Esempio n. 1
0
class TestHeftMethods(unittest.TestCase):
    """
	This class test HEFT on the same example graph presented by
	Topcuoglu et al
	"""
    def setUp(self):
        self.workflow = Workflow("{0}/{1}".format(
            current_dir, cfg.test_heuristic_data['topcuoglu_graph_nocalc']))
        env = Environment("{0}/{1}".format(
            current_dir, cfg.test_workflow_data['topcuoglu_graph_system']))
        self.workflow.add_environment(env)

    # self.workflow.load_attributes(cfg.test_heuristic_data['heft_attr'],calc_time=False)

    def test_rank(self):
        rank_values = [108, 77, 79, 80, 69, 63, 42, 35, 44, 14]
        # for task in self.workflow.tasks:
        # 	task.rank = rank_up(self.workflow, task)
        task_ranks = calculate_upward_ranks(self.workflow)
        # upward_rank(self.workflow)
        for task in self.workflow.tasks:
            task.rank = task_ranks[task]
        sorted_tasks = self.workflow.sort_tasks('rank')

        for node in sorted_tasks:
            self.assertTrue(rank_values[node.tid] == int(node.rank))

    def test_schedule(self):
        solution = heft(self.workflow)
        self.assertEqual(80, solution.makespan)
Esempio n. 2
0
 def plan(self, name, workflow, algorithm):
     workflow = ShadowWorkflow(workflow)
     available_resources = self.cluster_to_shadow_format()
     workflow_env = ShadowEnvironment(available_resources, dictionary=True)
     workflow.add_environment(workflow_env)
     plan = WorkflowPlan(name, workflow, algorithm, self.env)
     return plan
Esempio n. 3
0
def run_algorithm(arg, parser):
    if arg['algorithm'] == 'heft':
        pass
        wf = Workflow(arg['workflow'])
        env = Environment(arg['environment'])
        wf.add_environment(env)
        print(heft(wf))
        print(wf.machine_alloc)
Esempio n. 4
0
class TestHeftMethodLargeGraph(unittest.TestCase):
    def test_large_workflow(self):
        self.workflow = Workflow(
            "/home/rwb/Dropbox/PhD/writeups/observation_graph-model/json/3000Node.json"
        )
        env = Environment(
            "/home/rwb/Dropbox/PhD/writeups/observation_graph-model/json/3000Node_sys.json"
        )
        self.workflow.add_environment(env)
        heft(self.workflow)
Esempio n. 5
0
class TestFCFS(unittest.TestCase):
    def setUp(self):
        self.workflow = Workflow("{0}/{1}".format(
            current_dir, cfg.test_heuristic_data['topcuoglu_graph']))
        self.env = Environment("{0}/{1}".format(
            current_dir, cfg.test_workflow_data['topcuoglu_graph_system']))
        self.workflow.add_environment(self.env)

    def test_machine_availability(self):
        solution = Solution(self.env.machines)
        m = list(self.env.machines)[1]  # second machine
        t = Task(1)
        ast = 3
        aft = 11
        machine = m
        t.calculated_runtime[m] = 5
        solution.add_allocation(t, m, ast=ast, aft=aft)

        # Start time is 5, so machine should be unavailable
        self.assertFalse(
            _check_machine_availability(solution, m, start_time=5, task=t))
        # Start time is 2, so *technically* this fine; however, if we take into account task runtime it will overlap
        self.assertFalse(
            _check_machine_availability(solution, m, start_time=2, task=t))
        # Start time is now ahead of the two allocations
        self.assertTrue(
            _check_machine_availability(solution, m, start_time=11, task=t))

