Hi Jan,
I would like to introduce Hadoop which (at least in my opinion) is a perfect solution for ingesting and processing large amount of data. (NASA build a solution on top of Hadoop [1])
Hadoop in a nutshell: Java-based storage and analytics framework for structured and unstructured data on commodity hardware. Hadoop is designed as fail-save, long term storage and processing framework based on papers about Googles distributed filesystem(GFS) and processing (MapReduce).
As far as i know you use OpenArchive for tape storage. I guess thats not needed with Hadoop because of it's low TcO.
What do you think about this?