def iperf_UDP_test_scenario():
# Simulates IPERF transfers at different file sizes
testbed = BasicTestbed(host_ip=HOST_IP)
# from 250k to 9.75 mb in 250kb steps
# we add one modest "Warm up" sessions to start the connections for d_pepsal and qpep which have high first packet costs but only
# experience these costs once, when the customer starts the respective applications
iperf_file_sizes = [25*1000, 50*1000, 100*1000, 150*1000]+[(i/4)*1000000 for i in range(1, 47)]
iperf_file_sizes.sort()
benchmarks = [IperfUDPBenchmark(file_sizes=iperf_file_sizes[4:], bw_limit="50M", iterations=2)]
plain_scenario = PlainScenario(name="Plain", testbed=testbed, benchmarks=copy.deepcopy(benchmarks))
vpn_scenario = OpenVPNScenario(name="OpenVPN", testbed=testbed, benchmarks=copy.deepcopy(benchmarks))
pepsal_scenario = PEPsalScenario(name="PEPSal", testbed=testbed, benchmarks=copy.deepcopy(benchmarks), terminal=True, gateway=False)
distributed_pepsal_scenario = PEPsalScenario(name="Distributed PEPsal", gateway=True, terminal=True, testbed=testbed,benchmarks=copy.deepcopy(benchmarks))
qpep_scenario = QPEPScenario(name="QPEP", testbed=testbed, benchmarks=copy.deepcopy(benchmarks))
scenarios = [qpep_scenario, distributed_pepsal_scenario, vpn_scenario, plain_scenario, pepsal_scenario]
for scenario in scenarios:
logger.debug("Running iperf test scenario " + str(scenario.name))
iperf_scenario_results = {}
scenario.run_benchmarks()
for benchmark in scenario.benchmarks:
logger.debug("Running Iperf Test Scenario (", str(scenario.name), ") with file sizes: " + str(benchmark.file_sizes))
iperf_scenario_results = benchmark.results
print(iperf_scenario_results)
scenario.print_results()
benchmark.save_results_to_db(str(scenario.name),"opensand")
def plr_plt_scenario():
testbed = BasicTestbed(host_ip=HOST_IP)
sites_to_check = ["https://www.nasa.gov"]
plr_levels = list(numpy.unique(list(numpy.geomspace((1*pow(10, -7)), (1*pow(10,2)), num=20)) + list(numpy.geomspace((1*pow(10,-7)), (1*pow(10,2)), num=10))))
plain_scenario = PlainScenario(name="Plain", testbed=testbed, benchmarks=[])
vpn_scenario = OpenVPNScenario(name="OpenVPN", testbed=testbed, benchmarks=[])
pepsal_scenario = PEPsalScenario(name="PEPSal", testbed=testbed, benchmarks=[], terminal=True, gateway=False)
distributed_pepsal_scenario = PEPsalScenario(name="Distributed PEPsal",terminal=True, gateway=True, testbed=testbed,benchmarks=[])
qpep_scenario = QPEPScenario(name="QPEP", testbed=testbed, benchmarks=[])
scenarios = [plain_scenario, pepsal_scenario, distributed_pepsal_scenario, qpep_scenario, vpn_scenario]
for scenario in scenarios:
benchmarks = [SitespeedBenchmark(hosts=sites_to_check, scenario=scenario, iterations=1, sub_iterations=int(os.getenv("PLR_PLT_ITERATIONS")))]
if scenario.name == os.getenv("SCENARIO_NAME"):
logger.debug("Running PLR PLT scenario" + str(scenario.name))
plr_scenario_results = {}
for plr_level in plr_levels[int(os.getenv("PLR_MIN_INDEX")):int(os.getenv("PLR_MAX_INDEX"))]:
plr_string = numpy.format_float_positional(plr_level, precision=7, trim='-')
plr_scenario_results[str(plr_string)]=[]
for j in range(0, int(os.getenv("PLR_META_ITERATIONS"))):
scenario.benchmarks = copy.deepcopy(benchmarks)
scenario.deploy_scenario()
scenario.testbed.set_plr_percentage(plr_string, st_out=False, gw_out=True)
logger.debug("Running PLR PLT for " + str(scenario.name) + " at " + str(plr_string) + " batch " + str(j) +" of " + str(os.getenv("PLR_META_ITERATIONS")))
scenario.run_benchmarks(deployed=True)
for benchmark in scenario.benchmarks:
plr_scenario_results[str(plr_string)].append(benchmark.results)
logger.debug("Interim PLR/PLT Results (PLR: " + str(plr_string) + " meta_iteration: " + str(j) + "/" + str(int(os.getenv("PLR_META_ITERATIONS"))) + " Scenario: " + str(scenario.name) +"): " + str(plr_scenario_results) + "\n")
