Examkin

Soru

Zorluk: OrtaSelecting and Designing Compute and Application Platforms

An global logistics enterprise is architecting an event-driven supply chain routing service on Google Cloud. The workload handles stateless HTTP webhook payloads containing container tracking events, which scale rapidly from zero during burst peak hours to thousands of requests per second. The application requires standard HTTP/2 transport and zero infrastructure management overhead. Additionally, the development team requires built-in automatic scaling to zero when no events are processing to minimize operational costs. Which two GCP compute platform design choices, when combined, fulfill these requirements with the lowest operational complexity?

  1. Deploy the stateless web service as a fully managed Cloud Run containerized service.Cevap
  2. Configure Cloud Run concurrency settings to allow multiple simultaneous requests per container instance.Cevap
  3. C
    Provision an Autopilot Google Kubernetes Engine (GKE) cluster with a Horizontal Pod Autoscaler (HPA) configured for CPU utilization.
  4. D
    Deploy the workload across an Instance Group of Compute Engine Spot VMs behind a Network Load Balancer.
  5. E
    Host the application on App Engine Flexible Environment configured with manual scaling.

Cevap

Deploy the stateless web service as a fully managed Cloud Run service, and configure Cloud Run concurrency settings to process multiple simultaneous requests per container instance.
For stateless HTTP workloads that experience unpredictable burst traffic and require zero operational overhead alongside scale-to-zero capability, Cloud Run is the optimal Google Cloud platform. Enabling container concurrency further maximizes throughput and minimizes cost by allowing each running container instance to process multiple requests concurrently.

Adım Adım Çözüm

1
Analyze workload state and scaling requirements
The application is a stateless HTTP microservice with bursty traffic patterns requiring scale-to-zero functionality and minimal management overhead.
Identifying statefulness and scaling characteristics determines whether serverless or container orchestration platforms are optimal.
2
Evaluate compute platforms against operational overhead constraints
Cloud Run provides serverless container hosting that automatically scales down to zero when idle and manages all underlying infrastructure.
Alternative platforms like GKE or Compute Engine MIGs introduce unnecessary management overhead and baseline execution costs for simple stateless webhooks.
3
Optimize throughput and cost efficiency for HTTP workloads
Configuring request concurrency on Cloud Run enables a single container instance to serve multiple concurrent HTTP connections efficiently.
Concurrency tuning reduces the number of container instances needed during sudden traffic spikes, lowering costs while maintaining responsiveness.

Anahtar Kavram

Selecting serverless compute platforms (Cloud Run) over container orchestrators (GKE) for stateless HTTP workloads requiring scale-to-zero and low operational overhead.
Bu soruyu puanla

Konular

Selecting and Designing Compute and Application Platforms