By default, Koa users are granted access to two storage locations:

1. home storage (home)

2. scratch storage (Koa Scratch)

The two additional types of storage locations could be granted to a user, but are typically tied to a lab or project:

1. Lab storage (free)

2. KoaStore storage (for fee)

Common Storage Locations

Home (Free)

Each user is provided a home storage location on Koa. Each user's home is provided 50 GB of space that can be used as the user chooses. Upon initialization, user homes are setup with an examples directory, a symlink for Koa Scratch and other basic configurations needed to allow users to take advantage of Koa.

Performance

Home is not a high performance filesystem. Network File System (NFS) is tried and true but it is not designed to handle the stresses an High Performance Computing cluster can place on it. As a result, home should not be used for writing output from multiple jobs or jobs that can generate a lot of data. Jobs of this nature should take advantage of Scratch instead, which is a file system designed for the stresses an HPC can generate.

Home can be suitable for users with very modest/light needs and files that need to persist on Koa.

Details

Home is utilizing a Zoned File System (ZFS) over NFS Remote Direct Access (RDMA). Currently, home is served out utilizing a single Virtual Host that is connected to Koa at 200 Gbit. ZFS is configured to use zstd-3 compression allowing users to take full advantage of the inline compression and space savings zstd provides. As of this writing, on average users see a 2.16x compression ratio.

Open

Diagram showing how the home ZFS file system utilizes file based disks that are backed on Koa Storage

Scratch (Free)

Scratch, also known by the symlink “koa_scratch”, provides each user access to an 800TB pool of storage on which files may live 90 days on from the last time they were modified. In total, this file system can support up to 400,000,000 files and directories. Users are not provided an individual quota allowing for flexibility based on need. Scratch directly accesses the underlying Koa Storage system providing the high performance possible from the storage system.

Performance

Scratch provides direct access to the underlying high performance file system: Koa Storage, which utilizes Lustre. Lustre is designed for situations where many servers and workloads need to read and write data as quickly as possible. While Lustre works best with long sequential read/writes, and exhibits poorer performance with small random reads, methods exist to work around this limitation. Workarounds would include the use of squashfs as we cover in our documentation or archive formats that combine multiple small files that align with your workflow, such as HDF5.

For scalability and performance, files written to Scratch use a progressive file layout in which at certain size boundaries switch to different types of storage medium or recruit more storage targets to store parts of a file.

Automatic File Purging

Parameters of the automated file purge

• Only files under ~/koa_scratch/ or /mnt/lustre/koa/scratch/${USER} will be subject to purge

• Files and folders not modified for 90 days will be deleted from scratch

• The purge process will run daily

Details

Scratch, or more specifically, Koa Storage utilizes Lustre with ZFS as its underlying file system. For transport to Koa, it also utilizes RDMA using multiple servers all connected at 200Gbit Infiniband. The ZFS components are setup to provide the same zstd-3 compression seen on the home file system providing space savings and in some cases faster access to files as you need to read less data from a slower storage medium. The underlying storage system utilizes a mixture of spinning enterprise hard drives (HDD) and Solid State storage, (SAS SSD and NVMe).

Metadata

Metadata is stored in separate targets from the file data, which for performance reasons is entirely using Solid State Drives. The metadata is split up among multiple targets, which are then served out by different servers. In case of a server failure, “failing-over” the storage to another server is possible allowing for minimal downtime. Each folder is assigned to one of the Metadata storage targets and all files under it are also assigned to that target. Load balancing is done in some cases to try and balancing the different metadata targets so they do not grow too out of sync size wise.

Object Data

Object data is stored on a mixture of spinning enterprise hard drives, SAS Solid State Drives and NVMe.

Storage targets are split up into different configurations providing at least a 2 disk parity. Currently, Koa has 58 Object Storage Targets, of which 10 targets are SAS SSD or NVMe.

Object data is written to Scratch using a progressive file layout (PFL). The current PFL setup used by scratch follows the following rules:

• The first 512K of every file is written to a single solid state (SSD or NVMe) storage target

• Next, file data up to 64MB is written to a single HDD target

• Next, file data up to 512MB is written to two HDD targets in 4 MB stripes

• Next, file data up to 1 GB is written to four HDD targets in 4 MB stripes

• Finally, any remaining data for a file is written to eight HDD targets in 4 MB stripes

Open

Diagram showing how Scratch, or more specifically the Koa Storage system is divided and connected to Koa providing direct access to

Per lab or project storage

Lab and purchased storage processing

Lab (Free)

Lab storage is limited to one free storage location per Lab at UH. A lab will be granted 500GB of space and a limit of 512,000 files and directories. The lab will be assigned a posix group which users can be added and removed from upon request. A lab space will allow them to share applications, data and environments between members of the lab.

Details

Lab storage has direct access to the Koa Storage system, just like Scratch. It follows a different progressive file layout, in which solid state storage is not utilized.

KoaStore (For Fee Storage)

KoaStore is storage that is leased year to year on Koa in 1 TB increments with a limit of 1,024,000 files and folders per TB leased. Funds for this storage must come from an account at UH. Access control to this storage for a given group, lab or project is handled through a posix group which users can be added and removed from upon request.

Details

Lab storage has direct access to the Koa Storage system, just like Scratch. It follows a different progressive file layout, in which solid state storage is not utilized.