Object Mount Fusion

Overview

Object Mount Fusion is designed to enhance local, LAN-based, high-performance, attached storage solutions (like Amazon’s Elastic File System, EFS) with the throughput of object storage.

It is a cheaper and faster solution when compared to using EFS alone.

Object Mount Fusion takes an attached storage filesystem and an (initially) empty object storage bucket/directory and exposes a single interface for both. Object Mount Fusion will migrate files between the object storage location and the local filesystem location depending on their best fit for both performance and cost.

How it Works

Object Mount Fusion combines both the local file storage and the cloud object storage into a single, virtual mount (a FUSE mount).

The files on object storage are represented as hidden links from the host filesystem to the object store.

Unlike other solutions, Object Mount enables cloud-based object storage to be used as a first-class, high-throughput tier, rather than a traditional slow archival tier.

Object Mount Fusion automatically and dynamically migrates files between the two locations according to your application behavior and usage. New files may be written into either location depending on predicted and observed access properties.

Object Mount Fusion supports multiple users, each simultaneously accessing files on multiple nodes, by combining and sharing the attached storage and mount locations.

Setting up Object Mount Fusion

Object Mount Fusion should be used within the same high-speed LAN as your object storage.

For example: If you use AWS S3, Object Mount Fusion should only be set up on EC2 nodes within the same AWS Region as the bucket to to be accessed.

If you are using an S3-compatible on-premises object storage solution, Object Mount Fusion should be set up on a computer on the same high-speed local area network (LAN).

Note: If you have already set up and are using an attached storage system (such as EFS), you may skip ahead to Mounting a Object Mount Fusion filesystem.

Otherwise, follow the steps below to configure Object Mount Fusion:

Setup an Empty Bucket on Object Storage

To use the object storage location when running Object Mount Fusion mount, you will need a location on object storage that is empty.

This will be used as the location that the Fusion filesystem will dynamically migrate data to when it’s more suitably stored on object storage.

This can either be an entirely empty bucket, or an empty directory on an existing bucket. This location must not be modified by anything but Object Mount Fusion’s filesystems.

Review the configuration steps below for your object storage provider:

Setup a New Compute Node

Object Mount Fusion needs to be set up on a compute node within the same LAN/region as the object storage bucket to be used.

You will also need to attach a high-speed file storage solution to the instance. It is recommended to configure this while creating the new instance/image.

To set up a compute node in the same region as your bucket, follow the relevant steps below for your provider:

Create an Empty Directory on Attached Storage

You will need an empty directory on your attached file storage that is writable by your user.

If you don’t have this set up yet, follow the relevant steps for your platform to attach a writable storage device to your compute node:

Mounting an Object Mount Fusion Filesystem

To set up Object Mount Fusion, you need Object Mount installed on the compute node, and you need to Import S3 Credentials so that the bucket/container is accessible.

Your next steps depend on the filesystem you are using and the options used when mounting/attaching it.

Refer to the relevant section below:

• Filesystems without extended attribute support (e.g. EFS)
• Filesystems with extended attribute support (e.g. NFSv4 with Linux Kernel 5.9+)

Filesystems Without Extended Attribute Support (e.g. EFS)

Object Mount Fusion is set up and accessed through Object Mount on FUSE.

To do this, additional options must be passed to cuno mount when it is run.

You will need to:

• Use the --fusion option to specify the path to the associated attached storage directory
• Use the --root option to specify the path to the object storage

Use a command similar to the one below to enable both these options:

cuno mount \
    --fusion "<path to attached-storage backing directory>" \
    --root "<path to object storage backing directory>" \
    "<mount location>"

cuno mount \
    --fusion "<path to attached-storage backing directory>" \
    --root "<path to object storage backing directory>" \
    "<mount location>"

Example:

`cuno mount --fusion /dev/sdf/fusion-store --root s3://bucket/fusion-store $HOME/my-fusion-filesystem`

`cuno mount --fusion /dev/sdf/fusion-store --root s3://bucket/fusion-store $HOME/my-fusion-filesystem`

By running the above command, the <mount location> becomes the proper way to access the Fusion filesystem.

All operations, workflows, pipelines, etc. should be pointed to <mount location>.