        # Add a spanner in the works and have TWO allocations

        u = Task(2)
        ast = 14
        aft = 17
        machine = m
        u.calculated_runtime[m] = 4
        solution.add_allocation(u, m, ast, aft)
        # THis should be false, as start-time+runtime overlaps with 14-17
        self.assertFalse(
            _check_machine_availability(solution, m, start_time=13, task=u))
        # If we change the runtime, however, and start a little earlier, this should 'slot' inbetween the two allocations
        u.calculated_runtime[m] = 2
        self.assertTrue(
            _check_machine_availability(solution, m, start_time=11, task=u))

    def test_fcfs_allocations(self):
        # The order should be [0, 5, 4, 3, 2, 6, 1, 8, 7, 9]
        solution = fcfs(workflow=self.workflow)
        for t in self.workflow.tasks:
            if t.tid == 0:
                alloc = solution.task_allocations[t]
                self.assertEqual(0, alloc.ast)
                self.assertEqual(11, alloc.aft)
            if t.tid == 4:
                continue
        self.assertEqual(solution.makespan, 110)
Esempio n. 6
0
class TestWorkflowClass(unittest.TestCase):
    def setUp(self):
        self.workflow = Workflow("{0}/{1}".format(
            current_dir, cfg.test_workflow_data['topcuoglu_graph']))
        self.env = Environment("{0}/{1}".format(
            current_dir, cfg.test_workflow_data['topcuoglu_graph_system']))
        self.workflow.add_environment(self.env)

    def test_execution_order(self):
        correct_order = [0, 3, 2, 4, 1, 5, 6, 8, 7, 9]
        retval = heft(self.workflow)
        order = self.workflow.solution.execution_order
        for i, alloc in enumerate(order):
            self.assertEqual(correct_order[i], alloc.tid)
Esempio n. 7
0
class TestNSGAIIMethods(unittest.TestCase):
    def setUp(self):
        self.wf = Workflow("{0}/{1}".format(
            current_dir, cfg.test_metaheuristic_data['topcuoglu_graph']))
        env = Environment("{0}/{1}".format(
            current_dir, cfg.test_metaheuristic_data['graph_sys_with_costs']))
        self.wf.add_environment(env)
        self.rng = 10

        # These two are generated in the above tests, so we can garauntee their correctness
    def test_dominates(self):
        pop = generate_population(self.wf,
                                  size=4,
                                  rng=self.rng,
                                  skip_limit=100)
Esempio n. 8
0
class TestAddEnvironment(unittest.TestCase):

	def setUp(self):
		self.wf = Workflow("{0}/{1}".format(current_dir, cfg.test_workflow_data['topcuoglu_graph']))
		self.env = Environment("{0}/{1}".format(current_dir, cfg.test_workflow_data['topcuoglu_graph_system']))

	def test_time_false(self):
		"""
		When we read in the environment file and add it to to the workflow, we do some
		pre-processing of the default computing values. If we have errors in the workflow or environment
		config files, we need to ensure we exit appropriately.
		"""

		# We have a workflow with time is false, but time is actually true. This means there is more
		# than one value in the 'comp' attribute, which is incorrect.

		# We havea workflow with time: true, but time is actually false; i.e. there is only one value
		# stored in 'comp' attribute, when there should be multiply. REMEMBER, time: true implies that runtime
		# has been previously calculated for each machine.

		workflow_true_but_false = Workflow('test/data/workflow/exception_raised_timeistrue.json')
		self.assertRaises(TypeError, workflow_true_but_false.add_environment, self.env)
		pass

	def test_add_environment(self):

		retval = self.wf.add_environment(self.env)
		self.assertEqual(retval, 0)
		for task in self.wf.graph.nodes:
			if task.tid == 5:
				self.assertEqual([24, 28, 15], list(task.calculated_runtime.values()))

		self.assertEqual(self.wf.graph.edges[Task(3), Task(7)]['data_size'], 27)
Esempio n. 9
0
class TestPHeftMethods(unittest.TestCase):
    def setUp(self):
        self.workflow = Workflow(cfg.test_heuristic_data['pheft_graph'])
        env = Environment(cfg.test_workflow_data['topcuoglu_graph_system'])
        self.workflow.add_environment(env)
        self.up_oct_rank_values = [72, 41, 37, 43, 31, 41, 17, 20, 16, 0]
        self.up_rank_values = [169, 114, 102, 110, 129, 119, 52, 92, 42, 20]

    def tearDown(self):
        return -1

    def test_up_rank(self):
        task_ranks = calculate_upward_ranks(self.workflow)
        # upward_rank(self.workflow)
        for task in self.workflow.tasks:
            task.rank = task_ranks[task]
        sorted_tasks = self.workflow.sort_tasks('rank')
        for node in sorted_tasks:
            self.assertTrue(int(node.rank) == self.up_rank_values[node.tid])

