Google Associate Cloud EngineerEnsuring successful operation of a cloud solutionMedium
A company is migrating a legacy application to Google Cloud. The application uses a monolithic architecture and requires significant compute resources, including GPUs for specific workloads. The team needs full control over the operating system and software stack. They also anticipate needing to scale these resources up and down manually based on project phases. Which Google Cloud compute service is the most appropriate choice?
- ACompute Engine
- BCloud Run
- CApp Engine Flexible environment
- DGoogle Kubernetes Engine (GKE)
Show answer & explanationAnswer & explanation
Correct answer: A. Compute Engine
Compute Engine provides Infrastructure as a Service (IaaS), offering virtual machines with full control over the OS and software stack. It supports GPUs and allows manual scaling, making it ideal for monolithic applications requiring specific hardware and granular control.
Why the other options are wrong
- B. Cloud Run is a serverless platform for stateless containers, not suitable for monolithic applications requiring full OS control, GPUs, or manual scaling of underlying infrastructure.
- C. App Engine Flexible is a PaaS that offers more control than Standard but still abstracts away much of the infrastructure, which might not be sufficient for 'full control over OS and software stack' and specific GPU requirements.
- D. GKE is for containerized applications and, while powerful, might introduce unnecessary complexity for a monolithic legacy application that isn't containerized and requires full OS control.
Compute Engine
Google Cloud's Infrastructure as a Service (IaaS) offering that allows users to create and run virtual machines on Google's infrastructure.
- Provides full control over the operating system and software stack.
- Supports various machine types, including those with GPUs and custom configurations.
- Billing is per second for most resources, offering cost flexibility.
Memory trick: Compute Engine: Control Everything.