https://catalogartifact.azureedge.net/publicartifacts/Microsoft.AmlFileSystems-1.0.4/LargeFileSystem.png

Azure Managed Lustre File system

Microsoft

Azure Managed Lustre File system

Microsoft

Azure Managed Lustre provides an easy-to-deploy, fully managed Lustre® file system for on-demand use with data stored in Azure Blob.

Lustre is the world's most popular open-source parallel distributed file system. It provides the high throughput needed for computationally intensive workloads. Azure Managed Lustre is an easy to deploy, cloud-based Lustre file system that's integrated with Azure Blob. Store your data in Azure Blob and create a Lustre file system only when needed - when your workload is done, simply move output back to your long-term blob storage and delete the Lustre system.

Azure Managed Lustre provides an easy-to-deploy, fully managed Lustre® file system for on-demand use with data stored in Azure Blob.

Lustre is the world's most popular open-source parallel distributed file system. It provides the high throughput needed for computationally intensive workloads. Azure Managed Lustre is an easy to deploy, cloud-based Lustre file system that's integrated with Azure Blob. Store your data in Azure Blob and create a Lustre file system only when needed - when your workload is done, simply move output back to your long-term blob storage and delete the Lustre system.

Azure Managed Lustre gives you:

  • A customizable Lustre file system that can be deployed on demand in minutes
  • Full monitoring and maintenance for the Lustre system while you use it
  • The ability to store your data long-term in cost-effective Azure Blob storage - and access that data via NFS 3.0
  • The flexibility to create a small Lustre file system for a small job, or a massive system for a large task
  • The ability to hydrate your file system with only the data you need for the processing job, saving on storage costs and avoiding conflicts with other workflow demands
  • Integration with Azure services including Azure high-performance computing, Azure machine learning, and Azure HDInsight