    def test_oct_rank(self):
        oct_rank_matrix = generate_ranking_matrix(self.workflow)
        # upward_oct_rank(self.workflow, oct_rank_matrix)
        for task in self.workflow.tasks:
            sum = 0
            for (t, p) in oct_rank_matrix:
                if t is task:
                    sum += oct_rank_matrix[(t, p)]

            rank = int(sum / len(self.workflow.env.machines))
            task.rank = rank

        sorted_tasks = self.workflow.sort_tasks('rank')

        for node in sorted_tasks:
            self.assertEqual(self.up_oct_rank_values[node.tid], node.rank)

    def test_heft_schedule(self):
        # upward_rank(self.workflow)
        solution = heft(self.workflow)
        self.assertTrue(solution.makespan == 133)

    def test_pheft_schedule(self):
        # upward_rank(self.workflow)
        solution = pheft(self.workflow)
        self.assertTrue(solution.makespan == 122)
Esempio n. 10
0
class TestHeftMethodCalcTime(unittest.TestCase):
    def setUp(self):
        self.workflow = Workflow("{0}/{1}".format(
            current_dir, cfg.test_workflow_data['topcuoglu_graph']))
        env = Environment("{0}/{1}".format(
            current_dir, cfg.test_workflow_data['topcuoglu_graph_system']))

        # self.workflow = Workfow(cfg.test_heuristic_data['topcuoglu_graph'],
        # cfg.test_heuristic_data['topcuoglu_graph_system'])

        self.workflow.add_environment(env)

    # self.workflow.load_attributes(cfg.test_heuristic_data['flops_test_attr'])

    def test_schedule(self):
        solution = heft(self.workflow)
        self.assertEqual(98, solution.makespan)
Esempio n. 11
0
    def _initialise_shadow_workflows(self, observation, cluster):
        """
        Use the SHADOW library workflow model to build the graph

        Parameters
        ----------
        observation

        Returns
        -------
        workflow : shadow.models.workflow.Worflow
            A wrapper for NetworkX DiGraph object with additional information

        """
        workflow = Workflow(observation.workflow)
        available_resources = self._cluster_to_shadow_format(cluster)
        workflow_env = Environment(available_resources, dictionary=True)
        workflow.add_environment(workflow_env)
        return workflow
Esempio n. 12
0
# Copyright (C) 26/6/20 RW Bunney

# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.

# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.

# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <https://www.gnu.org/licenses/>.
from shadow.models.workflow import Workflow
from shadow.models.environment import Environment
import shadow.visualiser.graph as sgraph
from IPython.display import Image

workflow_file = 'TestAskapCont_channels-10_shadow.json'
sys = 'final_heft_sys.json'
workflow = Workflow(workflow_file)
env = Environment(sys)
workflow.add_environment(env)

# png = sgraph.visualise_graph(workflow, workflow_file.strip('.json')+'.png')

graphviz = sgraph.convert_to_graphviz(workflow)
graphviz.render('output.gz')
Esempio n. 13
0
from queue import Queue
from shadow.algorithms.heuristic import heft

from shadow.models.workflow import Workflow, Task
from shadow.models.environment import Environment

HEFTWorkflow = Workflow('heft.json')
env = Environment('sys.json')
HEFTWorkflow.add_environment(env)

print(heft(HEFTWorkflow))

DelayWorkflow = Workflow('heft_delay.json')
DelayWorkflow.add_environment(env)
print(heft(DelayWorkflow))


def calc_task_delay(task, delay, workflow):
    t = workflow.graph.tasks[task]
    aft = t.aft
    update_list = list(workflow.graph.successors(t))
    # add 10 to the start and finish time of each of these successors, and their successors
    update_queue = Queue()
    update_queue.put(update_list)
    print(update_queue)
    return workflow


task = HEFTWorkflow.tasks[0]
calc_task_delay(task, 10, workflow=HEFTWorkflow)
Esempio n. 14
0
class TestPopulationGeneration(unittest.TestCase):
    """
	Tests how we generate solutions for NSGAII and SPEAII
	"""
    def setUp(self):
        self.wf = Workflow("{0}/{1}".format(
            current_dir, cfg.test_metaheuristic_data['topcuoglu_graph']))
        env = Environment("{0}/{1}".format(
            current_dir, cfg.test_metaheuristic_data['graph_sys_with_costs']))
        self.wf.add_environment(env)
        self.rng = np.random.default_rng(40)

    def test_generate_exec_orders(self):
        """
		Expected results from creating a topological sort of `self.wf` for a
		given population size. The `skip_limit` in this example is 1,
		which means that if we were to generate a population of execution
		orders, we would go through all topological sorts without skipping any.

