Scalability Experiment
Long fat networks (LFNs) have high data rates and large two-way
propagation delays. For congestion control algorithms (CCAs), it can be
challenging to fully utilize bandwidth resources in networks with a
large bandwidth-delay product (BDP) given by BDP=
bw \(\times\)
RTT, where \(\texttt{bw}\)
is the bandwidth and \(\texttt{RTT}\)
the two-way propagation delay. In this experiment, the influence of the
BDP on the operating point of a CCA is evaluated. A CCA should fully
exhaust the available bandwidth irrespective of the BDP.
Scenario
A static dumbbell topology with a single flow that uses greedy source
traffic is set up. The experiment parameters are the bandwidth
bw and two-way propagation delay rtts. To
evaluate if that is the case, the experiment is repeated for different
BDPs.
To summarize the setup:
Topology: Dumbbell topology (\(K=1\)) with static network parameters
Flows: A single flow (\(K=1\)) that uses a CCA
Traffic Generation Model: Greedy source traffic
Experiment Results
Experiment #21
Parameters
Command: ns3-dev-ccperf-responsiveness-default --experiment-name=scalability --db-path=benchmark_TcpNewReno.db '--parameters={aut:TcpNewReno,bw:256Mbps,rtt:15ms}' --aut=TcpNewReno --stop-time=15s --seed=42 --bw=256Mbps --time-series=0s,3.75s --rate-series=16Mbps,256Mbps --rtt-series=15ms,15ms --loss-series=0.0,0.0 --loss=0.0 --qlen=320p --qdisc=CoDelQueueDisc --rtts=15ms --sources=src_0 --destinations=dst_0 --protocols=TCP --algs=TcpNewReno --recoveries=TcpPrrRecovery --start-times=0s --stop-times=15s '--traffic-models=Greedy(bytes=0)'
Flows
| src | dst | transport_protocol | cca | cc_recovery_alg | traffic_model | start_time | stop_time |
|---|---|---|---|---|---|---|---|
| src_0 | dst_0 | TCP | TcpNewReno | TcpPrrRecovery | Greedy(bytes=0) | 0.00 | 15.00 |
Metrics
The following tables list the flow, link, and network metrics of experiment #21. Refer to the the metrics page for definitions of the listed metrics.
Flow Metrics
Flow metrics capture the performance of an individual flow. They are measured at the endpoints of a network path at either the source, the receiver, or both. Bold values indicate which flow achieved the best performance.
| Metric | flow_1 |
|---|---|
| cov_in_flight_l4 | 0.68 |
| cov_throughput_l4 | 0.66 |
| flow_completion_time_l4 | 15.00 |
| mean_cwnd_l4 | 213.06 |
| mean_delivery_rate_l4 | 146.69 |
| mean_est_qdelay_l4 | 3.09 |
| mean_idt_ewma_l4 | 0.27 |
| mean_in_flight_l4 | 212.58 |
| mean_network_power_l4 | 8998.53 |
| mean_one_way_delay_l7 | 606.00 |
| mean_recovery_time_l4 | 39.86 |
| mean_sending_rate_l4 | 147.03 |
| mean_sending_rate_l7 | 148.83 |
| mean_srtt_l4 | 18.09 |
| mean_throughput_l4 | 146.81 |
| mean_throughput_l7 | 146.81 |
| mean_utility_mpdf_l4 | -0.02 |
| mean_utility_pf_l4 | 4.52 |
| mean_utilization_bdp_l4 | 1.08 |
| mean_utilization_bw_l4 | 0.79 |
| total_retransmissions_l4 | 16.00 |
| total_rtos_l4 | 0.00 |
Link Metrics
Link metrics are recorded at the network links of interest, typically at bottlenecks. They include metrics that measure queue states. Bold values indicate which link achieved the best performance.
| Metric | btl_forward |
|---|---|
| mean_qdisc_delay_l2 | 2.81 |
| mean_qdisc_length_l2 | 20.66 |
| mean_sending_rate_l1 | 152.40 |
| total_qdisc_drops_l2 | 15.00 |
Network Metrics
Network metrics assess the entire network as a
whole by aggregating other metrics, e.g., the aggregated throughput of
all flows.
Hence, the network metrics has only one column named net.
| Metric | net |
|---|---|
| mean_agg_in_flight_l4 | 546.72 |
| mean_agg_throughput_l4 | 94.47 |
| mean_agg_utility_mpdf_l4 | -1.16 |
| mean_agg_utility_pf_l4 | 22.07 |
| mean_agg_utilization_bdp_l4 | 1.26 |
| mean_agg_utilization_bw_l4 | 0.94 |
| mean_entropy_fairness_throughput_l4 | 2.30 |
| mean_jains_fairness_throughput_l4 | 0.93 |
| mean_product_fairness_throughput_l4 | 4779807472.98 |
Figures
The following figures show the results of the experiment #21.Time Series Plot of the Operating Point
Time series plot of the number of segments in flight, the smoothed round-trip time (sRTT), and the throughput at the transport layer.
In Flight vs Mean Operating Point
The mean throughput and mean smoothed round-trip time (sRTT) at the transport layer of each flow. The optimal operating point is highlighted with a star (magenta). The joint operating point is given by the aggregated throughput and the mean sRTT over all flows
Mean Operating Point Plane
The mean throughput and mean smoothed round-trip time (sRTT) at the transport layer of each flow.
Distribution of the Operating Point
The empirical cumulative distribution function (eCDF) of the throughput and smoothed round-trip time (sRTT) at the transport layer of each flow.
Comparison of Congestion Control Algorithms (CCAs)
Figures
