Archive dataset type¶
What is the new dataset type for?¶
Since September 2024, openBIS instances at Empa offer a new dataset type archive. The archive dataset type implements a direct archiving policy: any data uploaded as dataset with the archive type will be directly archived instead of being stored on the DSS (data store server) storage of openBIS. This means that datasets of the type archive cannot be directly accessed from the ELN UI anymore but must be unarchived first. This dataset type can be used for large data that does not need to be readily available and needs to be directly stored on the Strongbox at ETH.
Important to know¶
As for the regular archiving feature in openBIS, the minimum data bundle size for archival is 10 GB. If you try to archive a smaller dataset, it will be queued for archival but only stored when more datasets in the same space or project or collection/experiment are selected for archival, either through the choice of the archive dataset type or through a manual archiving request through the archive helper in the ELN UI.
When to use archive type¶
Use the archive dataset type for any dataset that needs long-term archival but does not need to be readily available.