 |
 |
I can't say that we conducted any formal comparison or analysis, but the primary reasons HathiTrust has kept the archival data we steward out of the cloud are around the sensitivity of the content (about 85% copyrighted). One aspect of this is avoiding potential legal liabilities with accidental exposure of the content. Contract terms with Google also prohibit distribution of materials to any party that is not a higher education institution, so the use of cloud storage would need a guiding legal opinion. That issue would be further exacerbated by the stringent data security terms of the Google settlement, should it be approved. For example, we might easily find ourselves painted into a corner if we were unable to have the cloud storage provider meet those terms (which we almost certainly wouldn't, because they require eg 5-year retention of audit logs for ANY operation against the data--uh-huh.) Handling content that is the subject of a lawsuit of such magnitude tends to lead to conservative legal positioning. :) On 01/24/2011 09:59 AM, Bryan Beecher wrote: > We've been using a few different "cloudy" technologies at ICPSR. I wrote up > a short blog post in November that might be the sort of thing you'd find > useful: > > http://techaticpsr.blogspot.com/2010/11/cloud-and-archival-storage.html > > Our storage needs are relatively small compared to the guys below (< 10TB), > so what we've chosen to do so far might not be a good fit for an > organization that has lots and lots of content (say,> 100TB). > > -- bryan > > On Mon, Jan 24, 2011 at 9:32 AM, Michael J. Giarlo<[log in to unmask]> wrote: > >> I apologize for not saying this on the last teleconference -- my (obv. >> broken) phone was stuck on mute so I could only listen -- but I would also >> be interested in hearing about how other folks on NDSA Infrastructure are >> dealing with archival storage. >> >> It will be useful to hear about particular cloud/grid solutions such as >> Duracloud, LOCKSS, and iRODS, and just as useful to hear about how folks not >> using one of these solutions are accommodating their archival needs at the >> storage layer and why they chose *not* to go the cloud/grid direction. >> >> I have folks like, e.g., HathiTrust and CDL in mind. >> >> -Mike >>