		Population size: 4
		Skip limit - 1
		rng = default_rng(40) (This is specified in setUp).

		With `skip_limit=1`, the first top_sort created will be:

			[0, 5, 4, 3, 2, 6, 1, 8, 7, 9]

		Test 1: Check to make sure the expected topological sort generated is
		the one we get from `generate_exec_orders`

		Test 2: Check that the next topological sort is different - that is,
		we are not generating the same TopSort over and over again.

		Returns
		-------
		Pass/Fail
		"""

        # Test 1
        top_sort = generate_exec_orders(self.wf,
                                        popsize=4,
                                        rng=self.rng,
                                        skip_limit=1)
        order = [0, 5, 4, 3, 2, 6, 1, 8, 7, 9]
        curr = next(top_sort)
        sum = 0
        for i in range(len(order)):
            if curr[i].tid != order[i]:
                sum += 1
        self.assertEqual(sum, 0)

        # Test 2
        curr = next(top_sort)
        for i in range(len(order)):
            if curr[i].tid != order[i]:
                sum += 1
        self.assertGreater(sum, 0)

    def test_calc_start_finish_time(self):
        """
		Given a 'randomly' generated Solution, ensure that we calculate the
		correct start and finish times for the random allocation.

		Test 1: Duplicates `test_generate_exec_orders` to confirm RNG returns
		the same values for the same function parameters.

		Test 2: Initial conditions, where there are no allocations already in
		the solution


		Returns
		-------
		Pass/Fail
		"""

        # Test 1
        top_sort = generate_exec_orders(self.wf,
                                        popsize=4,
                                        rng=self.rng,
                                        skip_limit=1)
        curr = next(top_sort)
        self.assertListEqual([x.tid for x in curr],
                             [0, 5, 4, 3, 2, 6, 1, 8, 7, 9])
        machine = 0

        solution = GASolution
        ast, aft = calc_start_finish_times(curr[0], 0, self.wf, solution)

    def test_generate_allocations(self):
        """
		After generating an execution order, we need to allocate tasks to
		machines whilst maintaining precedence constraints.

		Test 1: Duplicates `test_generate_exec_orders`


		Test2: Generate a list of allocations, and creates a set from this
		list of IDs. This compare to a list of sets; the difference between
		each set (which represents task-machin pairings) should be 0.


		Returns
		-------

		"""
        top_sort = generate_exec_orders(self.wf,
                                        popsize=4,
                                        rng=self.rng,
                                        skip_limit=1)
        curr = next(top_sort)
        self.assertListEqual([x.tid for x in curr],
                             [0, 5, 4, 3, 2, 6, 1, 8, 7, 9])
        machines = list(self.wf.env.machines)
        # Solution should contain allocations between tasks and machines
        soln = generate_allocations(machines, curr, self.wf, self.rng)

        # first should be 0,0,1,2,0
        # For each Task in soln, we will have an allocation
        # e.g. tid=0, m='cat0_m0'.
        # If we go through each machine, and each task, the order should be
        ##################  cat0_m1  |cat1_m1 | cat2_m2
        # task_alloc_order = [0, 1, 4, 5, 2, 6, 8, 3, 7, 9]
        # task_alloc_order = [0, 5, 4, 1, 2, 8, 3, 7, 9]
        alloc_sets = [
            {0, 5, 4, 8, 7},
            {6, 9},
            {3, 2, 1},
        ]
        i = 0
        for machine in machines:
            x = len(alloc_sets[i]
                    & set(soln.list_machine_allocations(machine)))
            self.assertEqual(x, 0)

        self.assertEqual(soln.makespan, 107)