print("Final PLR PLT results for " + str(scenario.name))
print("***********************************************")
print(plr_scenario_results)
print('\n*********************************************')
logger.success("PLR/PLT test complete")
def plt_test_scenario(testbed=None):
if testbed is None:
testbed = BasicTestbed(host_ip=HOST_IP)
benchmarks = [SitespeedBenchmark(hosts=['https://www.sina.com.cn'])]
plain_scenario = PlainScenario(name="Plain",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
vpn_scenario = OpenVPNScenario(name="OpenVPN",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
pepsal_scenario = PEPsalScenario(name="PEPSal",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks),
terminal=True,
gateway=False)
distributed_pepsal_scenario = PEPsalScenario(
name="Distributed PEPsal ",
terminal=True,
gateway=True,
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
qpep_scenario = QPEPScenario(name="QPEP",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
#scenarios = [plain_scenario, pepsal_scenario, distributed_pepsal_scenario, qpep_scenario, vpn_scenario]
scenarios = [distributed_pepsal_scenario]
for scenario in scenarios:
logger.debug("Running PLT test scenario " + str(scenario.name))
scenario.deploy_scenario()
scenario.run_benchmarks(deployed=True)
for benchmark in scenario.benchmarks:
print("Results for PLT " + str(scenario.name))
print(benchmark.results)
def plt_test_scenario(testbed=None):
if testbed is None:
testbed = RealWorldTestbed()
alexa_top_20 = [
"https://www.google.com", "https://www.youtube.com",
"https://www.tmall.com", "https://www.facebook.com",
"https://www.baidu.com", "https://www.qq.com", "https://www.sohu.com",
"https://www.taobao.com", "https://www.360.cn", "https://www.jd.com",
"https://www.yahoo.com", "https://www.amazon.com",
"https://www.wikipedia.org", "https://www.weibo.com",
"https://www.sina.com.cn", "https://www.reddit.com",
"https://www.live.com", "https://www.netflix.com",
"https://www.okezone.com", "https://www.vk.com"
]
with open(str(os.getenv("TESTBED_FILE"))) as file:
testbed_name = file.readlines()[0]
plain_scenario = PlainScenario(name="plain",
testbed=testbed,
benchmarks=[])
vpn_scenario = OpenVPNScenario(name="ovpn", testbed=testbed, benchmarks=[])
pepsal_scenario = PEPsalScenario(name="pepsal",
testbed=testbed,
benchmarks=[],
terminal=True,
gateway=False)
distributed_pepsal_scenario = PEPsalScenario(name="dist_pepsal",
terminal=True,
gateway=True,
testbed=testbed,
benchmarks=[])
qpep_scenario = QPEPScenario(name="qpep", testbed=testbed, benchmarks=[])
scenarios = [
plain_scenario, pepsal_scenario, distributed_pepsal_scenario,
qpep_scenario, vpn_scenario
]
for scenario in scenarios:
scenario.benchmarks = [
SitespeedBenchmark(
hosts=alexa_top_20[int(os.getenv("ALEXA_MIN")
):int(os.getenv("ALEXA_MAX"))],
scenario=scenario,
iterations=int(os.getenv("PLT_ITERATIONS")),
sub_iterations=int(os.getenv("PLT_SUB_ITERATIONS")))
]
logger.debug("Running PLT test scenario " + str(scenario.name))
scenario.deploy_scenario()
scenario.run_benchmarks(deployed=True)
for benchmark in scenario.benchmarks:
print("Results for PLT " + str(scenario.name))
print(benchmark.results)
benchmark.save_results_to_db(str(scenario.name), testbed_name)
for scenario in scenarios:
if scenario.name == os.getenv("SCENARIO_NAME"):
scenario.print_results()
def iperf_test_scenario():
# Simulates IPERF transfers at different file sizes
testbed = RealWorldTestbed()
# from 250k to 9.75 mb in 250kb steps
# we add one modest "Warm up" sessions to start the connections for d_pepsal and qpep which have high first packet costs but only
# experience these costs once, when the customer starts the respective applications
iperf_file_sizes = [25 * 1000, 50 * 1000, 100 * 1000, 150 * 1000
