Today I went to the iRODS user conference in Chapel Hill, NC. iRODS stands for Integrated Rule Oriented Data System. the system is intended to be used as a data and digital object curation system that supports data curation activities (e.g. preservation activities) and access activite (e.g. create, access, replace, update, version). iRODS is usable as a subsystem for Dspace and Fedora as well as a number of other products and is flexible enough to serve as a platform for other file system services (e.g. Dropbox).
The morning session focused on new features in version 2.5. As I was not familiar with the system I spent the time stepping through a tutorial on the iRODS website. The afternoon focused on use cases which were rather interesting.
The first use case focused on continuous integration using maven/git/hudson which are some tools we have been discussing internally here at ZSR. The development environment at RENCI is called GForge & appears to be a solid approach to enabling distributed development of services. We heard from DataDirect Networks that discussed how they are using iRODS to change how get away from file system work. I also got to see some demonstrations of use cases in Archives and file management systems. Lots more information about iRODS and the conference is available on the iRODS wiki.
One fun project I learned about. National Climatic Data Center. For me the really incredible idea was that an iRODS approach to file system storage combined with lots of good metadata would enable a distributed (multi-institution, multi-site) structure that researchers could use to do retrievals and run processes based on common metadata elements or other file identifiers – for more info see the TUCASI Project.