Review Reports - Agent-Based Virtual Machine Migration for Load Balancing and Co-Resident Attack in Cloud Computing

Round 1

Reviewer 1 Report

This paper presents an Agent-based Virtual Machine Migration for Load Balancing and Co-resident Attack in Cloud Computing, which are based on all areas of cloud computing. Thus, this paper is directly related to the theme of this journal.

Overall, the paper is organized properly; the concept and future research directions are extensively explained. So, the paper is accepted after the following minor changes:

1. Problem statement is given in the research aim and the problem statement in heading 3 but the problem statement must be part of the introduction without subheading

2. Figure 3 is not clear, redraw figures and provide clear images

3. Paper contains a few grammar mistakes which will be cooperated in the final version.

4. add a few references related to cloud computing, which are mentioned below

Laghari, Asif Ali, Hui He, Asiya Khan, Rashid Ali Laghari, Shoulin Yin, and Jiachi Wang. "Crowdsourcing platform for QoE evaluation for cloud multimedia services." Computer Science and Information Systems 00 (2022): 38-38.

He, H., Khan, A., Kumar, N., & Kharel, R. (2018). Quality of experience framework for cloud computing (QoC). IEEE Access, 6, 64876-64890.

Awais Khan Jumani, and Rashid Ali Laghari. "Review and State of Art of Fog Computing." Archives of Computational Methods in Engineering (2021): 1-13.

Kaishan Wu, Rashid Ali Laghari, Mureed Ali, and Abdullah Ayub Khan. "A Review and State of Art of Internet of Things (IoT)." Archives of Computational Methods in Engineering (2021): 1-19.

Author Response

Dear Reviewer,

We would like to thank you for your positive feedback on our manuscript, and we appreciate your valuable suggestions for improving the paper. We have addressed each point as follows:

Overall, the paper is organized properly; the concept and future research directions are extensively explained. So, the paper is accepted after the following minor changes:

Point 1: I Problem statement is given in the research aim and the problem statement in heading 3 but the problem statement must be part of the introduction without subheading

Response 1: Thank you for your feedback on the placement of the problem statement. We agree that the problem statement should be part of the introduction without a subheading. We have revised the manuscript to incorporate this change.

Point 2: Figure 3 is not clear, redraw figures and provide clear images

Response 2: We appreciate your feedback on the quality of Figure 3. We have redrawn the figure to improve its clarity and provide a clear image in the revised manuscript.

Point 3: Paper contains a few grammar mistakes which will be cooperated in the final version.

Response 3: We appreciate your feedback regarding grammar mistakes in the manuscript. We have thoroughly proofread the manuscript and corrected all the grammar mistakes we could find in the revised version.

Point 4: add a few references related to cloud computing, which are mentioned below

Response 4: Thank you for your suggestion regarding additional references related to cloud computing. We have reviewed the references you provided and included the relevant ones in the revised manuscript to improve the literature review section.

Once again, we appreciate your feedback and suggestions for improving our manuscript. We hope that the revised version meets your expectations, and we look forward to hearing back from you.

Sincerely,

Biao

Author Response File: Author Response.docx

Reviewer 2 Report

The problems described in the article are interesting and relatively uncommonly addressed in such a compilation of articles. The article discusses practical issues, and the solutions, due to their simplicity, can be successfully applied in practice.

The article lacks innovative elements from a scientific point of view, nevertheless, I consider it worth publishing. As for the identified problems:

- There is no reference to Table 1, which makes lines 218 to 245 unreadable.

- In many formulas (lines 218 to 233), the designation X and x with different indices are used, but this designation has a different meaning in each equation. Please use different symbols to indicate different meanings.

- Numbering should be introduced for formulas from lines 218, 226, and 227,

- In the "Related work" and "load balancing" section, the latest solutions in the field of load distribution in cloud computing are missing, especially those using artificial intelligence solutions, including neural networks. In my opinion, the article discusses the issue of load distribution rather than load balancing.

Author Response

Dear Reviewer,

We would like to thank you for your constructive feedback on our manuscript. We have carefully reviewed your comments and have addressed each point as follows:

The article lacks innovative elements from a scientific point of view, nevertheless, I consider it worth publishing. As for the identified problems:

Point 1: There is no reference to Table 1, which makes lines 218 to 245 unreadable.

Response 1: We apologize for the oversight of not referencing Table 1 in the manuscript. We have rectified this issue by referencing Table 1 in the text where it is mentioned and provided a detailed description of the table in the caption.

Point 2: In many formulas (lines 218 to 233), the designation X and x with different indices are used, but this designation has a different meaning in each equation. Please use different symbols to indicate different meanings.

Response 2: We appreciate the feedback regarding the use of the same symbols with different indices in the formulas. We have revised the formulas to use different symbols for different meanings, to avoid any confusion.

Point 3: Numbering should be introduced for formulas from lines 218, 226, and 227

Response 3: We agree with the suggestion of introducing numbering for the formulas in lines 218, 226, and 227. We have added numbering to these formulas to improve clarity.

Point 4: In the "Related work" and "load balancing" section, the latest solutions in the field of load distribution in cloud computing are missing, especially those using artificial intelligence solutions, including neural networks. In my opinion, the article discusses the issue of load distribution rather than load balancing.

Response 4: Thank you for pointing out the gap in our literature review regarding the latest solutions in load distribution in cloud computing. We also agree that the article focuses more on load distribution rather than load balancing. Currently, the primary approach for addressing load balancing problems is through the use of artificial intelligence. The reason for adopting a simple heuristic method for load distribution is primarily due to the constraints of the simulation platform. Furthermore, NetLogo was chosen as the simulation platform due to its utilization in a previous study on the overall reliability simulation evaluation of wireless sensor networks (WSNs). As such, we were somewhat limited in our selection of load distribution methods.

Once again, we appreciate your feedback and suggestions for improving our manuscript. We hope that the revised version meets your expectations and look forward to hearing back from you.

Sincerely,

Biao

Author Response File: Author Response.docx

Reviewer 3 Report

In the paper titled “Agent-based Virtual Machine Migration for Load Balancing and Co-resident Attack in Cloud Computing” authors propose VM live migration to resolve co-resident security issues in cloud computing. The article needs minor organization/presentation changes with the following points.

· The abstract states that VM live migration are used to low potential of CRA. While migrating VM, the data can also be at risk! Have authors studied the trade-off?.

· The VM migrations induce extra load on the cloud networking resources. Is it goeng to effect the performance of cloud based applications? SLA violations?

· The objective of the VM migration is not clear in the introduction. Is the objective load balancing OR security?

· What is agent based load balancing? The introduction must add details to the proposed method.

· The section 3.1 should be renamed. It contains some initial modeling of the problem.

· Some of the figures have very poor pixel quality such as figure 3.

· The algorithm of page 10 should be written in algorithmic format.

· Why NEtlogo is used for simulation? How is it suitable for the proposed strategy? Have nay articles listed in section 2 used this tool for performance evaluation?

· The organization of section 6 is poor and many subsections are utilized without numbering.

· Are SLA violations considered in this article? What are the results in terms of SLA violations?

· Some very relevant articles in the same direction are missing such as

SLA-Aware Best Fit Decreasing Techniques for Workload Consolidation in Clouds (IEEE Access)

Multi-factor nature inspired SLA-aware energy efficient resource management for cloud environments (Cluster Computing)

Author Response

Dear Reviewers,

We thank you for taking the time to review our manuscript and providing us with valuable feedback. We have carefully considered each of the points raised in your comments and have made the following changes to the manuscript.

In the paper titled “Agent-based Virtual Machine Migration for Load Balancing and Co-resident Attack in Cloud Computing” authors propose VM live migration to resolve co-resident security issues in cloud computing. The article needs minor organization/presentation changes with the following points.

