Service Specific Assignment

The Demonstration Scenario defines a structure SWAP_Order that must be handed over to each service to be executed. For the Service Specific Assignment, a Device Registry is required, as well as an external assignment agent. The assignment agent will be locally registered in the execution engine for a single service, so that it is considered only there. The remaining assignment steps are accomplished with the default assignment strategy of the execution engine.

To register the local assignment agent, an AssignmentAgent structure is defined in the PFDL for the Gluing service and is handed over to the execution engine as service input argument.

PFDL Definition

Struct SWAP_Order
    order_id:number
    stand:Stand_Segment
    segments: Light_Segment[]
    number_light_segments: number
End

Struct Stand_Segment
            stand_shape:string
            stand_height:number
            stand_id:string
End

Struct Light_Segment
            color: string
            diameter: number
            segment_id:string
End

Struct AssignmentAgent
        agent:string
End

Task productionTask
            manufacture_light_segments
                In
                    SWAP_Order
                    {
                        "order_id":1000,
                        "stand":{
                            "stand_shape":"plate",
                            "stand_height":3,
                            "stand_id": "Default"
                        },
                        "segments":
                        [
                        {
                            "color": "red",
                            "diameter": 5,
                            "segment_id": "Default"
                        },
                        {
                            "color": "green",
                            "diameter": 5,
                            "segment_id": "Default"
                        }
                        ],
                        "number_light_segments": 5
                    }
End

Task manufacture_light_segments
            In
                order: SWAP_Order
            Parallel Loop i To order.number_light_segments
                manufacture_light_segment
                    In
                        order
                    Out
                        order:SWAP_Order
            Loop i To order.number_light_segments
                Gluing
                    In
                        AssignmentAgent
                        {
                            "agent":"opc.tcp://localhost:10000"
                        }
                        order
                    Out
                        order:SWAP_Order
End

Task manufacture_light_segment
            In
                order: SWAP_Order
            GetPartsFromWarehouse
                In
                    order
                Out
                    order: SWAP_Order
            Coating
                In
                    order
                Out
                    order: SWAP_Order
            Out
                order
End

Process Execution

To execute the above-defined PFDL-process, a small python script is required to set up the execution engine and the docker environment:

import time, os, sys
#add directories to the system path
sys.path.append(os.path.join(os.path.dirname(__file__), "../."))
sys.path.append(os.path.join(os.path.dirname(__file__), "../tests/test_helpers/."))
#import the main function
from main import main
#import the docker environment
from util.start_docker_compose import DockerComposeEnvironment

if __name__ == '__main__':
    #configure the required docker environment
    env = DockerComposeEnvironment(["Device_Registry", "Service_Server", "Dashboard", "Assignment_Agent"])
    #start the docker environment
    env.run_docker_compose("./Tutorial/docker-compose.yaml")
    #wait until the server started
    time.sleep(10)
    #run the execution engine
    main()
    #shut down the docker environment
    env.stop_docker_compose()

Here, we first need to add the root directory of the repository to the system path to import the main function of the execution engine. Beside, we can re-use the docker-environment from the test section.

For this Tutorial, the terminal command must be extended with three of the optional arguments. First, a global Device Registry is set with the argument “device_registry_url”=”opc.tcp://localhost:8000”. Besides, each service server registers itself in the Device Registry with the URI from the docker compose environment, so that the URI returned from the Device Registry must ab adjusted, to replace the docker-internal URI with a docker-external URI. Here, the optional argument “custom_url”=”opc.tcp://localhost:” is added, so that each server from the docker-environment can be reached from outside. In addition to that, we will set another argument, to increase the number of default clients for the execution engine’s control interface. Here, the argument “number_default_clients”=5 is provided, so that the execution engine starts with a set of 5 control interface clients. The assignment agent argument “assignment_agent_url”=”opc.tcp://localhost:10000” is not provided, since the assignment should be accomplished with the default behavior. However, it is also possible to define a global assignment agent, which can be locally replaced for single services

The process can then be executed from the command line with:

python Tutorial/service_specific_assignment.py "opc.tcp://localhost:4840" "Tutorial/PFDL/service_specific_assignment.pfdl" "dashboard_host_address"="http://localhost:8080" "device_registry_url"="opc.tcp://localhost:8000" "custom_url"="opc.tcp://localhost:" "number_default_clients"=5

so that we can execute the process and map it on the SWAP-IT Dashboard. In case that the log messages of the execution engine should be captured, the command line argument can be simply adjusted to:

python Tutorial/service_specific_assignment.py "opc.tcp://localhost:4840" "Tutorial/PFDL/service_specific_assignment.pfdl" "dashboard_host_address"="http://localhost:8080" "log_info"=True "device_registry_url"="opc.tcp://localhost:8000" "custom_url"="opc.tcp://localhost:"