Queue¶

The queue behavior of a given pipe can be customized in case of more complex queueing operations are needed in addition to the simple FIFO buffer. With this purpose, two decorators can be used to determine the push and pop behavior of the queue. The two decorators are @simpype.queue.push and @simpype.queue.pop (see push() and pop() for more details). There are two ways of customizing the behavior of a queue:

Work directly on the simulation scenario and perform an inline customization;

Create a queue model to be included in the simulation scenario through a custom pipe model.

Either approaches are valid, however inline customization is more suited for small customizations while queue model is more suited for larger customizations and code re-usability (you can include the smae model multiple times in different simulations).

Inline customization¶

In this example, a LIFO discipline is implemented.

import simpype
import random

sim = simpype.Simulation(id = 'simple')
gen0 = sim.add_generator(id = 'gen0')
gen0.message.property['priority'] = {
    0: lambda: random.randint(0,1)
}
res0 = sim.add_resource(id = 'res0')
res0.pipe.add_queue(id = 'lifo')
res0.random['service'] = {
    0: lambda: 2.0
}

@simpype.pipe.enqueue(res0.pipe)
def enqueue(self, message):
    return self.queue['lifo'].push(message)

@simpype.pipe.dequeue(res0.pipe)
def dequeue(self):
    return self.queue['lifo'].pop()

@simpype.queue.push(res0.pipe.queue['lifo'])
def push(self, message):
    return self.buffer.append(message)

@simpype.queue.pop(res0.pipe.queue['lifo'])
def pop(self):
    return self.buffer.pop(-1)

sim.run(until = 10)

Custom model¶

Alternatively, a separate pipe model and queue model can be created to implement the same discipline:

Edit mylifo.py with a text editor and create a pipe model in the following way:

import simpype

class MyQueue(simpype.Queue):
    def __init__(self, sim, pipe, id):
        super().__init__(sim, pipe, id)

   @simpype.queue.push
   def push(self, message):
       return self.buffer.append(message)

   @simpype.queue.pop
   def pop(self):
       return self.buffer.pop(-1)


class MyPipe(simpype.Pipe):
    def __init__(self, sim, resource, id):
        super().__init__(sim, resource, id)
        self.add_queue(id = 'lifo', model = 'mylifo')

    @simpype.pipe.enqueue
    def enqueue(self, message):
        return self.queue['lifo'].push(message)

    @simpype.pipe.dequeue
    def dequeue(self):
        return self.queue['lifo'].pop()

# Do NOT remove. This is required for SimPype to build your model.
queue = lambda *args: MyQueue(*args)
pipe = lambda *args: MyPipe(*args)

Create your simulation scenario including the new model:

import simpype
import random

sim = simpype.Simulation(id = 'simple')
gen0 = sim.add_generator(id = 'gen0')
gen0.message.property['priority'] = {
    0: lambda: random.randint(0,1)
}
res0 = sim.add_resource(id = 'res0', pipe = 'mylifo')
res0.random['service'] = {
    0: lambda: 2.0
}

sim.run(until = 10)

Make sure that the file and directory structure is the following:

<working directory>
|-- simple.py
|-- mylifo.py

If you want to change the directory where SimPype looks for custom models, set the following variable in the simulation environment:

import simpype

sim = simpype.Simulation(id = 'simple')
sim.model.dir = '<your model dir>'

Please make sure you have reading permissions for <your model dir>. In this case, the file and directory structure would look like:

<working directory>
|-- simple.py

<your model dir>
|-- mylifo.py