larpix-daq 0.2.0rc3

LArPix DAQ system

larpix-daq

LArPix DAQ is the data acquisition system for LArPix. It handles the data flow between the "data boards" and offline storage and includes data monitoring and operator control functionality built on the xylem DAQ framework.

LArPix DAQ consists of a set of scripts which are responsible for individual parts of the DAQ system's functionality, as well as an operator interface API which can be run in an interactive python session or used as a basis for a more sophisticated interactive program. The scripts can be run from the same or from different computers, as long as the IP addresses of the various computers are known.

System states

There are three states the system can be in: READY, RUN, and STOP. The state is controlled through the Operator object using the methods prepare_run (transition to READY), begin_run (transition to START), and end_run (transition to STOP).

• STOP: Default state on startup. All components are not expecting data.
• READY: Components should prepare to receive data. Data may arrive at the component before the instruction to transition to the RUN state (though this is expected to be rare). The component should treat that data as if it were received in the RUN state.
• RUN: Components should expect to receive data. Data should not be produced in any other state.

To mark the start and end of a run in the data flow, the producer.py script produces INFO messages with contents "Beginning run" and "Ending run", respectively.

Operator

The LArPix DAQ Operator module provides the interface into the DAQ core for all DAQ operations.

Operator methods interact with the DAQ core to accomplish the desired behavior. For the simplest interactions, a single request and response exchange occurs, and the result is returned. (TODO!!! unify this interface) For most interactions, there are multiple responses for a single request - e.g. an immediate acknowledgement of receipt and then the eventual result. The methods implementing these interactions return generator iterators rather than values. The way to call these functions usually looks like

o = Operator()
final_responses = []
for response in o.run_routine('example'):
    print(response)
    # interact with response object within loop
# When the loop ends, the last response received is still saved in
# the response object
final_responses.append(response)