Organise and access data at scale with Rucio

What is it?

Built on more than a decade of experience in LHC experiments, Rucio serves the data needs of any modern scientific experiments. Rucio can manage large amounts of data, countless numbers of files, heterogeneous storage systems, globally distributed data centres, with monitoring and analytics.

Rucio allows management of data with expressive statements. You to say what you want, and Rucio will figure out the details of how to do it. For example, three copies of my file on different continents with a backup on tape. You can also automatically remove copies of data after a set period or once its access popularity drops.

While Rucio is extremely scalable, the STFC Rucio Data Management Service is designed for smaller communities, with expected data needs up to tens of Petabytes. The fact that the underlying Rucio infrastructure is managed by STFC, allows communities to easily start using and/or test Rucio with little setup cost.

Requirements to consider to use Rucio

For Rucio to manage your data in this setup, Rucio will need X.509 certificate access, or soon, through EGI Check-in to:

  • Your experiments Storage Element that is X.509 capable
  • Your experiments Tape Archive that is X.509 capable
  • Your experiments VOMS server information to check Users credentials against

Rucio is a system that sits on top of already established storage elements to unify users access for data management, and retrieval. Rucio consists of a database of the storage element’s details, users and their access credential information, access levels, the data, and its location, Rucio is not a direct data storage solution.

Rucio Use Cases

Rucio is a data management software, that integrates with your experiments currently provisioned storage. This section will highlight some use cases that Rucio will fill for your experiment.

Management of data for archive when no longer used actively

A simple use case for Rucio is to manage data between ‘hot’ storage, made of HDDs or SSD, and ‘cold’ storage made up of tape. When your experiment generates data that data will be accessed much more than older data as your colleagues work with the data. Within Rucio the data will be registered to be on the ‘hot’ and ‘cold’ storage. This ensures the integrity of the data by providing multiple copies of the data, one on the slower tape archive, and one copy on the more easy to access HDDs and SSDs. Then as the usage of this data declines, the data on the ‘hot’ storage can be removed to make way for newer data that is more frequently accessed. Should the archived data be requested again Rucio can stage the data from tape back to disk making it available for users.

Management of data between sites for jobs

Another useful use case for Rucio is to manage the data between different sites within your experiment. This can provide users with better access to the data that they want to work on. Another option is to have Rucio integrated with your workflow management software (Panda and DIRAC both have integrated with Rucio), so data can be moved to sites as job slots are available, streamlining the data flow for the user.

Official Rucio Pages

Multi-VO specific pages

Next topics:
Getting Started with Rucio

How to get started with Rucio

Rucio Command-Line Interface

The most common Rucio commands

Administering Rucio

Help Rucio admins understand and perform actions for their VO

Dteam Specific Documentation

How to get set up with the dteam VO