Resource Allocation Strategy for Satellite Edge Computing Based on Task Dependency
Round 1
Reviewer 1 Report
1. Rewrite the abstract and also specify the objectives for which method outperforms.
2. Include clear motivation for the paper and rewrite the contributions.
3. All of the abbreviations should be provided with the full definition at their first occurrence, and refrain from repeating the use of the full form after the first time.
4. Include a comparision table with recent works(2021-2023).
5. Improve the quality of Figure 1 and need explanation.
6. For Dependency task scheduling, we have different scientific workflow applications for cloud computing. In this paper what type of scheduling applications are used?
for reference
Pillareddy, V.R.; Karri, G.R. MONWS: Multi-Objective Normalization Workflow Scheduling for Cloud Computing. Appl. Sci. 2023, 13, 1101. https://doi.org/10.3390/app13021101.
7. Equations citations are required in the text.
Minor editing of English language required
Author Response
Responses to reviewers are in the word file
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This study proposed solutions to address improper resource allocation problem in satellite edge computing. I have following comments/opinions:
- What is DAG? Is it established universal term? Most of the outputs are analyzed based on DAG. But there is no any brief explanation and abbreviation of DAG.
- Line 77 & 82>>Section 2: related research and section 5: related work is summarized. But section 5 has conclusion. << Please refine the section organization.
- The proposed use case architecture in Figure 2 leads to communications between satellite ground stations with the space control section as an obvious structure. It does not reflect the hierarchical computing scenarios from cloud to fog to edge computing. Please provide the complete setup of the use case diagram in term of edge computing.
- What is this reverse learning (RL)? The effect of using and not using RL is not clearly presented form the experimental/analytical point of view.
- The line 11 of algorithm 1 mentions about pointing to Algorithm 2, which has not been outlined before, and it comes only after Alg.1, please make a proper sequence of tasks in the algorithm.
- All algorithms presented in the manuscript are expected to present into standard algorithm format e.g. follow the latex package (algirithm2e) for standard presentation of algorithm.
- Contribution one has highlighted the optimization in terms of computing as well as commutation resources. Edge computing has the major concerns to reduce latency/propagation delays. I wonder, there is no particular analysis or experimentation with respect to address such concern. Please clarify it.
- L344: please mention table number instead of the term “following.”
Moderate review of language is required.
Author Response
word file has been uploaded
Author Response File: Author Response.pdf
Reviewer 3 Report
This paper address the scheduling problem in resource allocation for satellite network edge computing. The resource allocation problem is defined and then the optimization is pursued with an improved sparrow search algorithm. Task completion rates are analyzed with different satellite computing resources, communication resources, number of tasks, and ground computing capacity.
Some acronyms need to be defined before they are used. Examples include TBID, DAG, STK, and TDISSA, making it difficult to understand. In fact, TBID has never been defined. I found it odd to compare the algorithm optimization results with random search mechanism. Is random processing typical for scientific edge computing scheduling?
The English writing is good in general. There are a few places where transition words may not be necessary. For example, in the second paragraph of section 1, the word 'however' is unnecessary; in the fourth paragraph, the word "therefore" is unnecessary.
Please define acronyms before they are used.
Author Response
Responses to reviewers are in the word file
Author Response File: Author Response.pdf
Reviewer 4 Report
This manuscript proposes a satellite-ground joint deployment network architecture based on task dependency, and then schedules the subtasks of the application DAG with dependency, and then proposes a scheduling algorithm of TBID. In addition, to overcome the problem that the traditional sparrow search algorithm falls into local optimality in the search process, this manuscript improves it and uses the improved sparrow search algorithm to allocate resources. In this manuscript, reverse learning, random search and Cauchy mutation operators are introduced into the Sparrow search algorithm to improve the convergence and global search ability of the algorithm. The simulation results are also presented. The talked issue is interesting, but there are several concerns.1) What is the main question addressed by the research?
2) What does it add to the subject area compared with other published material?
3) The fitness function should be described in detail. why the authors used this fitness function.
4) The fitness function references should be highlighted.
5) Writing should be improved.
6) Conclusion should be supported by data.
Moderate editing of English language required
Author Response
word file has been uploaded
Author Response File: Author Response.pdf
Reviewer 5 Report
1.- It is difficult to follow the article, especially since from time to time you have to be looking for what certain acronyms mean that are not declared, for example:
Page 2: What does TBID mean?
Page 2: What does DAG stand for?
Page 2: What does STN mean?
Page 11: What does TDISSA stand for?
2.- It seems to me that in general the equations are written with a font size that does not correspond, it is much larger than the one used by the authors in the development of the text.
3.- Improve the resolution of figures 1 and 2, is figure 1 your own or was it taken from another place? If figure 1 was extracted from another place, it must be properly referenced.
4.- I don't understand the use of the dot product in equations 13, 14, 15 and 16 because all the values that are involved with scalars are not vectors.
They should review the English in greater detail, there are inaccuracies throughout the text, but they are minor. A finer reading or perhaps requesting help from an interpreter would suffice.
Author Response
word file has been uploaded
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
- Line 148 .. ?
This can be accepted but Still figure 1 looks like a simple client/server communication instead of considering to satellite edge network if taking some references:
1. Sensors | Free Full-Text | SatEC: A 5G Satellite Edge Computing Framework Based on Microservice Architecture (mdpi.com)
2. (PDF) Dynamic Game-Based Computation Offloading and Resource Allocation in LEO-Multiaccess Edge Computing (researchgate.net)
3. https://d3i71xaburhd42.cloudfront.net/130c971e3b37370e8b5359a8cb59466c481d51e5/2-Figure1-1.png
maybe the figure as an use case can be improved.
need more refinement
Reviewer 4 Report
The authors try to do my concerns. This paper can be accepted now.
It is acceptable.
