Automated test and measurement is an important area of development across fields from biology to IC manufacturing. "Big science" has evolved into "big data", which necessarily entails "big measurement". In science the problem is amplified by the experimental nature of measurements; the user often does not know a priori what region of parameter space is of interest. Instrumentation which offers immediate feedback may be ill-suited to repetitive, automated measurements.
Mezurit 2 [2] is cross-platform Free Software (GPL) designed to make the setup and operation of experiments common in nanoscience and low-temperature physics quick and easy, while providing mechanisms for complex automation. In a typical scenario, the user begins by defining several virtual channels, each corresponding to a physical quantity present in the device under test, as functions of available input and outputs provided by DAQ boards and GPIB-connected external hardware. Then, the user can control several virtual instruments which provide mechanisms to view and record data, sweep outputs, and configure event triggers. After working out the optimal procedure, the user can then automate everything using the comprehensive Python-based scripting interface.
A brief overview of the history, development, and multi-threaded architecture of Mezurit 2 will be presented, along with a live demonstration using simulated hardware. In addition, ongoing efforts to achieve soft real-time operation (in concert with the PREEMPT_RT patch) will be discussed. The talk will conclude with a discussion of a hypothetical open-source data acquisition platform with the goal of producing similar applications with more flexibility and much less effort, without resorting to proprietary software such as LabVIEW.
Speakers: Brian Standley