System Details


Arctos is a multi-tenant database using expansions of a data model developed by the University of California’s Museum of Vertebrate Zoology in the early 1990s. It operates in an arena where tools and standards are evolving rapidly, and where savvy curators and collections staff are a continuous source of good ideas.  Thus, it is a dynamic system under continuous development.

Hosting

Arctos data and applications are hosted on the Corral system at the Texas Advanced Computing Center (TACC), whose mission is to enable discoveries that advance science and society through the application of advanced computing technologies. TACC provides Arctos with secure off-site storage, OCR processing, media storage and processing, and other services. This arrangement alleviates the need for collections personnel to administer servers, patch operating systems, deal with failed drives, or develop backup strategies. 

Oracle Database

The guts of Arctos consist of an Oracle Relational Database Management System (RDBMS). Along with holding tables of data, the Oracle RDBMS maintains referential integrity, enforces controlled vocabularies, provides a secondary security and user management layer, and separates data into Virtual Private Databases (VPDs) for the different collections. Oracle’s VPD architecture allows sharing of equipment, expertise, and authority data, while allowing users to maintain exclusive row-level control of collections data.

ColdFusion Platform

Arctos applications are written in Adobe ColdFusion, which provides a bridge between users, applications, and the RDBMS by converting CFML to HTML forms via an Apache webserver, and converting form submissions back to DDL. ColdFusion also provides a method to rapidly develop dynamic webservices. Various JavaScript components and the jQuery library are employed heavily, together with webservices, to improve the user experience without requiring specialized software.

Scheduled tasks and Oracle Jobs perform functions such as emailing reminders to users, maintaining flattened query data, publishing data via the Integrated Publishing Toolkit (IPT),  querying GenBank for Arctos specimens, and creating media for various specialized imaging projects. 

As a centralized web application, clients require a modern browser on a reasonably brisk network connection.  The FireFox browser is recommended for database operators, but all public features and most operator features are supported by most of the popular browsers. There are effectively no restrictions on number of users, or limit to the number of concurrent users.

Arctos code is managed in Github, a fully-featured versioning system that allows roll-backs, separate development threads, etc. Code updates are added frequently by seamless push. Changes to data are logged by Oracle, and continuous backups are stored in two secure, widely-separated locations.

Arctos software and data definition language (DDL) are freely available, and local installations are possible. One such endeavor (MCZBase) uses forked code that is derived from, but  no longer contributing to, Arctos. We can provide no support for such installations. 

More technical resources: