Migration to EPCCfs storage¶
Important
This information was last updated on 26 Aug 2026.
This section of the documentation covers the process of moving from the current
storage for compute jobs (/work file system) to new storage (/epccfs file
system).
The current /work storage is coming to its end of life so on Wed 16 Sep 2026 we
are moving to new storage mounted on Cirrus login and compute nodes as /epccfs.
Migration process¶
There will be a full maintenance session (Wed 16 Sep 2026) with no jobs running on compute nodes for this switch. During this maintenance session, the following high level steps will be followed:
- All jobs Pending in the queue will be deleted
- This step is necessary as jobs in the Pending queue before the switch will
expect to be able to write to
/work. As this will not be possible after the switch, any pending jobs would fail. - Current
/workstorage will be changed to read-only mode /epccfsstorage will be made available in read/write-mode- Final testing of
/epccfs - Service returned to users
The current /work storage will remain available in read-only mode until at least
21 Nov 2026 for users to copy over any data they wish to access/keep on the
new /epccfs storage.
Impacts for users¶
- All current Cirrus projects and user accounts will get directories on the new storage
with quotas set as they are on the current
/workstorage. - Any jobs in Pending state at the start of the maintenance session for the switch
will be deleted. You must resubmit them (using the new
/epccfsstorage) once the system returns from maintenance. - Any jobs submitted after the maintenance session must write to locations in
/epccfs. Any jobs that attempt to write to locations in/workwill fail. - Following the maintenance session, you will no longer be able to write data to
any locations in
/work. This data will remain available in read-only mode until at least 21 Nov 2026. - Users should copy any data they wish to keep off
/workbefore 21 Nov 2026.
EPCCfs storage¶
Note
This is a copy of the documentation in the Data section of the Cirrus User Guide.
Every project has an allocation on the EPCCfs storage and your
project's space can always be accessed via the path
/epccfs/[project-code]. The EPCCfs storage provides a large
capacity (more than 30 PB) and is currently implemented using
the VAST technology.
Warning
EPCCfs is not backed up at all.
You can find your directory on the EPCCfs at:
For example, if your username is auser and you are in the e05 project, then
your EPCCfs directory will be at:
Copying data to /epccfs from /work file systems¶
You can use the standard Linux cp command to copy data from other Cirrus file
systems to EPCCfs or vice versa. For example, to
transfer the file important-data.tar.gz from the /work file system to
/epccfs you would use the following command (assuming you are user auser
in project e05):
(remember to replace the project code and username with your own username and project code).
Use rclone parallel local data transfers for large datasets
If you are transferring a large amount of data to EPCCfs, you should consider
using rclone local data transfer (perhaps in a
serial job submission script) rather than using the basic cp command.
Quotas on EPCCfs¶
As for the other Cirrus storage systems, all projects are assigned a quota on EPCCfs. The project PI or manager can split this quota up between users or groups of users if they wish.
You can view any EPCCfs quotas that apply to your account by logging into SAFE and navigating to the page for your Cirrus login account.
- Log into SAFE
- Use the "Login accounts" menu and select your Cirrus login account
- The "Login account details" table lists any user or group quotas that are linked with your account. (If there is no quota shown for a row then you have an unlimited quota for that item, but you may still may be limited by another quota.)
Tip
Quota and usage data on SAFE is updated twice daily so may not be exactly up to date with the situation on the systems themselves.
You can also query quotas that apply to your current EPCCfs directory from the
the command line using the df -h $PWD command, for example: