The Dakota team is committed to supporting users by providing documentation, examples, direct interaction through our mailing list, and fostering community engagement.
When seeking help with Dakota, consider these support mechanisms:
The Dakota team maintains extensive documentation, which is updated with every release.
- Review the Dakota Users' Manual for guidance on methods and input file structure.
- Additional mathematical/implementation detail for some methods is included in the Theory Manual.
- The recently revamped Dakota Reference Manual lists and describes every keyword in the Dakota input specification in a helpful hierarchical format.
- Consult Frequently Asked Questions to resolve some commonly encountered difficulties.
- Information on integrating Dakota as a library is found in the Developer's manual
Numerous examples are provided with Dakota to demonstrate its capabilities.
- The examples folder in your Dakota installation provides categorized Dakota input files and interface scripts for many use cases.
- The test folder, which contains our entire regression test suite, is also a great source of input files demonstrating both simple and advanced Dakota features. Each *.in file can contain multiple tests, with lines corresponding to each test denoted by #N or #pN. Use the script dakota_test.perl (also located in the test folder) to extract individual tests by number. E.g. to extract test 10 from the test file dakota_sbo_rosen.in, which demonstrates surrogate-based local optimization:
perl dakota_test.perl -e 3 dakota_sbo_rosen.in
The file dakota_sbo_rosen.in_ (original test file name + an appended underscore) is produced. Add the --parallel (-p) option to extract a parallel test or run with --help for a full list of options.
Public User Community
Dakota is an open source and publicly available project with a significant worldwide user base. Our vision is that Dakota users will increasingly help each other with questions and issues. We encourage general usage questions about how to use Dakota, including how to install, choose algorithms, or interface to an application code, be posted to our public users' email list. Simply email your question to firstname.lastname@example.org. Your first message will be held for moderation, and after it is approved, you will be subscribed to the list. Please refer to the guidance in our FAQ when preparing your request. Also, note that while we make every attempt to respond to all support inquiries, funding and staffing limitations demand that we prioritize requests as follows:
- Sandia National Laboratories users and integrators
- CRADA and other DOE users and partners, e.g., NNSA/ASC labs, NEAMS and CASL Nuclear Energy initiatives
- Close research or software collaborators
- Other users
Internal Sandia User Community
After logging in using your Sandia credentials, a "Dakota @ SNL" link will appear on the main menu. Follow it to submit a support request.