] + [(i / 4) * 1000000 for i in range(1, 47)]
iperf_file_sizes.sort()
with open(str(os.getenv("TESTBED_FILE"))) as file:
testbed_name = file.readlines()[0]
benchmarks = [
IperfBenchmark(file_sizes=iperf_file_sizes[
int(os.getenv("IPERF_MIN_SIZE_INDEX")
):int(os.getenv("IPERF_MAX_SIZE_INDEX"))],
iterations=int(os.getenv("IPERF_ITERATIONS")))
]
plain_scenario = PlainScenario(name="plain",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
vpn_scenario = OpenVPNScenario(name="ovpn",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
pepsal_scenario = PEPsalScenario(name="pepsal",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks),
terminal=True,
gateway=False)
distributed_pepsal_scenario = PEPsalScenario(
name="dist_pepsal",
gateway=True,
terminal=True,
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
qpep_scenario = QPEPScenario(name="qpep",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
scenarios = [
qpep_scenario, distributed_pepsal_scenario, vpn_scenario,
plain_scenario, pepsal_scenario
]
for scenario in scenarios:
logger.debug("Running iperf test scenario " + str(scenario.name))
iperf_scenario_results = {}
scenario.run_benchmarks()
for benchmark in scenario.benchmarks:
logger.debug("Running Iperf Test Scenario (", str(scenario.name),
") with file sizes: " + str(benchmark.file_sizes))
iperf_scenario_results = benchmark.results
print(iperf_scenario_results)
benchmark.save_results_to_db(str(scenario.name), testbed_name)
scenario.print_results()
def iperf_test_scenario():
testbed = BasicTestbed(host_ip=HOST_IP)
# from 250k to 9.75 mb in 250kb steps
# we add one modest "Warm up" sessions to start the connections for d_pepsal and qpep which have high first packet costs but only
# experience these costs once, when the customer starts the respective applications
iperf_file_sizes = [25 * 1000, 50 * 1000, 100 * 1000, 150 * 1000
] + [(i / 4) * 1000000 for i in range(1, 40)]
#iperf_file_sizes = [(i/2)*1000000 for i in range(1, 20)]
iperf_file_sizes.sort()
logger.debug("Running Iperf Test Scenario with file sizes: " +
str(iperf_file_sizes))
benchmarks = [IperfBenchmark(file_sizes=iperf_file_sizes)]
plain_scenario = PlainScenario(name="Plain",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
vpn_scenario = OpenVPNScenario(name="OpenVPN",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
pepsal_scenario = PEPsalScenario(name="PEPSal",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks),
terminal=True,
gateway=False)
distributed_pepsal_scenario = PEPsalScenario(
name="Distributed",
gateway=True,
terminal=True,
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
qpep_scenario = QPEPScenario(name="QPEP",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
#scenarios = [qpep_scenario, distributed_pepsal_scenario, vpn_scenario, plain_scenario, pepsal_scenario]
scenarios = [vpn_scenario]
for scenario in scenarios:
logger.debug("Running iperf test scenario " + str(scenario.name))
iperf_scenario_results = {}
scenario.run_benchmarks()
for benchmark in scenario.benchmarks:
iperf_scenario_results = benchmark.results
print(iperf_scenario_results)
up_speeds = []
down_speeds = []
for key in iperf_scenario_results:
up_speeds.append(iperf_scenario_results[key]["sent_bps"])
down_speeds.append(iperf_scenario_results[key]["received_bps"])
print(scenario.name)
print(" Up: ", up_speeds)
print(" Down:", down_speeds)
def plr_test_scenario():
testbed = BasicTestbed(host_ip=HOST_IP)
iperf_file_sizes=[1000000, 2000000, 5000000, 1000000]
