AI-102 Question 200
Single answerYou are designing a Microsoft Azure AI solution to process large batches of incoming image data and perform real-time object detection. The solution must be scalable and cost-efficient, and it must process the images in parallel. Which Azure service or combination of services should you use to achieve this?
- A
Azure Cognitive Services Computer Vision API
- B
Azure Functions with Azure Blob Storage
- C
Azure Machine Learning with Azure Kubernetes Service (AKS)
- D
Azure Batch with custom object detection models
Show answer and explanation
Correct answer: D
Explanation
Azure Batch is specifically designed for running large-scale parallel and high-performance computing (HPC) workloads. In this scenario, the requirement is to process large batches of images with custom object detection models in a scalable and cost-efficient manner. Azure Batch allows you to run tasks across a pool of compute nodes, making it ideal for batch processing. Custom models can be integrated into the Batch workflow to meet the real-time object detection requirement.
- A. Incorrect.
The Azure Cognitive Services Computer Vision API is a pre-built AI service suited for general object recognition tasks. However, it may not be suitable for custom object detection scenarios requiring large-scale parallel processing.
- B. Incorrect.
Azure Functions with Azure Blob Storage is suitable for event-driven processing but is not ideal for large-scale, parallelized image processing due to its limitations in handling high throughput.
- C. Incorrect.
Azure Machine Learning with AKS is a good choice for real-time inferencing, but it can be more expensive and complex to manage for batch processing scenarios.
- D. Correct.
Azure Batch with custom object detection models is the best option for processing large batches of images in parallel. It is cost-efficient and designed for high-performance parallel computing, making it ideal for this scenario.