def evaluator(candidate, args):
type_info_price = args.get("type_info_price")
type_info_ecu = args.get("type_info_ecu")
task_base_time = args.get("task_base_time")
task_preds = args.get("task_preds")
comm_speeds = args.get("comm_speeds")
comm_sizes = args.get("comm_sizes")
n_tasks = args.get("n_tasks")
n_nodes = args.get("n_tasks")
n_types = args.get("n_types")
fitness = evaluate(candidate, type_info_price, type_info_ecu,
task_base_time, task_preds, comm_speeds, comm_sizes,
n_tasks, n_nodes, n_types)
return inspyred.ec.emo.Pareto(fitness)
def evaluator(candidate, args):
type_info_price = args.get("type_info_price")
type_info_ecu = args.get("type_info_ecu")
task_base_time = args.get("task_base_time")
task_preds = args.get("task_preds")
comm_speeds = args.get("comm_speeds")
comm_sizes = args.get("comm_sizes")
n_tasks = args.get("n_tasks")
n_nodes = args.get("n_tasks")
n_types = args.get("n_types")
fitness = evaluate(candidate, type_info_price,
type_info_ecu,
task_base_time,
task_preds,
comm_speeds,
comm_sizes,
n_tasks,
n_nodes,
n_types)
return inspyred.ec.emo.Pareto(fitness)
if __name__ == '__main__':
from sys import argv
from workflow import Workflow
from pool import AWS
from emo_tool import get_info, evaluate
from log import get_details
from plot import stat, trans, show
wf = Workflow(argv[1])
pool = AWS("aws.info")
info = get_info(wf, pool)
_, loc, ts = heft2(*info[2:])
c = get_chromosome(loc, ts)
makespan, cost = evaluate(c,
*info[2:-1],
n_tasks=len(loc),
n_nodes=len(loc),
n_types=len(info[1]))
scheme = get_details(c,
*info[:-1],
n_tasks=len(loc),
n_nodes=len(loc),
n_types=len(info[1]))
scheme = [(n, ) + v for n, v in scheme.iteritems()]
print "Makespan: %.0fs\tCost: $%.2f" % (makespan, cost)
# stat(scheme)
# show(*trans(scheme))
break
if len(types) > n_tasks:
types = types[:n_tasks]
c = array.array("I", [])
for l, t in itertools.izip_longest(loc, types, fillvalue=0):
c.append(l)
c.append(t)
return c
if __name__ == '__main__':
from sys import argv
from workflow import Workflow
from pool import AWS
from emo_tool import get_info, evaluate
from log import get_details
from plot import stat, trans, show
wf = Workflow(argv[1])
pool = AWS("aws.info")
info = get_info(wf, pool)
_, loc, ts = heft2(*info[2:])
c = get_chromosome(loc, ts)
makespan, cost = evaluate(c, *info[2:-1], n_tasks=len(loc), n_nodes=len(loc), n_types=len(info[1]))
scheme = get_details(c, *info[:-1], n_tasks=len(loc), n_nodes=len(loc), n_types=len(info[1]))
scheme = [(n,)+v for n,v in scheme.iteritems()]
print "Makespan: %.0fs\tCost: $%.2f" % (makespan, cost)
# stat(scheme)
# show(*trans(scheme))