Exam AWS Certified Machine Learning - Specialty topic 1 question 35 discussion

Ans: A (S3) is most cost effective

A : S3 cost effective + athena ( not c redshift dont support unstructured data)

Amazon S3 (Simple Storage Service) is the best choice because it: Scales automatically to store an arbitrary number of datasets. Is cost-effective, as S3 charges only for storage used, unlike provisioned databases. Supports querying datasets with SQL using Amazon Athena. Is highly durable (99.999999999% durability) and optimized for large datasets. How It Works in This Scenario? Store datasets in S3 as files in Parquet, ORC, or CSV format. Use AWS Glue Data Catalog to create table metadata. Query the datasets using Amazon Athena (serverless SQL querying on S3). Automatically scale without worrying about storage limits.

'cost effective' --> AWS S3

A. YES - S3 + Athena/Presto B. NO - no SQL support C. NO - expensive to scale D. NO - DynamoDB is NoSQL

AWS S3 + Athena will do it

The most appropriate storage scheme for this scenario is option A: Store datasets as files in Amazon S3. Amazon S3 is a highly scalable and cost-effective object storage service that can store a large amount of data. S3 can scale automatically to accommodate a large number of datasets, making it a good option for storing the training data used in machine learning models. Additionally, S3 supports SQL querying through Amazon Athena or Amazon Redshift Spectrum, allowing data scientists to easily explore the data.

"store a large amount of training data commonly used in its machine learning models".. well it cannot be anything other than S3. Athena can query S3 cataloged data with SQL commands. Anwser is A

Amazon Redshift is not cost-effective.

I would say C https://docs.aws.amazon.com/redshift/latest/mgmt/working-with-clusters.html "For workloads that require ever-growing storage, managed storage lets you automatically scale your data warehouse storage capacity without adding and paying for additional nodes."

s3 is right

A, S3 is most cost effective