# a balacned distribution across log scale of 20 points, plus key powers of ten, total length of 28 checkpoints from 10e-7 to 100
plr_levels = list(numpy.unique(list(numpy.geomspace((1*pow(10, -7)), (1*pow(10,2)), num=20)) + list(numpy.geomspace((1*pow(10,-7)), (1*pow(10,2)), num=10))))
benchmarks = [IperfBenchmark(file_sizes=iperf_file_sizes, reset_on_run=True, iterations=1)]
# test with pepsal vs qpep vs plain
pepsal_scenario = PEPsalScenario(name="PEPSal", testbed=testbed,benchmarks=copy.deepcopy(benchmarks))
qpep_scenario = QPEPScenario(name="QPEP", testbed=testbed, benchmarks=copy.deepcopy(benchmarks))
plain_scenario = PlainScenario(name="Plain", testbed=testbed, benchmarks=copy.deepcopy(benchmarks))
vpn_scenario = OpenVPNScenario(name="OpenVPN", testbed=testbed, benchmarks=copy.deepcopy(benchmarks))
distributed_pepsal_scenario = PEPsalScenario(name="Distributed PEPsal", gateway=True, terminal=True, testbed=testbed,benchmarks=copy.deepcopy(benchmarks))
scenarios = [qpep_scenario, plain_scenario, vpn_scenario, pepsal_scenario, distributed_pepsal_scenario]
for scenario in scenarios:
if scenario.name == os.getenv("SCENARIO_NAME"):
logger.debug("Running packet loss rate scenario " + str(scenario.name))
iperf_scenario_results = {}
for plr_level in plr_levels[int(os.getenv("PLR_MIN_INDEX")):int(os.getenv("PLR_MAX_INDEX"))]:
plr_string = numpy.format_float_positional(plr_level, precision=7, trim='-')
iperf_scenario_results[str(plr_string)] = []
for j in range(0, int(os.getenv("PLR_META_ITERATIONS"))):
logger.debug("Running PLR for " + str(scenario.name) + " at " + str(plr_string) + " batch " + str(j) + " of " + str(os.getenv("PLR_META_ITERATIONS")))
scenario.deploy_scenario()
scenario.testbed.set_plr_percentage(plr_string, st_out=False, gw_out=True)
for i in range(0, int(os.getenv("IPERF_ITERATIONS"))):
scenario.benchmarks = copy.deepcopy(benchmarks)
scenario.run_benchmarks(deployed=True)
for benchmark in scenario.benchmarks:
iperf_scenario_results[str(plr_string)].append(benchmark.results)
# if the link breaks, we need to restart the ip routes
for key in benchmark.results.keys():
if(benchmark.results[key]["sent_bps"]) == 0:
scenario.deploy_scenario()
scenario.testbed.set_plr_percentage(plr_string, st_out=False, gw_out=True)
logger.warning("Failed Iperf Run @ " + str(plr_string))
break
logger.debug("Interim PLR Results (PLR: " + str(plr_string) + " sub_iter: " + str(i) + "/" + str(int(os.getenv("IPERF_ITERATIONS"))) + " Scenario: " + str(scenario.name) +"): " + str(iperf_scenario_results))
print("Final PLR Results for ", scenario.name)
print("*********************************")
print(iperf_scenario_results)
print("\n******************************")
logger.success("PLR Test Complete")
def attenuation_test_plt_scenario():
testbed = BasicTestbed(host_ip=HOST_IP)
attenuation_levels = [i * 0.5 for i in range(0, 11)]
benchmarks = [
SitespeedBenchmark(hosts=["https://www.bbc.co.uk"], iterations=5)
]
# test with pepsal vs qpep vs plain
# NB - due to the networking requirements of PEPsal, a special testbed launch order is required in attenuation_test_pepsal_scenario to allow access to web content
qpep_scenario = QPEPScenario(name="QPEP Attenuation ",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
plain_scenario = PlainScenario(name="Plain Attenuation ",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
vpn_scenario = OpenVPNScenario(name="OpenVPN",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
scenarios = [plain_scenario, qpep_scenario, vpn_scenario]
for scenario in scenarios:
scenario_results = []
scenario_errors = []
for attenuation_level in attenuation_levels:
scenario.deploy_scenario()
scenario.benchmarks = copy.deepcopy(benchmarks)
logger.debug("Running attenuation scenario test for attenuation=" +
str(attenuation_level))
scenario.testbed.set_downlink_attenuation(attenuation_level)
