Answer-first summary for fast verification
Answer: Create an Amazon S3 bucket to store the raw data. Create an Amazon FSx for Lustre file system that uses persistent SSD storage. Select the option to import data from and export data to Amazon S3. Mount the file system on the EC2 instances.
## Explanation **Correct Answer: B** **Key Requirements Analysis:** 1. **Sub-millisecond latencies** - Requires high-performance storage with low latency 2. **Minimum throughput of 6 Gbps** - Requires high-throughput storage 3. **8 TB of data** - Moderate data size 4. **Hundreds of Amazon EC2 instances** - Parallel processing needs **Why Option B is Correct:** 1. **Amazon FSx for Lustre with persistent SSD storage** provides: - **Sub-millisecond latencies** (typically 0.2-0.5 ms) - **High throughput** (up to hundreds of Gbps depending on configuration) - **Parallel file system** optimized for high-performance computing workloads 2. **Integration with Amazon S3** allows: - Cost-effective long-term storage in S3 - Efficient data import/export between S3 and Lustre - S3 as a data lake with Lustre as a high-performance cache 3. **Persistent SSD storage** vs HDD: - SSDs provide much lower latency and higher IOPS - HDDs cannot meet sub-millisecond latency requirements **Why Other Options are Incorrect:** **Option A (FSx for NetApp ONTAP with tiering policy ALL):** - FSx for ONTAP is designed for enterprise NAS workloads, not HPC - Tiering policy ALL moves cold data to capacity pool storage (S3), which increases latency - Not optimized for sub-millisecond latencies required by HPC workloads **Option C (FSx for Lustre with persistent HDD storage):** - HDD storage cannot provide sub-millisecond latencies - HDDs have higher latency (typically 3-10 ms) - Throughput may be limited compared to SSD-based Lustre **Option D (FSx for NetApp ONTAP with tiering policy NONE):** - FSx for ONTAP is not designed for HPC workloads - Even with tiering disabled, it doesn't provide the same performance characteristics as Lustre - Lustre is specifically designed for parallel processing across hundreds of instances **Additional Considerations:** - FSx for Lustre is purpose-built for high-performance computing, machine learning, and media processing - The solution scales to support hundreds of EC2 instances accessing data concurrently - 6 Gbps throughput is easily achievable with FSx for Lustre (can scale to hundreds of Gbps) - The 8 TB data size is well within Lustre's capabilities
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
No comments yet.
A research laboratory needs to process approximately 8 TB of data. The laboratory requires sub-millisecond latencies and a minimum throughput of 6 Gbps for the storage subsystem. Hundreds of Amazon EC2 instances that run Amazon Linux will distribute and process the data.
Which solution will meet the performance requirements?
A
Create an Amazon FSx for NetApp ONTAP file system. Set each volume's tiering policy to ALL. Import the raw data into the file system. Mount the file system on the EC2 instances.
B
Create an Amazon S3 bucket to store the raw data. Create an Amazon FSx for Lustre file system that uses persistent SSD storage. Select the option to import data from and export data to Amazon S3. Mount the file system on the EC2 instances.
C
Create an Amazon S3 bucket to store the raw data. Create an Amazon FSx for Lustre file system that uses persistent HDD storage. Select the option to import data from and export data to Amazon S3. Mount the file system on the EC2 instances.
D
Create an Amazon FSx for NetApp ONTAP file system. Set each volume's tiering policy to NONE. Import the raw data into the file system. Mount the file system on the EC2 instances.