Mounting the Filesystem

Refer to your environment for specific commands:

$
cuno run mkdir "/cuno/s3/<bucket>/fusion-store"

$
cuno run mkdir "/cuno/s3/<bucket>/fusion-store"

$
mkdir "/mnt/fast/fusion-store"

$
mkdir "/mnt/fast/fusion-store"

$
cuno mount --fusion "/mnt/fast/fusion-store" --root "/cuno/s3/<bucket>/fusion-store" "~/my-fusion-filesystem"

$
cuno mount --fusion "/mnt/fast/fusion-store" --root "/cuno/s3/<bucket>/fusion-store" "~/my-fusion-filesystem"

Note: See tables below for additional cuno mount parameters.

Filesystems With Extended Attribute Support (e.g. NFSv4 with Linux Kernel 5.9+)

When using a filesystem that supports extended attributes you should “bind” it to a cloud location first, as this does additional error checking and makes for a simpler mount/unmount procedure.

Bind a Directory to a Cloud Location:

To bind a local Fusion directory to a cloud location, use cuno fusion.

This command will configure a local pointer targeted at the cloud mount we will make later. It will also save any relevant options used at this stage as metadata; to be used by default by any mounts that use this fusion binding.

$
cuno fusion "<path to attached storage backing directory>" "<path to object storage backing directory>"

$
cuno fusion "<path to attached storage backing directory>" "<path to object storage backing directory>"

Example:

$
cuno fusion /dev/sdf/fusion-store s3://bucket/fusion-store

$
cuno fusion /dev/sdf/fusion-store s3://bucket/fusion-store

Mounting the Filesystem (After Binding)

Object Mount Fusion is set up and accessed through Object Mount on FUSE.

To do this, additional options must be used when configuring Object Mount on FUSE.

If you have already bound a directory to a cloud location, you only need to use the --root option to specify the path to the associated attached-storage directory in order to use the mount as a Fusion mount:

$
cuno mount \
    --root "<path to attached-storage backing directory>" \
    "<mount location>"

$
cuno mount \
    --root "<path to attached-storage backing directory>" \
    "<mount location>"

Refer to your environment for specific commands:

$
cuno run mkdir "s3://<bucket>/fusion-store"

$
cuno run mkdir "s3://<bucket>/fusion-store"

$
mkdir "/mnt/fast/fusion-store"

$
mkdir "/mnt/fast/fusion-store"

$
cuno fusion "/mnt/fast/fusion-store" "s3://<bucket>/fusion-store"
$
cuno mount --root "/mnt/fast/fusion-store" "$HOME/my-fusion-filesystem"

$
cuno fusion "/mnt/fast/fusion-store" "s3://<bucket>/fusion-store"
$
cuno mount --root "/mnt/fast/fusion-store" "$HOME/my-fusion-filesystem"

Note: See tables below for additional cuno mount and cuno fusion parameters.

Using & Testing the Object Mount Fusion filesystem

All operations, workflows, pipelines, etc. should be pointed to your mount location.

Usage Examples

The following examples assume you have mounted a Fusion filesystem at $HOME/my-fusion-filesystem.

• Download data from the web into the Fusion filesystem:

  $
  cd $HOME/my-fusion-filesystem
  $
  wget http://vision.stanford.edu/aditya86/ImageNetDogs/images.tar
  

  $
  cd $HOME/my-fusion-filesystem
  $
  wget http://vision.stanford.edu/aditya86/ImageNetDogs/images.tar
  

  CopyCopied!

• List the files in the Fusion filesystem:

  $
  ls $HOME/my-fusion-filesystem/
  

  $
  ls $HOME/my-fusion-filesystem/
  

  CopyCopied!

• Unpack a tar archive:

  $
  cd $HOME/my-fusion-filesystem
  $
  tar -xf images.tar
  

  $
  cd $HOME/my-fusion-filesystem
  $
  tar -xf images.tar
  

  CopyCopied!

Testing

Check that access, modification and migration are working as expected.

We recommend measuring cost differences, and the overall time taken for equivalent jobs run on attached storage alone.

Note: You can set the CUNO_DEBUG environment variable to trace, or debug at the time of mounting for more details logging information.

Not seeing what you expect? Contact us at our 🌐 Support Team.

APPENDIX TABLES: Parameters, Arguments & Options

TABLE: cuno fusion Arguments

TABLE: cuno fusion Options

TABLE: cuno fusion Subcommands

TABLE: Relevant cuno mount Options for Object Mount Fusion: