Summary

GT.M is the database of record and “NoSQL” application development & deployment platform at some of the largest real-time core-banking and electronic health record systems deployments around the world. With its source code available under a free / open source software (FOSS) license, GT.M has been downloaded over 135,000 times from its primary site, is bundled with selected applications, and can be installed on contemporary Debian/Ubuntu systems with sudo apt-get install fis-gtm.

Although GT.M has been free to hack¹ for the entire twenty-first century, its complexity, which arises from its breadth of functionality, as well as its uncompromising robustness, performance and security, is an obstacle. The workshop leader, with decades of experience developing GT.M, gets you started.

Hacking GT.M is a very hands-on course. The emphasis is on learning by doing under the guidance of a workshop leader, rather than learning by being taught by an instructor.

Target Audience

Experienced C programmers will benefit from the Hacking GT.M workshop. An understanding of common tools such as makefiles, C compilers, and debuggers is required; knowledge of specific tools used to develop GT.M – cmake, gcc, gdb, ld – would be helpful. Expertise in shell scripting is required; knowledge of the POSIX API would be helpful. While specific experience with GT.M is not required, exposure to GT.M would be helpful to understand why things are the way they are.

Topics

The workshop leader will lead discussions on the following topics.

• The directory structure of a GT.M source code distribution.

• Compiling GT.M, including needed tools.

• The internal structure and organization of GT.M (API layers and relationships).

• Working with the FIS GT.M team, including: creating specifications and release notes; regression testing and tests suitable for incorporation into the GT.M automated regression test suite; submitting your changes; coding standards; and more.