Q56 — AWS SAA-C03 Ch.1

Question 56 of 65 | ← Chapter 1

Q56. A company wants to use high performance computing (HPC) infrastructure on AWS for financial risk modeling. The company's HPC workloads run on Linux. Each HPC workflow runs on hundreds of Amazon EC2 Spot Instances, is short-lived and generates thousands of output files that are ultimately stored in persistent storage for analytics and long term future use. The company seeks a cloud storage solution that permits the copying of on-premises data to long-term persistent storage to make data available for processing by all EC2 instances. The solution should also be a high performance file system that is integrated with persistent storage to read and write datasets and output files. Which combination of AWS services meets these requirements?

Correct Answer: A. Amazon FSx for Lustre integrated with Amazon S3

Explanation

The combination of AWS services that meets the requirements of high-performance computing (HPC) infrastructure for financial risk modeling, including data copying, persistent storage, and high-performance file system, is option A: Amazon FSx for Lustre integrated with Amazon S3. Here's why this solution is the most suitable: 1. Amazon FSx for Lustre: Amazon FSx for Lustre is a high-performance file system optimized for HPC workloads. It provides low-latency access to data and high throughput for data-intensive workflows. It is well-suited for parallel compute workloads that require fast access to shared data sets. 2. Amazon S3: Amazon S3 is a highly scalable and durable object storage service. It is designed for storing and retrieving large amounts of data. It provides high durability and availability, making it suitable for long-term storage of output files and analytics data. By integrating Amazon FSx for Lustre with Amazon S3, the company gains the following benefits: - Data copying: The company can copy on-premises data to Amazon S3 for long-term persistent storage. This makes the data available for processing by all EC2 instances in the HPC workflow. - High-performance file system: Amazon FSx for Lustre acts as the high-performance file system for the HPC workloads. It provides fast and parallel access to the shared datasets and output files generated by the HPC workflows. - Integration with persistent storage: Amazon S3 serves as the persistent storage for storing the output files and analytics data. It provides durability and scalability, ensuring that the data is securely stored and can be accessed by other applications and workflows. Option B suggests using Amazon FSx for Windows File Server integrated with Amazon S3. However, FSx for Windows File Server is designed for Windows workloads and may not provide the same level of performance and scalability as FSx for Lustre for Linux-based HPC workloads. Option C suggests using Amazon S3 Glacier integrated with Amazon Elastic Block Store (Amazon EBS). However, S3 Glacier is an archival storage service with longer retrieval times, which may not be suitable for the real-time processing needs of HPC workloads. Option D suggests using an Amazon S3 bucket with a VPC endpoint integrated with Amazon EBS General Purpose SSD (gp2) volumes. However, Amazon EBS is block-level storage and may not provide the high-performance file system capabilities required for efficient parallel access to shared datasets in HPC workloads.