Point 1: The abstract states that VM live migration are used to low potential of CRA. While migrating VM, the data can also be at risk! Have authors studied the trade-off?.

Response 1: The original intention of our research was to investigate the security risks of cloud computing. During our investigation, we discovered that live migration of virtual machines is one of the five types of mitigation methods that can be used to mitigate CRA. Given our specialty in reliability, we recognized that load balancing is an effective means to improve the reliability of distributed systems. This led us to the idea of using one method to solve both problems simultaneously. However, as you pointed out in your review, the activity of virtual machine migration itself can also bring security risks, which we did not thoroughly consider in our paper.

In light of your comments, we have recognized the importance of addressing these risks. We propose to incorporate the risk brought by migration into the method design as an additional consideration element. This modification will enhance our research and produce a more comprehensive and valuable article.

Point 2: The VM migrations induce extra load on the cloud networking resources. Is it going to effect the performance of cloud based applications? SLA violations?

Response 2: We agree that VM migrations can have an impact on cloud networking resources and subsequently on the performance of cloud-based applications. As previously stated, our initial objective was to tackle the two aforementioned issues concurrently. Nonetheless, our limited professional background precluded us from accounting for the network resource overhead and security risks inherent in the virtual machine migration process. In enhancing the study of virtual machine migration, we suggest incorporating two additional metrics, namely resource overhead and security risk, to enable a comprehensive analysis of the benefits and drawbacks of the process. Presently, research only takes into account the advantages of migration, which is inadequate.

Point 3: The objective of the VM migration is not clear in the introduction. Is the objective load balancing OR security?

Response 3: We have revised the introduction section to clearly state the objectives of our proposed VM migration strategy. We aim to achieve both load balancing and co-resident attack prevention through the use of an agent-based approach.

Point 4: What is agent based load balancing? The introduction must add details to the proposed method.

Response 4: We have added more details to the introduction section to explain the concept of agent-based load balancing and how it is employed in our proposed strategy.

Point 5: The section 3.1 should be renamed. It contains some initial modeling of the problem.

Response 5: We have renamed section 3.1 to better reflect its contents.

Point 6: Some of the figures have very poor pixel quality such as figure 3..

Response 6: We apologize for the poor quality of some of the figures. We have replaced figure 3 with a higher quality version and have reviewed all other figures to ensure they meet the required quality standards.

Point 7: The algorithm of page 10 should be written in algorithmic format.

Response 7: We have rewritten the algorithm presented on page 10 in algorithmic format for clarity.

Point 8: Why NetLogo is used for simulation? How is it suitable for the proposed strategy? Have nay articles listed in section 2 used this tool for performance evaluation?

Response 8: To clarify our selection of NetLogo, it was chosen because it was utilized in one of my doctoral dissertation studies to assess the reliability of wireless sensor networks (WSNs). In addition, NetLogo was employed as a simulation tool in both the perception and network layers of WSNs for another study. Thus, NetLogo was also used for cloud computing (though this explanation is beyond the scope of this paper). While NetLogo may lack the professional specificity of other cloud computing simulation platforms, its versatility remains its strength.

Point 9: The organization of section 6 is poor and many subsections are utilized without numbering.

Response 9: We have revised section 6 to improve its organization and have numbered all subsections for clarity.

Point 10: Are SLA violations considered in this article? What are the results in terms of SLA violations?

Response 10: Our paper neglected to address both the violation of service level agreements (SLAs) and the consequential impact of such violations, only briefly mentioning it in the context of our CRA mitigation measures. It was only when researching related studies that we became aware of this oversight. Introducing relevant research should include both the positive and negative aspects of the topic at hand. That's why we mentioned the SLA in the related work section.

Point 11: Some very relevant articles in the same direction are missing such as.

Response 11: We have reviewed the literature and have added the relevant articles to the reference list.

Once again, we thank you for your helpful feedback and hope that the revised manuscript meets your expectations.

Sincerely,

Biao

Author Response File: Author Response.docx