Steady-state Single Flow Experiment
Flows that use congestion control algorithms (CCAs) for rate control should be able to fully utilize network resources in the absence of competition. When a flow governed by its CCA operates alone on a path, the goal of the CCA should be to converge to an efficient steady-state behavior quickly. For example, NewReno converges to a periodic sawtooth function.
A CCA should fill the pipe, i.e., the bandwidth (bottleneck rate) of the network path should be fully utilized to maximize the throughput. Furthermore, self-inflicted queueing delay at the bottleneck queue should be avoided. At best, the operating point of the CCA maximizes throughput and minimizes delay.
Scenario
In this experiment a single flow governed by a CCA
operates in a static dumbbell network. Greedy
source traffic ensures that the flow is network-limited. To
evaluate how network parameters influence the performance of CCAs, the
experiment is repeated for different network paths by setting the
experiment parameter path. Each path defines its own set of
network parameters.
To summarize the setup:
Topology: Dumbbell topology (\(K=1\)) with static network parameters defined by the
pathparameterFlows: A single flow (\(K=1\)) with a CCA
Traffic Generation Model: Greedy source traffic
Experiment Results
Experiment #15
Parameters
Command: ns3-dev-ccperf-static-dumbbell-default --experiment-name=steady_state_single_flow --db-path=benchmark_TcpNewReno.db '--parameters={aut:TcpNewReno,path:static.india_to_aws_india}' --aut=TcpNewReno --stop-time=15s --seed=42 --bw=100.42Mbps --loss=0.0 --qlen=173p --qdisc=FifoQueueDisc --rtts=54ms --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 #15. 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 | 0.23 |
| cov_throughput_l4 | 0.24 |
| flow_completion_time_l4 | 15.00 |
| mean_cwnd_l4 | 359.43 |
| mean_delivery_rate_l4 | 75.43 |
| mean_est_qdelay_l4 | 0.56 |
| mean_idt_ewma_l4 | 0.14 |
| mean_in_flight_l4 | 359.16 |
| mean_network_power_l4 | 1386.08 |
| mean_one_way_delay_l7 | 395.19 |
| mean_recovery_time_l4 | 126.06 |
| mean_sending_rate_l4 | 76.11 |
| mean_sending_rate_l7 | 77.57 |
| mean_srtt_l4 | 54.56 |
| mean_throughput_l4 | 75.60 |
| mean_throughput_l7 | 75.60 |
| mean_utility_mpdf_l4 | -0.02 |
| mean_utility_pf_l4 | 4.29 |
| mean_utilization_bdp_l4 | 0.83 |
| mean_utilization_bw_l4 | 0.75 |
| total_retransmissions_l4 | 403.00 |
| 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 | 0.30 |
| mean_qdisc_length_l2 | 2.08 |
| mean_sending_rate_l1 | 78.60 |
| total_qdisc_drops_l2 | 402.00 |
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 | 251.80 |
| mean_agg_throughput_l4 | 27.72 |
| mean_agg_utility_mpdf_l4 | -1.17 |
| mean_agg_utility_pf_l4 | 7.98 |
| mean_agg_utilization_bdp_l4 | 0.40 |
| mean_agg_utilization_bw_l4 | 0.36 |
| mean_entropy_fairness_throughput_l4 | 1.61 |
| mean_jains_fairness_throughput_l4 | 0.91 |
| mean_product_fairness_throughput_l4 | 13712.03 |
Figures
The following figures show the results of the experiment #15.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
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.
Mean Operating Point Plane
The mean throughput and mean smoothed round-trip time (sRTT) at the transport layer of each flow.
Comparison of Congestion Control Algorithms (CCAs)
Figures
