Summary Digital Preservation at Bodleian Library Oxford University cdn.nationalarchives.gov.uk
1,651 words - PDF document - View PDF document
One Line
The Bodleian Library at Oxford University has developed a private cloud for storing its digital collections and is actively seeking partners for data preservation, with plans to scale up to Petabyte capacity.
Slides
Slide Presentation (10 slides)
Key Points
- The Bodleian Library at the University of Oxford provides a 'private cloud' local infrastructure for its digital collections.
- The library holds a wide range of archives and special collections, including hybrid collections of analogue and digital material.
- The library offers archival storage for deposits, while the Oxford e-Research Centre offers storage with access to the academic Data Grid.
- Digital preservation is of major importance to the library, and they are actively looking for partners for a consortium approach.
- The library uses various software and tools for preservation, including FEDORA, PRONOM, DROID, and Forensic Toolkit.
- The main issues in a business case for cloud storage for digital preservation are the cost and risk profiles.
- Archiving and digital preservation is often less developed than people think, and using external providers can have asymmetric risks.
- The growth in research data will require a significant re-engineering of provision at all levels across the University in the future.
Summaries
26 word summary
The Bodleian Library at Oxford University created a 'private cloud' to store its digital collections, with plans for Petabyte scale archiving. Seeking partners for data preservation.
77 word summary
The Bodleian Library at Oxford University has created a 'private cloud' infrastructure to store its digital collections. It currently holds 300 Tb of data and plans to archive data at a Petabyte scale in the coming years. The library manages its own storage and compute for preservation, and is seeking partners for a consortium approach to data preservation. Cost and risk profiles are key considerations, and the growth in research data will require re-engineering at the university.
158 word summary
The Bodleian Library at the University of Oxford has implemented a 'private cloud' local infrastructure for its digital collections, including books, images, multimedia, and research data. The library is involved in various digitization projects, including Google Books and a project with the Vatican. It currently has 300 Tb of digital material, but plans to start archiving data at a Petabyte scale over the next three years. The library manages its own local virtualized storage and compute for preservation, rather than using external cloud providers. They are actively seeking partners for a consortium approach to preserve data in geographically distant locations. The library uses FEDORA, PRONOM, and DROID for their technical infrastructure. Cost and risk profiles are the main concerns in a business case for cloud storage, and the library believes that a shared private cloud or consortium approach may be more viable for archives. The growth in research data will require significant re-engineering across the University of Oxford.
403 word summary
The Bodleian Library at the University of Oxford has implemented a 'private cloud' local infrastructure for its digital collections, including digitized books, images, multimedia, research data, and catalogs. The library has been a legal deposit library for 400 years and holds a wide range of archives and special collections. It is involved in various digitization projects, including Google Books and a project with the Vatican to digitize maps. The library also supports researchers by offering services for managing their research data.
The library, Oxford e-Research Centre, and central IT service all offer their own storage services, which have been described as 'private cloud'. The library offers archival storage for deposits, the IT service offers conventional storage for research data, and the e-Research Centre offers storage with access to the academic Data Grid. Digital preservation is of major importance to the Bodleian Library. They are required to comply with Research Council mandates for retention of data, typically for ten years.
The library currently has 300 Tb of digital material, but over the next three years, they plan to start archiving data at a Petabyte scale. The content is mostly images and text, but they are expanding to include AV and other file formats over time.
Oxford manages its own local virtualized storage and compute for preservation, which they refer to as private cloud. They are not procuring cloud storage from an external service because most external cloud providers do not match their funding model and they are concerned about the costs of retrieving data. They are actively looking for partners for a consortium approach and preservation advantages in having a copy of the data stored in a geographically distant location.
The library uses FEDORA, home-grown software, PRONOM, and DROID for their technical infrastructure. The private cloud is based on VMware ESX running across multiple sites with a mix of disk and tape storage tiers.
The main issues in a business case for cloud storage for digital preservation are the cost and risk profiles. The library needs the ability to pay up front for perpetual and/or 10+ year storage of research data to fit in with fixed term research project grant funding. They believe that using a shared private cloud or a consortium of like-minded institutions might be more viable for archives.
In the future, the growth in research data will require a significant re-engineering of provision at all levels across the University of Oxford.
543 word summary
The Bodleian Library at the University of Oxford has implemented a 'private cloud' local infrastructure for its digital collections, including digitized books, images, multimedia, research data, and catalogs. The library has been a legal deposit library for 400 years and holds a wide range of archives and special collections. It is involved in various digitization projects, including Google Books and a project with the Vatican to digitize maps. The library also supports researchers by offering services for managing their research data.
The library, Oxford e-Research Centre, and central IT service all offer their own storage services, which have been described as 'private cloud'. The library offers archival storage for deposits, the IT service offers conventional storage for research data, and the e-Research Centre offers storage with access to the academic Data Grid. The library provides a quote for archiving digital material generated from a research grant, and for larger or more complex datasets, they can collaborate with other institutions.
Digital preservation is of major importance to the Bodleian Library. They are required to comply with Research Council mandates for retention of data, typically for ten years. The library is working mostly with humanities for digitized materials, but born digital research datasets are coming from different disciplines such as sciences and social sciences, which require new skills in the library. Over the next 4-5 years, the largest collections will be born digital rather than digitized.
The library currently has 300 Tb of digital material, but over the next three years, they plan to start archiving data at a Petabyte scale. The content is mostly images and text, but they are expanding to include AV and other file formats over time.
Oxford manages its own local virtualized storage and compute for preservation, which they refer to as private cloud. They are not procuring cloud storage from an external service because most external cloud providers do not match their funding model and they are concerned about the costs of retrieving data. They are actively looking for partners for a consortium approach and preservation advantages in having a copy of the data stored in a geographically distant location.
The library uses FEDORA, home-grown software, PRONOM, and DROID for their technical infrastructure. The private cloud is based on VMware ESX running across multiple sites with a mix of disk and tape storage tiers.
The main issues in a business case for cloud storage for digital preservation are the cost and risk profiles. The library needs the ability to pay up front for perpetual and/or 10+ year storage of research data to fit in with fixed term research project grant funding. They feel there can be asymmetric risk in using an external provider, and they are concerned about the longevity of commercial providers.
The library has learned that archiving and digital preservation is often less developed than people think. They have found that the Open Archival Information System (OAIS) Reference Model is a mixed blessing and that working with very large data volumes is not simple. They believe that using a shared private cloud or a consortium of like-minded institutions might be more viable for archives.
In the future, the growth in research data will require a significant re-engineering of provision at all levels across the University of Oxford.