How FSx for Lustre Works
FSx for Lustre is a fully managed service that makes it easy and cost-effective to launch and run the high-performance Lustre file system1
.
It is designed for applications that require fast storage, providing sub-millisecond latencies, up to hundreds of GBps of throughput, and up to millions of IOPS1
.
FSx for Lustre eliminates the traditional complexity of setting up and managing Lustre file systems, enabling you to spin up and run a high-performance file system in minutes1
.
It provides multiple deployment options so you can optimize cost for your needs1
.
FSx for Lustre is POSIX-compliant, so you can use your current Linux-based applications without having to make any changes1
.
Benefits of FSx for Lustre
It accelerates compute workloads with shared storage that provides sub-millisecond latencies, up to hundreds of GBs/s of throughput, and millions of IOPS
2
.
It allows you to access and process Amazon S3 data from a high-performance file system by linking your file systems to S3 buckets
2
.
It offers a range of deployment options to optimize price and performance
2
.
It provides a native file system interface and works as any file system does with your Linux operating system1
.
Limitations of FSx for Lustre
There are quotas for Amazon FSx for Lustre per AWS account, per AWS Region
3
.
There are limits on Amazon FSx for Lustre resources for each file system in an AWS Region
3
.
Creating large files with low stripe counts can cause I/O performance bottlenecks
4
.
Use Cases for FSx for Lustre
It can be used to accelerate machine learning (ML) by reducing training times with maximized throughput to your compute resources and seamless access to training data stored in Amazon S3
2
.
It enables high performance computing (HPC) by powering the most demanding HPC workloads with fast, highly scalable storage, natively integrated with AWS compute and orchestration services
2
.
It unlocks big data analytics by supporting thousands of compute instances running complex analytics workloads on petabytes of data2
.