Zebra is great of bibliographic data (Score 1)

The data could easily be converted into MARC bibliographic records and indexed with Zebra. You could then use zebra has a stand-alone Z29.50 server, or run Koha on top for easy searching. Zebra can search millions of records in seconds, so it would be ideal, considering this is essentially bibliographic data. I am a public library IT guy, and would seriously be willing to help out if can use me. Just send me a email ( kyle DOT m DOT hall AT gmail.com ). I can take the raw data and convert it to a bulk marc record set for you, and probably even offer alternative hosting if you'd like.