Towards Fair and QoS-Aware Bandwidth Allocation in Next-Generation Multi-Gigabit WANs

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The opinions are as follows: 1 How does the FQ-DBA algorithm specifically integrate traffic classification, fairness enforcement, and QoS-aware allocation to optimize network performance? 2 In what ways does the proposed FQ-DBA framework demonstrate scalability and energy efficiency in the context of next-generation WANs? 3. Lack of comparative analysis with existing algorithms currently proposed. 4 If feasible, other language programs (such as Matlab) can be provided for easy reproduction. 5 Please increase the number of references

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Dear authors, in the next paragraphs, my comments about your manuscript.

The article presents an explicit and up-to-date proposal for dynamic bandwidth allocation in multi-gigabit WAN networks, naming it FQ-DBA. The structure is very well organized, spanning from the literature review to a thorough description of the algorithm, its implementation in ns-3, and experimental comparison between reference algorithms.

Simulation results are consistent, accentuating clear gains under critical metrics (fairness, QoS, throughput, and latency). The clarity of the presentation of those results, including comparative tables and figures, adds to the robustness of the evaluation. The work is, moreover, relevant because it draws energy efficiency and scalability issues-the two prime concerns for advanced networks-into consideration.

 

Points for Improvement:

1.While it appears logically sound, a standpoint can be held about less than thorough literature review. Although some of the classic works are cited (max-min fairness, proportional fairness, WFQ, PONs), references need to be found more toward recent ones (2020-2024) on DBA in SDN, in the 5G/6G scenario, and for edge computing to make the framework much more current.

2.The methodology section offers a too generic presentation of the algorithms: it allegedly describes four DBA algorithms, while it focuses clearly only on FQ-DBA, creating some ambiguity. More concrete information including description of simulation parameters, such as network topology, simulation run times, and load scenario variation, need to be provided.

3.The above algorithm is presented with clear pseudocode, but the integration with actual ns-3 code is incomplete, containing obvious placeholders such as “replace with actual logic” that would greatly hinder reproducibility. It would be much appreciated if the final validated code is made available.

4.The paper lacks sufficient discussion on the limitations of FQ-DBA. For example, the computational overhead introduced by this algorithm has not been evaluated, nor has the scalability of FQ-DBA been assessed in scenarios with thousands of flows, something that is paramount in very large WANs.

5.Energy, while stated as a goal, is never really quantified in the results; no concrete metrics are given that would certainly prove any energy gains.

6.Further lacking is the investigation into heterogeneous scenarios like mobile traffic or hybrid networks (WAN + wireless access), which would better establish the wider applicability of the algorithm.

7.Finally, it would be good to see the exploration of integration possibilities with machine learning techniques (reinforcement learning, traffic prediction) so as to keep the work aligned with the ongoing trend toward intelligent networks.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The add all the comments mentioned by me