Soru

Zorluk: OrtaSelecting and Designing Compute and Application Platforms

An enterprise logistics organization is designing the architecture on Google Cloud for two new distinct internal applications. The first application is an event-driven stateless HTTP webhook service that receives unpredictable bursty traffic and must scale to zero instances during quiet hours to minimize idle costs. The second application is a legacy backend routing daemon that requires custom OS kernel network parameters (sysctl tuning) and raw non-HTTP TCP protocol bindings. Which TWO compute architecture options should you select to meet these technical requirements while minimizing operational overhead? (Select TWO)

  1. Deploy the stateless event-driven HTTP webhook service to fully managed Cloud Run.Cevap
  2. Deploy the legacy backend routing daemon requiring kernel sysctl tuning to Compute Engine Virtual Machines.Cevap
  3. C
    Deploy the stateless event-driven HTTP webhook service onto a dedicated Google Kubernetes Engine (GKE) Standard cluster.
  4. D
    Deploy the legacy backend routing daemon to Cloud Run by packaging the custom sysctl kernel modules inside the container image.
  5. E
    Purchase 3-year Committed Use Discounts (CUDs) for the peak resource requirements of the stateless HTTP webhook service.

Cevap

Select Cloud Run for the stateless HTTP webhook service and Compute Engine VMs for the legacy backend routing daemon.
Cloud Run provides serverless stateless container execution with scale-to-zero functionality, perfectly matching the HTTP webhook workload while minimizing operational overhead. Compute Engine provides root-level OS access and kernel customizability needed for sysctl tuning and custom network protocols.

Adım Adım Çözüm

1
Analyze requirements for the first application (stateless HTTP, bursty traffic, scale-to-zero requirement).
Fully managed Cloud Run fits stateless HTTP microservices with zero-instance scaling capability and zero baseline operational cost.
Cloud Run abstracts cluster management while providing pay-per-use request scaling down to zero.
2
Analyze requirements for the second application (custom OS kernel parameters, sysctl tuning, raw TCP).
Compute Engine Virtual Machines provide full OS access needed to modify kernel sysctl parameters and run non-HTTP network protocols.
Serverless container platforms like Cloud Run run in restricted sandboxes and do not allow host kernel modification.

Anahtar Kavram

Selecting GCP Compute Platforms based on statefulness, OS control, scaling characteristics, and operational overhead.
Bu soruyu puanla