Storing Data¶
Available storage¶
Summary
By default, each user on Savio is entitled to a 50 GB home directory that receives regular backups; in addition, each Faculty Computing Allowance-using research group can request 50 GB of project space and each Condo-using research group can request 200 GB of project space to hold research specific application software shared among the group's users. All users also have access to the large Savio high performance scratch filesystem for working with non-persistent data.
| Name | Location | Quota | Backup | Allocation | Description |
|---|---|---|---|---|---|
| HOME | /global/home/users/ | 50 GB | Yes | Per User | HOME directory for permanent code, executables, small data |
| GROUP | /global/home/groups/ | 50/200 GB | No | Per Group | GROUP directory, must be requested, for shared code, executables (50 GB for FCA, 200 GB for Condo) |
| SCRATCH | /global/scratch/users | none | No | Per User | SCRATCH directory with Lustre FS for data for active compute. See below for details of scratch usage limits/fair usage. |
Group Directory Access
Group directories are not created by default. Group directories must be requested by contacting brc-hpc-help@berkeley.edu. See the documentation on group directories here.
For information on making your files accessible to other users (in particular members of your group), see these instructions.
Checking your disk usage¶
To see your overall usage in your home directory (base filepath /global/home/users) in relation to your quota, you can run either of these commands:
check_quota.sh
check_quota.sh -u ${USER}
The same command can be used to check a project or group storage in /global/home/groups:
check_quota.sh -p [PROJECT_NAME]
To see your own overall scratch usage in bytes and number of files (base filepath /global/scratch, so including any condo storage you have access to):
grep ${USER} /global/scratch/scratch-usage.txt
If you have access to condo storage, you can see its available and used space as follows:
cd /global/scratch/projects/
./quota.sh
As of fall 2026, we are piloting regular semi-weekly reports that give detailed information on usage by user and by condo project group within /global/scratch. Look for the most recent date, following the example below:
cd /global/home/users/starfish_reports/2026-10-20/usage/all-filesystems # For user-specific reports as of Oct. 20, 2026
cd /global/home/users/starfish_reports/2026-10-20/usage/summaries # For condo project-specific reports
In general it's best to use the above reports, but if need be (see the cautionary note below), you can use the du command to determine what is taking up a lot of space. Here are some examples:
du -h path_to_subdirectory # usage by file in a subdirectory
du -h -d 1 # summary of usage within each top-level subdirectory
Use du sparingly
The du command places a heavy load on the filesystem. Please use it sparingly, including using on specific subdirectories, rather than using it for your entire home or (most importantly) scratch directory. Also consider running once for a given subdirectory and saving the result rather than running repeatedly.
Scratch storage¶
Every Savio user has scratch space located at /global/scratch/users/<username>. Scratch is a 13 PB resource shared among all users.
To preserve this space for effective use by all users, we have generally asked users to keep their usage to reasonable levels (not more than 12 TB except for short bursts of usage during active computation) and to delete data no longer needed for active computation.
New quotas enforcing short-term heavy usage (as of November 2026)
We are now rolling out a new more automated/universal approach to managing scratch space, with the goal of providing fair and reliable access to all users. We will start with those users currently using the most space before including all users.
- Users will be allowed to exceed a "soft quota" of 12 TB of use in /global/scratch/users for at most 8 weeks, to allow for active project work needing large amounts of space.
- Similarly, users will be allowed to exceed a count of 1 million files and directories for at most 8 weeks.
- If a user is above 12 TB (or 1 million files) continually for 8 weeks, they will no longer be able to write additional data until they bring their usage under 12 TB (and under 1 million files).
- If, after 8 weeks, a user has an exceptional need for an extension, they can request an extension for another 8 weeks via this form that documents their need and their plan to reduce usage by the 16th week.
- Users will not be able to exceed 100 TB or 30 million files for any period of time, unless they request an exemption via this form with details on their needs (space and time period).
- Users can check their usage themselves, and we will automatically email users above the 12 TB / 1 million file count soft quota on a semi-weekly basis.
- Condo storage usage (
/global/scratch/projectsand/global/scratch/p2p3) is not subject to these limitations.
Group scratch directories are not provided. Users who would like to share materials in their scratch directory with other users can set UNIX permissions to allow access to their directories/files.
Savio condo storage¶
Condo storage is additional persistence storage available for purchase at a competitive price.
As of summer 2021, Condo storage directories are on the same filesystem as Savio's scratch storage. So for users accessing condo storage directories found in /global/scratch/projects, you do not need to (and should not) stage your data to your individual scratch directory. For condo users accessing condo storage directories in /clusterfs, you should still stage your data to your individual scratch directory when appropriate.
You can see the available and used space in the condo as follows:
cd /global/scratch/projects/
./quota.sh
More storage options¶
If you need additional storage during the active phase of your research, such as longer-term storage to augment Savio's temporary Scratch storage, or off-premises storage for backing up data, the consultants at the Research Data Management Program can help you identify suitable options.
Assistance with research data management¶
The campus's Research Data Management (RDM) service offers consulting on managing your research data, which includes the design or improvement of data transfer workflows, selection of storage solutions, and a great deal more. This service is available at no cost to campus researchers. To get started with a consult, please contact RDM Consulting.
In addition, you can find both high level guidance on research data management topics and deeper dives into specific subjects on RDM's website. (Visit the links in the left-hand sidebar to further explore each of the site's main topics.)
CGRL storage¶
The following storage systems are available to CGRL users. For running jobs, compute nodes within a cluster can only directly access the storage as listed below. The DTN can be used to transfer data between the locations accessible to only one cluster or the other, as detailed in the previous section.
| Name | Cluster | Location | Quota | Backup | Allocation | Description |
|---|---|---|---|---|---|---|
| Home | Both | /global/home/users/$USER |
30 GB | Yes | Per User | Home directory ($HOME) for permanent data |
| Scratch | Rosalind (Savio) | /global/scratch/users/$USER |
none | No | Per User | Short-term, large-scale Lustre storage for very high-performance computing |
| Condo User | Rosalind (Savio) | /clusterfs/rosalind/users/$USER |
none | No | Per User | Long-term, large-scale user storage |
| Condo Group | Rosalind (Savio) | /clusterfs/rosalind/groups/ |
none | No | Per Group | Long-term, large-scale group-shared storage |