scenario.testbed.run_attenuation_scenario()
scenario.testbed.connect_terminal_workstation()
scenario.run_benchmarks(deployed=True)
for benchmark in scenario.benchmarks:
if len(benchmark.results) > 0:
scenario_results.append(mean(benchmark.results))
scenario_errors.append(benchmark.errors)
print(scenario.name, "Results: ")
print(" PLT: ", scenario_results)
print(" ERR: ", scenario_errors)
logger.success("Attenuation PLT Test Complete")
def attenuation_test_iperf_scenario():
testbed = BasicTestbed(host_ip=HOST_IP)
attenuation_levels = [i * 0.25 for i in range(0, 21)]
# test a 10mb transfer
iperf_file_sizes = [10000000]
benchmarks = [
IperfBenchmark(file_sizes=iperf_file_sizes, reset_on_run=True)
]
# test with pepsal vs qpep vs plain
pepsal_scenario = PEPsalScenario(name="PEPsal Attenuation ",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
qpep_scenario = QPEPScenario(name="QPEP Attenuation ",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
plain_scenario = PlainScenario(name="Plain Attenuation ",
testbed=testbed,
benchmarks=copy.deepcopy(benchmarks))
scenarios = [plain_scenario]
for scenario in scenarios:
up_results = []
down_results = []
all_measurements = {}
for attenuation_level in attenuation_levels:
scenario.deploy_scenario()
at_up_measurements = []
at_down_measurements = []
scenario.testbed.set_downlink_attenuation(attenuation_level)
scenario.testbed.run_attenuation_scenario()
for i in range(0, 5):
scenario.benchmarks = copy.deepcopy(benchmarks)
logger.debug("Running attenuation scenario test #" + str(i) +
" for attenuation=" + str(attenuation_level))
scenario.run_benchmarks(deployed=True)
for benchmark in scenario.benchmarks:
for key in benchmark.results:
at_up_measurements.append(
benchmark.results[key]["sent_bps"])
at_down_measurements.append(
benchmark.results[key]["received_bps"])
if (benchmark.results[key]["sent_bps"]) == 0:
#if the attenuated link breaks you have to restart it
scenario.deploy_scenario()
scenario.testbed.set_downlink_attenuation(
attenuation_level)
scenario.testbed.run_attenuation_scenario()
logger.debug(
"Attenuation Result: " + str(attenuation_level) +
" --- " + str(benchmark.results[key]["sent_bps"]) +
"/" + str(benchmark.results[key]["received_bps"]))
all_measurements[str(scenario.name) + "_" +
str(attenuation_level) +
"_up"] = at_up_measurements
all_measurements[str(scenario.name) + "_" +
str(attenuation_level) +
"_down"] = at_down_measurements
# after running 5 sample tests, add their mean to our reported average
up_results.append(mean(at_up_measurements))
down_results.append(mean(at_down_measurements))
print("Current Up: ", up_results)
print("Current Down: ", down_results)
print("All Measurements: ", all_measurements)
print(scenario.name, "Results: ")
print(" Down:", down_results)
print(" Up:", up_results)
print(" All:", all_measurements)
logger.success("Attenuation Test Complete")
default=False,
action='store_true')
args = parser.parse_args()
# First define our OpenSAND testbed environment
testbed = None
if args.orbit == 'GEO':
testbed = BasicTestbed(host_ip=args.xhost, display_number=args.display)
else:
testbed = LeoTestbed(host_ip=args.xhost, display_number=args.display)
# Next set the scenario
scenario_dict = {
"plain":
PlainScenario(name="Plain", testbed=testbed, benchmarks=[]),
"qpep":
QPEPScenario(name="QPEP", testbed=testbed, benchmarks=[]),
"pepsal_distributed":
PEPsalScenario(name="Distributed PEPsal",
testbed=testbed,
gateway=True,
benchmarks=[]),
"pepsal_integrated":
PEPsalScenario(name="Integrated PEPsal",
testbed=testbed,
benchmarks=[]),
"open_vpn":
OpenVPNScenario(name="OpenVPN", testbed=testbed, benchmarks=[])
}
scenario = scenario_dict[args.scenario]
# Launch the testbed and deploy the PEP/VPN if relevant