Next Article in Journal
Improved Reconstruction Algorithm of Wireless Sensor Network Based on BFGS Quasi-Newton Method
Previous Article in Journal
Theoretical Analysis of the Time Transient of the THz Self-Mixing Rectification Voltage in a Semiconductor Barrier
Previous Article in Special Issue
Application-Aware Flow Forwarding Service for SDN-Based Data Centers
 
 
Review
Peer-Review Record

A Comprehensive Survey of In-Band Control in SDN: Challenges and Opportunities

Electronics 2023, 12(6), 1265; https://doi.org/10.3390/electronics12061265
by David Carrascal, Elisa Rojas *,†, Jose M. Arco †, Diego Lopez-Pajares, Joaquin Alvarez-Horcajo and Juan Antonio Carral
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Electronics 2023, 12(6), 1265; https://doi.org/10.3390/electronics12061265
Submission received: 31 January 2023 / Revised: 21 February 2023 / Accepted: 5 March 2023 / Published: 7 March 2023

Round 1

Reviewer 1 Report

This paper surveys the different approaches that 11 have been proposed so far towards the advance in in-band SDN control, based on four main 12 categories: automatic routing, fast failure recovery, network bootstrapping, and distributed control.

Although the article has the merit of being published, it is better to see and take into consideration my remarks:

1. The introduction is too short and needs to be improved. it is important to put a comprehensible state of the art and a primary motivation. 

2. We summarized the main advantages of each model  below. It is recommended to make a table where you describe not only the advantages but also the disadvantages.

3. Related work and contribution of this survey. In this section, the surveys already done are very poorly explained and you summarize this part too much. You have to detail each article and then compare them to your work. Also, a table is needed for a comparison. I recommend this article

Singh, S., Jha, R.K. “A Survey on Software Defined Networking: Architecture for Next Generation Network,” J Netw Syst Manage vol. 25, pp. 321–374, 2017.

4. Please add a section where you discuss only SDN and openflow.

5. A flow chart in the form of a picture is important to show the organisation of your survey.

6. in the automation routing sub-section, it is good to add articles that discuss the AI aspect.  I recommend the following articles 

   Y. -R. Chen, A. Rezapour, W. -G. Tzeng and S. -C. Tsai, "RL-Routing: An SDN Routing Algorithm Based on Deep Reinforcement Learning," in IEEE Transactions on Network Science and Engineering, vol. 7, no. 4, pp. 3185-3199, 1 Oct.-Dec. 2020.

M. A. Ouamri, G. Barb, D. Singh and F. Alexa, "Load Balancing Optimization in Software-Defined Wide Area Networking (SD-WAN) using Deep Reinforcement Learning," 2022 International Symposium on Electronics and Telecommunications (ISETC), Timisoara, Romania, 2022, pp. 1-6.

7.  The conclusion is poor for a survey 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This survey deals with a specific, but, nevertheless, interesting topic (in-band control SDN). The set of paper considered is wide and covers the full spectrum of problems connected to the considered topic. Anyway , in my opinion, the paper could be improved considered the following aspects:

 a) I think that the topology discovery is a key element in the set of needed network functions and the explanation of its exclusion is not clear. So, I suggest to include the topology discovery in the main list of in-band SDN control proposals i.e., together with the automatic routing, the fast recovery, and the network bootstrapping.

b) The authors should enlarge the presentation of the possible control protocols usable for the interaction between a switch and the controller and among switches. Control protocols are key parts of a SDN architecture and deserve a more detailed presentation. In practice, in my opinion, the concepts illustrated in the rows 209-225 (page 7 of 23) merit a deeper discussion focused on the main differences among them.

c) The short presentations of the papers dealing with the four categories of network functions (i.e., automatic routing, fast recovery and so on) are quite arid. The authors should try to present a taxonomy and a synoptic picture of the various solutions/proposals and discuss the main differences among them. Such a taxonomy could help the readers to a better orienteering in the topic.

d) Performance aspects are almost completely neglected, a brief discussion of performance offered by the various solutions could be a useful completion of the survey.

In summary, the authors should consider the previous suggestions and include them in a new version of the paper.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

the authors have improved the article and followed the reviewers' instructions 

Back to TopTop