We will offer a storage backend using CEPH technology. 6 Terra bytes of available storage space will server public and private data repository’s. Block storage can as well be mounted into instances for extra capacity. Once stored on block storage you will be able to clone disks or take snapshots of work in progress. All data stored on the CEPH backend will be mirrored, meaning we can loose half of the drives and no data will be lost.
Shared public datasets.
As there are already enormous amounts of data available to the general public most of these can be made available on request as a repository. Tell us what dataset you need by sending in a small whitepaper explaining what workload you will execute on it. If your request gets approved we will download and make it accessible on the public repository within your OpenStack project.
Private datasets.
Most users will work on private datasets. So there will be several possibility’s to upload, you will get a private repository within your OpenStack project. This can be a networked repository or data stored on block storage disks.
Generated data.
There can be a lot of extra or temporary data created while doing research, this can also be stored safely in the backend. Same principles for access, sharing and cloning of your generated datasets.