This is a simulation tool for microservice based SDN. It uses a executable jar built upon CloudSim Plus (https://cloudsimplus.org/) as the backend, specifically "org.cloudsimplus:cloudsim-plus:7.3.0"
The backend jar is available from https://drive.google.com/file/d/1PWtYCWDBRV02VcOD1kn_J-lLbsxyfXhT/view?usp=sharing.
The backend.jar must be put into the same directory as your program. The JRE is below (Other distribution will result in error, this is why I hate java...): https://community.chocolatey.org/packages/microsoft-openjdk11
Here is my new version of the simulation built purely upon Python: https://ulfaric.github.io/PyCloudSim-legacy/
import random
from copy import deepcopy
from typing import List
import pandas as pd
import plotly.graph_objects as go
import plotly.express as px
from dash import Dash, Input, Output, dcc, html
from PySDNSim.Backend import Backend
from PySDNSim.Config import Config
from PySDNSim.Experiment import Experiment
from PySDNSim.Host import Host
from PySDNSim.Job import Job
from PySDNSim.Microservice import Microservice
from PySDNSim.NetworkService import NetworkService, create_network_service
from treelib import Node, Tree
create simulation config
sim_config = Config(seed=1024, interval=1.0, step_size=0.01)
create host
host = Host(
cpus=64,
ram=102400,
bw=100000,
storage=1024000,
max_power=1600.0,
static_power=300.0,
replicas=3,
)
create microservices
microservices: List[Microservice] = list()
ms_mqtt_broker = Microservice(
name="mqtt_broker",
size=512,
cpus=2,
replicas=1,
max_replicas=3,
cpu_ratio=25,
ram_ratio=32,
bw_ratio=25,
)
ms_mqtt_broker.add_auto_scale("cpu",0.5)
ms_mqtt_broker.add_auto_scale("ram",0.5)
ms_mqtt_broker.add_auto_scale("bw",0.5)
microservices.append(ms_mqtt_broker)
ms_chirpstack_gateway = Microservice(
name="chirpstack_gateway",
size=128,
cpus=2,
replicas=1,
max_replicas=3,
cpu_ratio=25,
ram_ratio=32,
bw_ratio=25,
)
ms_chirpstack_gateway.add_auto_scale("cpu",0.5)
ms_chirpstack_gateway.add_auto_scale("ram",0.5)
ms_chirpstack_gateway.add_auto_scale("bw",0.5)
microservices.append(ms_chirpstack_gateway)
ms_chirpstack = Microservice(
name="chirpstack",
size=128,
cpus=4,
replicas=1,
max_replicas=3,
cpu_ratio=10,
ram_ratio=32,
bw_ratio=25,
)
ms_chirpstack.add_auto_scale("cpu",0.5)
ms_chirpstack.add_auto_scale("ram",0.5)
ms_chirpstack.add_auto_scale("bw",0.5)
microservices.append(ms_chirpstack)
ms_chirpstack_rest_api = Microservice(
name="chirpstack_rest_api",
size=128,
cpus=2,
replicas=1,
max_replicas=3,
cpu_ratio=5,
ram_ratio=128,
bw_ratio=25,
)
ms_chirpstack_rest_api.add_auto_scale("cpu",0.5)
ms_chirpstack_rest_api.add_auto_scale("ram",0.5)
ms_chirpstack_rest_api.add_auto_scale("bw",0.5)
microservices.append(ms_chirpstack_rest_api)
ms_postgresql = Microservice(
name="postgresql",
size=2048,
cpus=2,
replicas=1,
max_replicas=3,
cpu_ratio=50,
ram_ratio=128,
bw_ratio=100,
)
ms_postgresql.add_auto_scale("cpu",0.5)
ms_postgresql.add_auto_scale("ram",0.5)
ms_postgresql.add_auto_scale("bw",0.5)
microservices.append(ms_postgresql)
ms_redis = Microservice(
name="redis",
size=2048,
cpus=2,
replicas=1,
max_replicas=3,
cpu_ratio=50,
ram_ratio=128,
bw_ratio=100,
)
ms_redis.add_auto_scale("cpu",0.5)
ms_redis.add_auto_scale("ram",0.5)
ms_redis.add_auto_scale("bw",0.5)
microservices.append(ms_redis)
create network services
ns_list = list()
register_device = create_network_service(
name="register_device",
microservices=["chirpstack", "redis", "chirpstack_gateway"],
schdeule=[0, 1, 1],
schedule_length=[10, 10, 10],
ms_pool=microservices
)
ns_list.append(register_device)
read_data = create_network_service(
name="receive_data",
microservices=["chirpstack_gateway", "mqtt_broker", "chirpstack", "postgresql"],
schdeule=[0, 1, 2, 3],
schedule_length=[10, 10, 10, 10],
ms_pool=microservices
)
ns_list.append(read_data)
retrive_data = create_network_service(
name="retrive_data",
microservices=["chirpstack_rest_api", "chirpstack", "postgresql", "chirpstack"],
schdeule=[0, 1, 2, 3],
schedule_length=[10, 10, 10, 10],
ms_pool=microservices
)
ns_list.append(retrive_data)
start a simulation that randomly select network services at each iteration.
for iter in range(10):
num_ns = random.randint(1, 20)
chosen_ns: List[NetworkService] = deepcopy(random.choices(ns_list, k=num_ns))
for ns in chosen_ns:
ns.offset_schedule(random.randint(0, 5))
experiment = Experiment(
name=f"{iter}",
config=sim_config,
host=host,
microservices=microservices,
network_services=chosen_ns,
)
backend.run_experiment(experiment=experiment, output_path="./results")