Staging Archivematica AIPs for syncing to DuraCloud

[courtne...@gmail.com]

Hi folks! I hope to see you at the A-TEX Office Hour on Friday the 13th: https://www.tdl.org/event/a-tex-office-hours-aug2021/

But, until then, I have a question from some TDL members about staging AIPs to sync into TDL DuraCloud. Currently, their process is as follows:

• Navigate to Archivematica dashboard AIP Storage tab and find AIP(s) to download
• Download AIPs to local machine
• move AIPs to Sync Folder (the folder the Sync Tool has access to)
• Sync to DuraCloud Space
• Repeat until space is ready to snapshot into Chronopolis

The issue is that the download of even 3-5gb can take more than a day.

So, what are you all doing to select AIPs to send to your DuraCloud spaces?

Thanks in advance,

Courtney

Courtney C. Mumma, MAS/MLIS she/her(s) pronounce

Deputy Director, Texas Digital Library

 Email: c.m...@austin.utexas.edu

Web: http://www.tdl.org/

Twitter: @TxDigLibrary

Facebook: http://www.facebook.com/texasdigitallibrary

Join the e-mail list: http://tdl.org/news/newsletters/newsletter-signup/

[Sean Buckner]

Howdy, folks

At Texas A&M, we use both Chronopolis and Glacier for our content. We have two local storage spaces where all AIPs are stored; Tier 1 content goes through Archivematica to one location, Tier 2 goes to another (I switch as needed before transfer/ingest by changing the "Store AIP location:" in the Processing Configuration of the Administration tab). Tier 2 content goes to a storage location that has a sync job set up with our DuraCloud space for Glacier and that job runs daily, so any new AIP in that storage location automatically syncs to the Glacier DuraCloud space. For Tier 1 content destined for Chronopolis, the new AIPs need to be batched, of course. So, to avoid the "download AIPs to local machine" issue, I indicate (to my IT person) the UUIDs of the AIPs that I need copied over to a separate sync folder, and the AIPs get copied over from their storage location to that folder. The AIPs copy over - without the onerous "download" process - and then get synced directly to the Chronopolis DuraCloud space for follow-on snapshot when ready.

From the procedures described above, it would seem to me that there needs to be a back-end process to identify the specific UUIDs and copy (or move) them from that "Store AIP location" directly to the sync folder. It sounds like the download via the Archivematica interface is the obstacle, and it essentially is an additional copy process that can be avoided.

If I've misunderstood the procedures or my particular processes won't work in the above mentioned environment, then my apologies. Hopefully this helped.

Sean Buckner, MSI

Clinical Associate Professor  |  Coordinator of Digital Preservation & Digitization

Preservation Unit  |  University Libraries  |  Texas A&M University

5000 TAMU  |  College Station, TX 77843-5000

ph: 979.845.1924  |  sbuc...@library.tamu.edu

- - - - - - - - - - - - - - - - - - - - - - - - 
TEXAS A&M UNIVERSITY |  www.tamu.edu