Offloading of Atomic Tasks in Satellite Networks: A Fast Adaptive Resource Collaboration Method

Round 1

Reviewer 1 Report

The paper presents a scheduling optimization method based on maximum flow theory and a dynamic caching model for satellite communications.

The paper is generally well written, based on 42 reference titles, most of them relatively new.

There are several weak points, as follows

- regarding Figure 1. Overview of our method - please explain it in the text - what is the role and tasks of each block, what does the colors represent within the task queue etc.

-there is not a clear connection between formulas (4) and (5) and the reminder of the paper. Please emphasize this connection.

- definitions should appear before they are discussed in the text.

- in Definition 1 - please give the physical interpretation and explain the role of variable  \rho_{q_i}^{s}(t)

- in Definition 2 - please give the physical interpretation and explain the role of variable  \phi_{q_i}^{s}(t)

- in equation 6 - it is \phi_{q}^{s}(t) or \phi_{Q}^{s}(t) (as described in the paragraph above?. Similarly, is \rho_{q}^{s}(t) or \rho_{Q}^{s}(t)? There is an ambiguity in notations here.

- same in rows 304-305

- in eq. (7) - who is D_{s}? who are C1/2/3?

- regarding the model presented in section 4.1. Problem Conversion - why have you chosen 6 satellites (4 MEO/LEO, 2 HEO)? Is that situation relevant? comprehensive?

- in eq.(8) the variables are not explained; who is f, u, v, I, s, t etc. Also plea

- Fig. 4  should be placed after it is explained in the text;

- in Table 2, explain the physical significance of gamma, n_clusters, n_neighbors, degree. Also the text in the table should be with italic, as the one in the text above it;

- same for table 3.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Nice paper suitable for publication in its current form.

Author Response

Thank you for your approval of my paper

Reviewer 3 Report

This paper proposes a scheduling optimization method containing an atomic task offloading model based on maximum flow theory and a dynamic caching model. The authors determined the task offloading upper limit in the satellite network.
The organization of some parts of the paper must be rethought. Introduction are related work have partial overlap. Section 3.5 must be improved to be clear the problem formation. It seems to be incomplete. What's the relation of this section with the previous ones? Moreover, what is the link with 4.1? A mix of definitions and concepts makes the problem that the authors are addressing unclear.
Another detail that is not discussed is the cache size. What is the influence of the cache size on the offloading decision? 
In terms of results, authors should improve the discussion on the datasets used. They were presented in table 1. But those datasets should be discussed to understand which information is included.
Lastly, the conclusion must be improved to reflect the contribution of this work.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The paper looks much better after the corrections made, therefore I consider that it can be published in present format.