    # self.assertAlmostEqual(soln.solution_cost, 110.6, delta=0.01)

    def test_pop_gen(self):
        pop = generate_population(self.wf, size=25, rng=self.rng, skip_limit=5)
        soln1 = pop[0]
        # First solution should be the same solution we
        # have been working with previously.
        self.assertEqual(107, soln1.makespan)
        soln2 = pop[2]
        self.assertNotEqual(114, soln2.makespan)
        soln5 = pop[4]
        self.assertEqual(145, soln5.makespan)
        soln25 = pop[24]
        self.assertEqual(130, soln25.makespan)

    # self.assertEqual(soln.makespan,107)
    # This means we are dealing with a
    # what our the costs?

    @unittest.skip
    def test_nondomsort(self):
        pop = generate_population(self.wf,
                                  size=4,
                                  rng=self.rng,
                                  skip_limit=100)
        seed = 10
        objectives = []
        # print(pop)
        non_dom_sort(pop, objectives)
        for p in pop:
            print(p.nondom_rank)

    def test_create_sample_pop(self):
        logger.debug("HEFT makespan {0}".format(heft(self.wf).makespan))
        pop = generate_population(self.wf, size=25, rng=self.rng, skip_limit=5)
        for soln in pop:
            self.assertEqual(soln.execution_order[-1].task.aft, soln.makespan)
        logger.debug("GA Initial Population")
        logger.debug("########################")
        for soln in pop:
            logger.debug(("Execution order: {0}".format(soln.execution_order)))
            logger.debug("Allocations: {0}".format(
                soln.list_all_allocations()))
            logger.debug("Makespan (s): {0}".format(
                calculate_fitness(['time'], soln)))
            logger.debug("Cost ($){0}".format(calculate_fitness(['cost'],
                                                                soln)))

            soln.fitness = calculate_fitness(['time', 'cost'], soln)
        fig, ax = plt.subplots()
        ax.set_xlim([90, 200])
        ax.set_ylim([100, 250])
        x = [soln.fitness['time'] for soln in pop]
        y = [soln.fitness['cost'] for soln in pop]
        ax.scatter(x, y, c='red')
        ax.set_axisbelow(True)
        ax.legend()
        ax.grid(True)
        plt.xlabel('Solution Runtime')
        plt.ylabel('Solution execution cost')
        plt.show()
Esempio n. 15
0
class TestGASelectionMethods(unittest.TestCase):
    def setUp(self):
        self.wf = Workflow("{0}/{1}".format(
            current_dir, cfg.test_metaheuristic_data['topcuoglu_graph']))
        env = Environment("{0}/{1}".format(
            current_dir, cfg.test_metaheuristic_data['graph_sys_with_costs']))
        self.wf.add_environment(env)
        self.rng = np.random.default_rng(40)

    def test_binary_tournament(self):
        pop = generate_population(self.wf, size=25, rng=self.rng, skip_limit=5)
        for soln in pop:
            soln.fitness = calculate_fitness(['time', 'cost'], soln)
        compare_prob = 0.5
        parent1 = binary_tournament(pop, compare_prob, self.rng)
        logger.debug(parent1.execution_order)
        compare_prob = 0.7
        parent2 = binary_tournament(pop, compare_prob, self.rng)
        # parent2 = binary_tournament(pop, self.rng)
        logger.debug(parent2.execution_order)
        self.assertSequenceEqual([0, 5, 3, 4, 2, 1, 6, 8, 7, 9],
                                 [t.task.tid for t in parent1.execution_order])
        logger.debug("Fitness: {0}".format(parent1.fitness))
        self.assertSequenceEqual([0, 5, 4, 1, 3, 2, 8, 6, 7, 9],
                                 [t.task.tid for t in parent2.execution_order])
        logger.debug("Fitness: {0}".format(parent2.fitness))
        #
        #
        # fig, ax = plt.subplots()
        # x = [soln.fitness['time'] for soln in pop]
        # y = [soln.fitness['cost'] for soln in pop]
        # ax.set_xlim([90,200])
        # ax.set_ylim([100,170])
        # ax.grid(True)
        # ax.scatter(x, y, c='red')
        # ax.set_axisbelow(True)
        # selectedx = [parent1.fitness['time'], parent2.fitness['time']]
        # selectedy = [parent1.fitness['cost'], parent2.fitness['cost']]
        # ax.scatter(selectedx, selectedy, c='blue')
        # ax.legend()
        # plt.xlabel('Solution Runtime')
        # plt.ylabel('Solution execution cost binary')

