Answer-first summary for fast verification
Answer: Opting for an AI Platform Training job with a custom scale tier specifying 4 V100 GPUs and Cloud Storage, which offers a fully managed, scalable, and cost-efficient solution for model training.
The most cost-effective and efficient option for this scenario is to use an AI Platform Training job with a custom scale tier specifying 4 V100 GPUs and Cloud Storage. This approach leverages a fully managed platform for training machine learning models, eliminating the need to manage underlying infrastructure while utilizing GPUs effectively. Cloud Storage offers a scalable and cost-efficient solution for storing training data. Incorrect options: - **A Deep Learning VM with 4 V100 GPUs and local storage**: Managing VMs incurs higher costs and effort, and local storage may not accommodate large datasets efficiently. - **A Deep Learning VM with 4 V100 GPUs and Cloud Storage**: While Cloud Storage improves data storage, the VM management overhead remains. - **A Google Kubernetes Engine cluster with a V100 GPU Node Pool and an NFS Server**: This setup introduces unnecessary complexity and higher costs compared to the managed AI Platform Training solution.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
No comments yet.
In the context of deploying a machine learning model for image classification using transfer learning with a pre-trained EfficientNet model, you are tasked with selecting the most cost-effective and efficient platform components and configuration environment. The model will be retrained daily on a dataset of 20,000 images. Key considerations include minimizing infrastructure costs, ensuring scalability for daily retraining, and leveraging cloud storage for the dataset. Which of the following options best meets these requirements? (Choose one correct option)
A
Deploying a Google Kubernetes Engine cluster with a V100 GPU Node Pool and an NFS Server for data storage, which offers flexibility but requires manual management and incurs higher costs.
B
Utilizing a Deep Learning VM instance with 4 V100 GPUs and local storage, which provides high performance but lacks scalability and cost-efficiency for daily retraining.
C
Opting for an AI Platform Training job with a custom scale tier specifying 4 V100 GPUs and Cloud Storage, which offers a fully managed, scalable, and cost-efficient solution for model training.
D
Choosing a Deep Learning VM with 4 V100 GPUs and Cloud Storage, which improves data storage scalability but still involves VM management overhead and higher costs compared to managed services.