and gRPC support. Selecting Compute Engine VMs for the legacy batch processing engine provides the necessary root OS access and kernel flexibility required to install custom drivers and utilize Local SSDs.
Step-by-Step Solution
Key Concept
Selecting GCP Compute Platforms based on OS kernel customization requirements, statefulness, protocol support, and operational overhead.
Estimated Time:2m 0s