        # plt.show()

    def test_crossover(self):
        pop = generate_population(self.wf, size=25, rng=self.rng, skip_limit=5)
        for soln in pop:
            soln.fitness = calculate_fitness(['time', 'cost'], soln)

        random.seed(self.rng)

        p1 = binary_tournament(pop)
        self.assertSequenceEqual([0, 5, 3, 2, 1, 4, 7, 8, 6, 9],
                                 [t.task.tid for t in p1.execution_order])
        p2 = binary_tournament(pop)
        self.assertSequenceEqual([0, 5, 4, 1, 3, 2, 8, 6, 7, 9],
                                 [t.task.tid for t in p2.execution_order])
        c1, c2 = crossover(p1, p2, self.wf)
        p1order = [t.task.tid for t in p1.execution_order]
        c1order = [t.task.tid for t in c1.execution_order]
        self.assertSequenceEqual(p1order, c1order)
        for m in c1.machines:
            for allocation in c1.list_machine_allocations(m):
                if allocation.task.tid == 4:
                    self.assertEqual('cat1_m1', allocation.machine.id)
        for m in c2.machines:
            for allocation in c2.list_machine_allocations(m):
                if allocation.task.tid == 4:
                    self.assertEqual('cat0_m0', allocation.machine.id)
        self.assertNotEqual(p2.makespan, c2.makespan)

        fig, ax = plt.subplots()
        c1.fitness = calculate_fitness(['time', 'cost'], c1)
        c2.fitness = calculate_fitness(['time', 'cost'], c2)
        crossx = [c1.fitness['time'], c2.fitness['time']]
        crossy = [c1.fitness['cost'], c2.fitness['cost']]
        ax.scatter(crossx, crossy, c='green')
        x = [soln.fitness['time'] for soln in pop]
        y = [soln.fitness['cost'] for soln in pop]
        ax.grid(True)
        ax.set_xlim([90, 200])
        ax.set_ylim([100, 170])
        ax.scatter(x, y, c='red')
        ax.set_axisbelow(True)
        selectedx = [p1.fitness['time'], p2.fitness['time']]
        selectedy = [p1.fitness['cost'], p2.fitness['cost']]
        ax.scatter(selectedx, selectedy, c='blue')
        ax.legend()
        plt.xlabel('Solution Runtime')
        plt.ylabel('Solution execution cost')
        plt.show()

    def test_mutation(self):
        pop = generate_population(self.wf, size=25, rng=self.rng, skip_limit=5)
        for soln in pop:
            soln.fitness = calculate_fitness(['time', 'cost'], soln)

        random.seed(self.rng)

        p1 = binary_tournament(pop, self.rng, 0.6)
        self.assertSequenceEqual([0, 5, 3, 2, 1, 4, 7, 8, 6, 9],
                                 [t.task.tid for t in p1.execution_order])

        mutated_child = mutation(p1, self.wf, 'swapping', rng=self.rng)
        mutated_order_swapped = [0, 5, 3, 1, 7, 4, 2, 8, 6, 9]
        self.assertSequenceEqual(
            mutated_order_swapped,
            [alloc.task.tid for alloc in mutated_child.execution_order])
        selected_machine = None
        for machine in mutated_child.machines:
            if machine.id == 'cat2_m2':
                selected_machine = machine
        mutated_alloc = mutated_child.list_machine_allocations(
            selected_machine)
        self.assertSequenceEqual([7, 2, 9],
                                 [alloc.task.tid for alloc in mutated_alloc])

    def test_overall(self):
        total_generations = 25
        crossover_probability = 0.5
        mutation_probability = 0.4
        popsize = 25
        pop = generate_population(self.wf,
                                  size=popsize,
                                  rng=self.rng,
                                  skip_limit=5)
        for soln in pop:
            soln.fitness = calculate_fitness(['time', 'cost'], soln)

        generations = []
        x = [soln.fitness['time'] for soln in pop]
        y = [soln.fitness['cost'] for soln in pop]
        weights = [200 * i for i in Counter(x).values() for j in range(i)]

