Q31Design High-Performing Architectures
A solutions architect must host a high performance computing (HPC) workload in the AWS Cloud. The workload will run on hundreds of Amazon EC2 instances and will need parallel access to a shared file system to support distributed processing of large datasets. Datasets will be accessed simultaneously across multiple instances. The workload requires access latency of less than 1 ms. After processing is complete, engineers will need access to the dataset for manual postprocessing. Which solution meets these requirements?
← → navigate · a answer
Community votes
Discussion · 5
C 9
Amazon FSx for Lustre is a fully managed, high-performance file system optimized for HPC workloads. It is designed to deliver sub-millisecond latencies and high throughput, making it ideal for applications that require parallel access to shared storage, such as simulations and data analytics.
C 4
high performance computing (HPC) workloads, shared file system= Amazon FSx for Lustre
C 3
EFS could meet the latency requirement for most (!) read (!) operations, but this is not enough here. FSx for Lustre ist specifically designed for HPC.
C 2
host a high performance computing (HPC) workload -> FSx lustre
2
FSx luster for HPC