Distributed Cross-Domain Optimization for Software Defined Industrial Internet of Things

Round 1

Reviewer 1 Report

In this article, the authors propose a model for optimizing data flows in IIoT networks managed by multiple SDN controllers. Here, the real-time delivery of data flows, path selection, and allocation of constrained resources are jointly optimized to maximize the overall utility of different domains. Using the distributed Prox-SADMM algorithm, the problem to be solved is decomposed into multiple intra-domain sub-problems that can be solved in parallel. The simulation results show that the algorithm can solve the optimization problem quickly and is scalable.

The article clearly presents its own contributions and is convincing in its results. The system architecture is clearly presented and the authors' approach is understandable. The quality of the presentation is successful except for minor details.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The Industrial Internet of Things (IIoT) is a promising paradigm for a smart industry that provides intelligent services through interconnected networks. The authors of this paper focus on designing a flexible flow control strategy in multi-region management scenarios to optimize the quality of these services. They establish a flow optimization model based on the IIoT network managed by multiple Software-Defined Networking (SDN) controllers. This model jointly optimizes real-time delivery, route selection, and resource allocation to maximize the total utilities of domains while protecting privacy. The authors decompose the problem into intra-domain subproblems and implement the interaction among domains through slack variables. The solution is a distributed Proximal Symmetric Alternating Direction Method of Multipliers (Prox-SADMM) algorithm. The numerical simulation shows that the Prox-SADMM algorithm is suitable for large-scale networks and provides guidance for setting appropriate parameter values. The paper's main contributions include the proposed architecture, joint optimization problem, and distributed algorithm for efficient and privacy-protected resource allocation in IIoT networks.

The proposal is interesting. Here my comments:

1. How does the algorithm ensure privacy expectations are met while solving the joint optimization problem in IIoT networks? Explain in the motivation.

2. The state of the art and related works could be complemented with updated references on the subject related to the maintenance of IIoT networks, such as:

- Bao, B., Yao, Q., Li, C., Sun, Z., Li, X., Bai, W., ... & Li, Y. (2022, October). Cross-Domain Resource Allocation Scheme with Unified Control Architecture in Software Defined Optical Access Network. In Photonics (Vol. 9, No. 10, p. 740). MDPI.

- Cortés-Leal, A., Cárdenas, C., & Del-Valle-Soto, C. (2022). Maintenance 5.0: Towards a Worker-in-the-Loop Framework for Resilient Smart Manufacturing. Applied Sciences, 12(22), 11330.

3. The model is clear, but the authors could better explain the significance of the slack variables introduced in the Prox-SADMM algorithm described in the paper and how they contribute to the privacy protection of IIoT users.

4. In the discussion, you could delve into the impact of performance metrics on the functioning of the network.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf