Service Migration Strategy Based on Multi-Attribute MDP in Mobile Edge Computing

Round 1

Reviewer 1 Report

1. The biggest problem with this paper is language and style. When reading the work, I often did not know what the author meant. The document's formatting is also unacceptable, e.g., some fragments are repeated (see abstract).

2. The literature cited in the paper is insufficient - 7 items are definitely not enough for such work.

 

 

3. The experiments presented in the paper concern the migration of one user - I think it is worth examining the behavior of the proposed mechanisms in the case of several users (this adds credibility to the results obtained and is closer to reality)

4. The authors do not consider the errors during the migration (e.g., in the case of exhaustion of resources during the migration itself or its interruption during the migration process). Such an aspect is essential in the process and affects the continuity of the service used by the user.

 

Author Response

Dear Reviewers and Editors:

    First of all, thank you very much for taking time out of your busy schedule to read and revise my paper. Thank you for your valuable advice. You have revised the structure, content, research methods and results of my paper in all aspects, which will play a very important role in improving the quality of my paper.

    I have carefully studied the reviewer's comments and made modifications to the paper one by one according to the suggestions. The specific modifications are shown in the attachment.

    Finally, thank you again for your guidance, and thank you for reviewing and revising my revised paper again. I hope that I can finish an excellent paper with your guidance and help, and sincerely hope that my paper can be published in your journal.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors of this paper studied the service migration problem in the mobile edge computing environment. Specifically, they focused on the mobility of one single user and migrated the user's service across different edge nodes according to the predicted motion trajectory. They considered resource demands, available resources on edge nodes, migration overheads, migration revenue, etc., to collectively determine the most suitable migration strategy by multi-dimensional Markov decision process (MDP). A comprehensive evaluation was conducted to investigate the effectiveness of the proposed approach. Overall, this paper studies a timely problem. The paper is well-organized and easy to follow. However, there are many concerns that need to be further addressed.

1. The problem formulation is relatively fair. It is better to provide more details of those parameter definitions, such as the compute resource C_a, C_bi, and latency T_bi. For example, which factors impact the value of a given network latency and how to calculate it is not clear.

2. Nearly all equations have no labels.

3. It is better to refer to an equation or a figure by its name, like "From Fig. 1 we can see that..." is better than "from next figure we can see that...".

4. This paper uses straight movement to model/calculate the gains/revenues. However, what happens if the user is snaking/walking around? This should be further discussed.

5. The authors should provide more precious definitions of Motion Gain, Movement Gain, Communication Distance, Total Return, Campaign Gain, etc., and explicitly if a greater value is better or a smaller value is better.

6. The authors need to provide a more quantitative comparison to help readers understand the advantage of your approach.

7. It is better to summarize the used metrics before reporting the experimental results.

8. The related work needs to be further extended. Many key aspects are missing. For example, considering that this paper studies the service migration problem in the MEC environment, it is useful to discuss the recent outcomes in service/application deployment.

READ: Robustness-oriented edge application deployment in edge computing environment, 2020.

Joint Shareability and Interference for Multiple Edge Application Deployment in Mobile-Edge Computing Environment, 2021.

Joint coverage-reliability for budgeted edge application deployment in the mobile edge computing environment, 2022.

9. The paper needs to be further polished. There are many grammar issues and typos. 

Taking the abstract as an example, there are several redundant sentences, like "In addition, in order to prevent users from migrating too frequently and affecting the", and "In order to solve the problem of service interruption caused by user movement and limited service range of edge nodes, a service migration". 

Many sentences are not complete, for example, "at the beginning of each time slot a? state detection and make a service migration decision" on page 6, and "The payoff of" on page 11.

The authors mix-used the USER and VEHICLE, sometimes they said "the vehicle..." on page 5 even if it is a user (human) in figure 2.2.

The first word in a sentence should be capitalized, similarly, there will be a full stop after a sentence.

Many sentences are too long to follow. It's better to convert them into different short sentences.

 

Author Response

    Dear Reviewers and Editors:

    First of all, thank you very much for taking time out of your busy schedule to read and revise my paper. Thank you for your valuable advice. You have revised the structure, content, research methods and results of my paper in all aspects, which will play a very important role in improving the quality of my paper.

    I have carefully studied the reviewer's comments and made modifications to the paper one by one according to the suggestions. The specific modifications are shown in the attachment.

    Finally, thank you again for your guidance, and thank you for reviewing and revising my revised paper again. I hope that I can finish an excellent paper with your guidance and help, and sincerely hope that my paper can be published in your journal.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

All my remarks are addressed. Thank you!

Author Response

Dear Reviewers and Editors,

    First of all, thank you very much for taking time out of your busy schedule to read and revise my paper. Thank you for your valuable advice. You have revised the structure, content, research methods and results of my paper in all aspects, which will play a very important role in improving the quality of my paper.

    Thank you for your advice. I have carried out a detailed English spelling check for this paper, and I ask my classmates who are good at English to help me polish it. In addition, I translated the Chinese documents in the references into English.

    Finally, thank you again for your guidance and for reviewing and revising my revised paper. I hope to finish an excellent paper under your guidance and help, and I sincerely hope that my paper can be published in your journal.

Reviewer 2 Report

The authors have addressed all my concerns. The only minor issue is that the reference published in Chinese should be translated into English.

Author Response

Dear Reviewers and Editors,

    First of all, thank you very much for taking time out of your busy schedule to read and revise my paper. Thank you for your valuable advice. You have revised the structure, content, research methods and results of my paper in all aspects, which will play a very important role in improving the quality of my paper.

    Thank you for your advice. I have carried out a detailed English spelling check for this paper, and I ask my classmates who are good at English to help me polish it. In addition, I translated the Chinese documents in the references into English.

    Finally, thank you again for your guidance and for reviewing and revising my revised paper. I hope to finish an excellent paper under your guidance and help, and I sincerely hope that my paper can be published in your journal.