The Syracuse University Libraries’ Digital Library stewards locally produced digital content by coordinating and managing the technical activities and infrastructure that produce, describe, manage, and preserve this content. It also supports research, teaching, and learning by facilitating access to content from program participants and partners while encouraging creative engagement and use.

Our Infrastructure

Tech Stack Infrastructure Diagram

The above diagram updated March 2024 is an abstracted model of the digital library infrastructure employed at Syracuse University Libraries. As such, it demonstrates the digital activities and workflows in a general sense, but may not account for specific use cases and collections.

The diagram shows activities happening in five vertical blocks and an additional horizontal block crossing the bottom of the image.

Production

Beginning at the center, the model identifies two areas of digital library content production. First, it acknowledges the image production that takes place in the Libraries Digital Production Unit; audio production that takes place in Belfer Labs; and other, multi-media content that may be digitized by vendors or our content stewards. In addition, the production block includes content added to the institutional repository (SURFACE, more below), including electronic theses and dissertations (imported from ProQuest) and other scholarly content created by faculty, staff, and students. Finally, web archive files or WARCs are identified as a production area, though this activity does not take place within the Libraries, it is simply managed and preserved by a third-party tool: Archive-it. Note that web-archiving is still in a very nascent stage at the Libraries.

Management

Moving to the right, the model identifies the different systems the Libraries use to manage its digital assets. Quartex is the primary digital asset manager. It stores all the descriptive, administrative, and technical metadata (excluding certain metadata recorded only for digital preservation or metadata stored in the files themselves). It also stores access files for all the Libraries digitized objects and drives access to published collections. Content specialists also use this platform to access unpublished, digitized materials for reference and teaching purposes.

The Libraries also manage a customized instance of Kaltura for its vast audio-visual access files, though these are run through Quartex for end-user access.

Digital Commons powers the institutional repository and open scholarship services at the Libraries.

Access

Moving to the block on the far right, the model shows how end-user access is accomplished in the digital library infrastructure. Most published collections managed in Quartex and Kaltura are made available through the Syracuse University Libraries Digital Collections portal, though we manage several bespoke websites as well, including: the IVMF Digital Library, the Marcel Breuer Digital Archive, the Ted Koppel Digital Collection, and Our Stories. Our institutional repository, SURFACE, provides access to digital objects created in the course of scholarly production on the Syracuse University campus.

Preservation

Returning to center and moving left from the production block, the model outlines digital preservation activity taking place in two distinct spheres. First, the Libraries uses Preservica for its internally managed digital preservation activity and master/primary file management. Objects stored in the institutional repository are preserved in a dedicated LOCKSS environment, managed by the vendor.

Storage

Moving to the block on the far left, the model identifies the current storage environment for archival primary files. Most are staged and stored temporarily on hard drives or network drives before being transferred into Preservica. Preservica, as the arrows demonstrate, then transfers files back into an Amazon Web Services storage environment, using a combination of Glacier (for dark storage) and S3 for assets retrieved on a regular basis. Currently, the Libraries has more than 200 terabytes of content and counting. We are still investigating the various possibilities for external redundancies for archival primary file storage.