Object Store

Online storage for large amounts of data

Object Store is an online storage service for storing large amounts of research data. This service is ideal for storing different types of data. Your storage space can grow indefinitely, while keeping your data quickly and easily accessible.
Jonge onderzoeker in laboratorium achter computer

Advantages

Scalable, sustainable and available

Ideal for storing different types of data that can reach very large volumes.

Custom solutions

We can support applications of personalised solutions adapted to local settings.

Do you have a question about Object Store? Get in touch.

Illustration of connections

Servicedesk Data & Computing Services

Telefoonnummer

Object Store is especially suitable for:

  • scale-out storage
  • storing large amounts of static data (many files)
  • streaming media (audio/video)
  • files that require linked metadata (e.g. for policy-driven data management)
  • backup, storage and archiving of data
  • a national storage facility for repositories of research publications (e.g. figshare, Dataverse)
  • online data storage within an iRODS application, allowing geographically dispersed data to be made available as a single entity.

How does object storage work?

Object storage infographic

Our Object Store service is a distributed architecture that manages data as objects. This makes expansion easy. Traditional storage systems often organise data in a certain hierarchy of folders and files. Object storage, on the other hand, has a very flat structure of containers and the objects they contain. The advantage of this is that object storage systems can scale up much better than file-based storage systems.

Access protocols

Object Store is based on Ceph RGW. In addition to its proprietary protocol, it also provides the general S3 protocol. The S3 REST API was developed by Amazon to provide access to their cloud storage system. As a result, many features and tools are available for this type of storage system. Data processing software Apache Spark and Elastic Search can communicate with S3. On the client side, tools are also available. Cyberduck provides graphical user interfaces for processing and managing data. Rclone is an rsync-like synchronisation tool for cloud storage systems.

Archiving and backup

You can also archive your data in our Data Archive. The Object Store provides a fast network connection directly to this archive. SURF does not automatically back up data located in the Object Store. Data are stored multiple times, significantly reducing the risk of loss. Nevertheless, you remain personally responsible for backing up your files.

Support and advice

As a user of SURF's Object Store, you can always ask us for help. Our helpdesk can be reached via the service desk portal.

Do you need specific advice on how best to use Object Store for your research? Then contact our consultancy service.

More (technical) information about this service and how to use it can be found in the Object Store documentation.