Longterm Preservation of Scientific Data
The ITMZ provides two different storage technologies designed and intended for digital longterm preservation of scientific data. The ITMZ only provides the service as bitstream preservation. In addition depending on the chosen technologie some metadata may be added to the stored digital artefacts (files/objects).
Blockstorage based Object Storage
This object storage system is physical high redundant and distributed over three locations. Access is provided via an Amazon S3 compatible REST-API. The API may be used direct using a user-developed software or using compatible third party software. The intended field of use is for longterm archives which require live access on the preserved data. Files represented as objects in this case should only be written once and not modified again. Those files may be read in relative high frequency. A usecase for example is a online repository of the scientific data accessed by different users.
Tape based Object Storage
This object storage is based on a tape library as physical media for the bitstream. Access is provided via an Amazon S3 compatible REST-API. The API may be accessed direct using a user-developed software or using compatible third party software. The intended field of application is for preserving scientific data that does not need to be frequently accessed. Beeing aware of the access latency due to the process of loading, moving and copy of the data to a cache disk the data can be accessed at any time. Additionally it is possible to remove tape media physically and store it at a different location. The data is stored in LTFS on the tapes. Intended field of application are data repositories where the data is accessed never or at least less frequent but has to be preserved for a long period. Due to the possibilitie of removing tapes an export of a copy is possible as well. Additional application may be the requirement of moving data to different scientific facilities and import them there.
Access to Storage and Implementation of Archive Projects
There are some pitfalls which should be avoided if realizing a larger archiving project. Therefore we advise to make use of the extensive support provided by the library. Afterwards there are plenty of options and scenarious to implement using the provided storage infrastructure. A complete list of all variants and possibilities is too large for this page. Nevertheless some examples are provided:
- realizing a 3-2-1 archive
- using automatic data migration from block to tape storage
- implementing archiving workflows using a push or pull strategie
Feel free to get in contact with us using the provided data on the right hand side.
In every circumstances a Data Management Plan must be created. Here as well the library will support you.
Additional valuable information regarding management of scientific data can be found here (in German): https://www.forschungsdaten.org
Regarding archiving (partly in English):https://www.langzeitarchivierung.de
Tel.: +49 (0) 381 498 - 5325