Microsoft Certified: Azure Developer Associate (AZ-204)Develop Azure compute solutionsHard
A global media company is deploying a new video streaming platform that requires a highly available and scalable backend. The platform consists of several microservices, some of which require GPU acceleration for video transcoding, while others are standard web APIs. The company needs to manage container orchestration, ensure high availability, and optimize cost by using different types of compute resources within the same cluster. Which Azure compute solution is MOST appropriate for this scenario?
- AAzure App Service
- BAzure Kubernetes Service (AKS)
- CAzure Container Instances (ACI)
- DAzure Functions Premium Plan
Show answer & explanationAnswer & explanation
Correct answer: B. Azure Kubernetes Service (AKS)
Azure Kubernetes Service (AKS) is ideal for complex microservices architectures requiring container orchestration, high availability, and the ability to combine different node pools (e.g., standard VMs for web APIs and GPU-enabled VMs for transcoding) within a single cluster for cost optimization and specialized workloads.
Why the other options are wrong
- A. Azure App Service is a PaaS for web apps and APIs, not primarily designed for container orchestration of multiple microservices, especially those needing GPU acceleration.
- C. Azure Container Instances are suitable for single, isolated containers or simple container groups, but lack the orchestration, advanced networking, and mixed node pool capabilities needed for complex microservices with specialized hardware requirements.
- D. Azure Functions Premium Plan is for serverless functions, not for managing and orchestrating a full microservices platform with containerized services and GPU requirements.
Azure Kubernetes Service (AKS)
A managed Kubernetes service that simplifies deploying, managing, and scaling containerized applications in Azure.
- Container orchestration platform
- Supports high availability and scaling
- Integrates with Azure services
- Allows for multiple node pools (e.g., CPU, GPU)
Memory trick: AKS Keeps Orchestrating All Kinds of Services