ECN Not Enabled on 100 Mbps Ethernet: BBR-v3 and BBR-n+ Performance Evaluation

Experimental data comparing the performance of TCP congestion control algorithms BBR-v3 and BBR-n+ (BBR Smart Exit) against TCP Cubic. The dataset was created by Muhammad Ahsan and last updated on April 24, 2026. It includes results from tests in wired and wireless connectivity scenarios, evaluating metrics like throughput, HTTP delay, and ping latency.

Use Cases

• Compare throughput gains between BBR-n+ and BBR-v3 based on receiver window limitation tests described in the paper.
• Analyze reductions in HTTP delay and ping latency for BBR-n+ versus BBR-v3 and Cubic as mentioned in the experimental results.
• Evaluate fairness between BBR-n+ and Cubic flows in scenarios with eight or more concurrent streams, a noted limitation.
• Study the interaction of congestion control algorithms with modern AQMs like CAKE and FQ_CoDel, which were used in testing.

Strengths

• Data is associated with a specific research paper evaluating the BBR-n+ algorithm, providing context.
• Results include quantified performance gains, such as a 15–20% median throughput improvement for BBR-n+ over BBR-v3.
• Experiments cover multiple scenarios, including wired and wireless connectivity, and use specific AQMs (CAKE and FQ_CoDel).

Limitations

• Dataset is very small at 5.5 KB, indicating limited scope or a summary of results rather than raw experimental data.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for statistical analysis.

Provenance

Source

Muhammad Ahsan via figshare

Collection Method

Likely contains results from network experiments comparing TCP congestion control algorithms.

Freshness

Last updated 2026-04-24 17:51:43; freshness should be verified.