Skip to main content

UQ Research Data Manager

Provides information about how to use the UQ Research Data Manager (UQRDM) to store and manage research data for research projects.

1. How to access the storage via HPC facilities

You will only be able to access your allocation folder via HPC facilities, if you indicated in your project record that "the project data needs to be mounted on HPC facilities." Note: These storage codes start with a "Q" (e.g. Q0453).

On HPC systems, your Collection will be available under “/QRISdata”, with the Collection name (Q-number) being the name of the folder, e.g. /QRISdata/Q0578. For SSH access to your Collection, you can access it at “data.qriscloud.org.au”, with the Collection available under “/data”. More information on Collection access is available on the QRIScloud portal at https://www.qriscloud.org.au/support/qriscloud-documentation/93-using-qrisdata-collections.

2. HPC Benefits

Polaris-based HPC systems (Awoonga, Flashlite, Tinaroo), QRIScloud SSH access servers, and the UQ Nextcloud service all have the collections mounted. Other UQ-based MeDiCI caches (by arrangement with the respective cache managers) can also have the collections mounted.

The UQ SMB file servers make collections accessible via the "R:" drive (on the UQ networks), SSH access is available at “data.qriscloud.org.au” (under “/data”).  Should you need high performance SMB access from instruments and large data generating devices from a Windows/Linux workstation or instrument, your collections can be mounted on high performance SMB file servers upon request.  A service representative will contact you to help you with the on-boarding procedures for high performance file serving infrastructure.

More information is available on the QRIScloud portal at https://www.qriscloud.org.au/support/qriscloud-documentation/93-using-qrisdata-collections.

3. Current limitations

These collections are not suitable for projects that contain identifiable human data.

Data that has not been accessed recently may have a brief pause before recalling at full speed.

The default file/inode count limit is 1,000,000. You will need to strictly apply data management practices to not exceed the 1,000,000 inode/file limit. If the limit is reached, it will lock you out of your files. To avoid this issue, please zip files off the share and copy the zip folder back on the share.

Currently for HPC collections, new requests for file counts in excess of 1,000,000 per collection per 1TB of storage will only be granted in exceptional circumstances.

4. Requesting a size or file increase for HPC collections

If you require an increase to your HPC Collection, file numbers or storage size, note that you will be required to provide additional information about the data that you are storing, as well as usage patterns and anticipated growth rates.

Requests will be assessed and approval will depend on availability of storage.