Skip to main content

Amazon FSx for OpenZFS

Before you start

Prepare your environment for this section:

~$prepare-environment fundamentals/storage/fsxz
This will make the following changes to your lab environment:
    

  • Create an IAM OIDC provider
    • 

  • Create a new security group with rules necessary to access the Amazon FSx for OpenZFS file system from the EKS cluster
    • 

  • Create the Amazon FSx for OpenZFS Single-AZ 2 file system
    • 

You can view the Terraform that applies these changes here.


Amazon FSx for OpenZFS provides a fully managed, high-performance shared file system accessible via both NFSv3 and NFSv4. You have access to your shared data sets at microsecond latencies with millions of IOPS and up to 21 GB/s of throughput. FSx for OpenZFS also includes many enterprise features such as zero space snapshots, zero space clones, data replication, thin provisioning, user quotas, and compression.


There are two different storage classes; an all SSD-based storage class and an Intelligent-Tiering storage class. File systems leveraging the SSD storage class provide consistent microsecond latencies. Intelligent-Tiering file systems provide microsecond latencies for cached reads, 1-2 milliseconds latency for writes, and tens of milliseconds latency for read cache misses. The Intelligent-Tiering storage class provides fully elastic storage capacity growing and shrinking with your dataset only charging for the capacity consumed all at S3 like pricing.


In this lab, you will:


    

  • Learn about persistent network storage with the assets microservice
    • 

  • Configure and deploy the FSx for OpenZFS CSI Driver for Kubernetes
    • 

  • Implement dynamic provisioning using FSx for OpenZFS in a Kubernetes deployment
    • 



This hands-on experience will demonstrate how to effectively use Amazon FSx for OpenZFS with Amazon EKS for fully managed, high-performance, enterprise featured, elastic, persistent storage solutions.