Answer-first summary for fast verification
Answer: Block Storage, Filestore
Google recommends against using block storage (such as persistent disks) and NAS solutions like Filestore for ML data storage due to their complexity compared to more streamlined options like Cloud Storage or BigQuery. It's also advised to avoid direct data reads from databases such as Cloud SQL. For optimal performance and ease of management, storing data in BigQuery and Cloud Storage is strongly recommended. For more details, refer to Google's best practices on ML in GCP: [https://cloud.google.com/architecture/ml-on-gcp-best-practices#avoid-storing-data-in-block-storage](https://cloud.google.com/architecture/ml-on-gcp-best-practices#avoid-storing-data-in-block-storage), [Cloud Storage documentation](https://cloud.google.com/bigquery/docs/loading-data), and [Vertex AI launch announcement](https://cloud.google.com/blog/products/ai-machine-learning/google-cloud-launches-vertex-ai-unified-platform-for-mlops).
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
You are a Machine Learning Engineer at a large consulting firm tasked with developing an NLP model to classify customer support needs and assess satisfaction levels from customer communications. The firm emphasizes cost-efficiency, scalability, and compliance with data protection regulations. The texts of various communications are stored across different types of storage in GCP. Considering the managed ML environments like Vertex AI and AI Platform, which two types of data storage should you avoid using to ensure optimal performance, ease of management, and adherence to Google's best practices? (Choose two)
A
BigQuery
B
Cloud Storage
C
Block Storage
D
Filestore
No comments yet.