        generations.append((x, y))
        parents1 = []
        parents2 = []
        for gen in range(total_generations):
            new_pop = []
            parent1 = None
            parent2 = None
            while len(new_pop) < len(pop):
                p1 = binary_tournament(pop, 0.5, self.rng)
                p2 = binary_tournament(pop, 0.5, self.rng)
                parent1 = p1
                parent2 = p2
                if random.random() < crossover_probability:
                    c1, c2 = crossover(p1, p2, self.wf)
                    new_pop.append(c1)
                    new_pop.append(c2)
                elif random.random() < mutation_probability:
                    c1 = mutation(p1, self.wf, 'swapping', rng=self.rng)
                    if c1 is None:
                        # The mutation didn't occur due to selection issue
                        continue
                    new_pop.append(c1)
                else:
                    new_pop.append(p1)
                    new_pop.append(p2)
                    continue
            tmp_pop = pop + new_pop
            for soln in tmp_pop:
                soln.fitness = calculate_fitness(['time', 'cost'], soln)
                # soln.total_fitness = soln.calc_total_fitness()
            tmp_pop.sort(key=lambda x: (x.fitness['time'], x.fitness['cost']))
            # tmp_pop.sort(key=lambda solution: solution.total_fitness)
            pop = tmp_pop[0:popsize]
            weights = [10 * i for i in Counter(x).values() for j in range(i)]
            x = [soln.fitness['time'] for soln in pop]
            y = [soln.fitness['cost'] for soln in pop]
            parent1_x = [parent1.fitness['time']]
            parent1_y = [parent1.fitness['cost']]
            parent2_x = [parent2.fitness['time']]
            parent2_y = [parent2.fitness['cost']]
            generations.append((x, y))
            parents1.append((parent1_x, parent1_y))
            parents2.append((parent2_x, parent2_y))
            plt.draw()
            plt.show()

        for soln in pop:
            logger.info(soln.fitness)

        fig, ax = plt.subplots()
        ax.set_xlim([80, 250])
        ax.set_ylim([100, 250])
        scatter = ax.scatter(generations[0][0],
                             generations[0][1],
                             c='blue',
                             alpha=.2,
                             label='New Pop')
        parent_scatter = ax.scatter(parents1[0][0],
                                    parents1[0][1],
                                    c='red',
                                    alpha=0.8,
                                    label='Parent 1')
        parent2_scatter = ax.scatter(parents2[0][0],
                                     parents1[0][1],
                                     c='red',
                                     alpha=0.8,
                                     label='Parent 2')
        ax.legend()

        def animate(i):
            scatter.set_offsets(np.c_[generations[i][0], generations[i][1]])
            parent_scatter.set_offsets(np.c_[parents1[i][0], parents1[i][1]])
            parent2_scatter.set_offsets(np.c_[parents2[i][0], parents2[i][1]])
            ax.set_xlabel('Runtime (s) \n Generation {0}'.format(i))
            ax.set_ylabel('Cost ($)')

        anim = FuncAnimation(fig, animate, interval=100, frames=25)

        plt.draw()

        anim.save('filename.mp4', fps=1)
Esempio n. 16
0
import pandas as pd
import seaborn as sns

from shadow.models.workflow import Workflow
from shadow.models.environment import Environment
from shadow.algorithms.heuristic import heft, fcfs

# df = pd.DataFrame(columns=["time", "channels", "algorithm", "diff"])
df = pd.DataFrame()
WORKFLOW = "routput/shadow_Continuum_ChannelSplit_10.json"
CLUSTER = "routput/system_spec_40_200-400_1.0"

wf_fcfs = Workflow(WORKFLOW)
env_fcfs = Environment(CLUSTER)
wf_fcfs.add_environment(env_fcfs)
soln = fcfs(wf_fcfs)
seq = -1

#
# for p in range(len(self.processors)):
#     comp = 0
#     for task in list(self.graph.nodes()):
#         comp = comp + self.comp_matrix[task.tid][p]
#     if (seq is -1) or (comp < seq):
#         seq = comp

for x in range(10, 100, 10):
    print("Scheduling {0} Channels".format(x))
    WORKFLOW = "routput/shadow_Continuum_ChannelSplit_{0}.json".format(x)
    CLUSTER = "routput/system_spec_40_200-400_1.0"