Educational Resource Private Cloud Platform Based on OpenStack

Round 1

Reviewer 1 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

The authors have addressed the major concerns raised by the reviewers. However, some  revisions are needed to further clarify and strengthen the presentation of the research. These include enhancing the related work section, simplifying the model description, providing more detailed assumptions and justifications, and elaborating on the simulation setup. Furthermore, a case study is highly recommended to validate the usefulness of this architecture.

Author Response

Response to Reviewer Comments

We sincerely thank the Editor and the anonymous reviewers for their detailed, insightful and valuable comments. We have carefully considered the reviewers’ comments and revised the paper(computers-3166593-v3) accordingly. The revised contents are indicated in blue font in the modified version, and the main corrections in the paper are explained with red boxes. In the following, please find our responses to each reviewer’s comment and summary of revisions. For the ease of reading, we include the original review comments in italicized blue font in a box and our response follows.

Responses to Reviewer #1

1.The authors have addressed the major concerns raised by the reviewers. However, some  revisions are needed to further clarify and strengthen the presentation of the research. These include enhancing the related work section, simplifying the model description, providing more detailed assumptions and justifications, and elaborating on the simulation setup. Furthermore, a case study is highly recommended to validate the usefulness of this architecture.

Thank you very much for pointing out the shortcomings of the article and for continuous improvement of the article. 

We have strengthened the descriptions of related work as required (in the OpenStack with Ceph section), simplified the model descriptions (OpenStack and Ceph), and elaborated on the assumptions and reasons in more detail(in the Introduction section). At the same time, we conducted a case study with more detailed simulation settings to verify the usefulness of this architecture(in the Introduction section).

Thanks again!   

 

Author Response File: Author Response.pdf

Reviewer 2 Report (Previous Reviewer 1)

Comments and Suggestions for Authors

The article has significantly improved, with the authors addressing all highlighted comments and implementing the necessary improvements. The articulation and visuals have also seen enhancements. However, there are a few additional suggestions for further improvement:

 

1. There is still a need to clearly identify the key necessity for this platform. For instance, many cloud platforms are publicly available for educational purposes, most of which are based on OpenStack. Relevant examples include:
• Design and implementation of private cloud for higher education using OpenStack (https://ieeexplore.ieee.org/abstract/document/8363161)
• Designing and Applying an Education IaaS System based on OpenStack (https://www.naturalspublishing.com/files/published/m1f7372r3823na.pdf)
• Using OpenStack to improve student experience in an H.E. environment (https://ieeexplore.ieee.org/abstract/document/6661847)

 

If the proposed platform is significantly different and is an enhanced or improved version, the authors should introduce a comparative analysis section. This section should counter existing claims and include a Table highlighting how their work differs from and extends previous work, demonstrating how their results or solutions are superior.

 

 

2. It would be a significant contribution if the authors decide to make the platform publicly available for evaluation or to a global audience as open source in a repository or through an MDPI data journal.

 

3. Figures 6 and 9 are blurred, and the text is not visible. Please enhance the figure quality, including pixels and size, for better clarity.

Author Response

Responses to Reviewer #2

1.There is still a need to clearly identify the key necessity for this platform. For instance, many cloud platforms are publicly available for educational purposes, most of which are based on OpenStack. Relevant examples include:  

We are very grateful to the reviewers who have continuously improved the quality of our paper.

We have clearly explained the necessity of platform construction as required(in the Introduction section), and added relevant references as requested. Eg. Ref. 6, Ref. 17, Ref. 20.

Please refer to the page 1-2.

Thanks again!   

 

2. 2.If the proposed platform is significantly different and is an enhanced or improved version, the authors should introduce a comparative analysis section. This section should counter existing claims and include a Table highlighting how their work differs from and extends previous work, demonstrating how their results or solutions are superior.

Many thanks to the reviewers who made continuous improvement in the quality of the paper. We have explained the shortcomings of the platform and the differences from other platforms more clearly as requested. (in the Test and Analysis of the Private Cloud Platform for Educational Resources  section)

Please refer to the page 13.

Thanks again!   

   

3. 3.It would be a significant contribution if the authors decide to make the platform publicly available for evaluation or to a global audience as open source in a repository or through an MDPI data journal.

Many thanks to the reviewers for giving such good suggestions, we are considering making the platform publicly available for evaluation either as a global audience as open source in a repository or through an MDPI data journal. 

 

4. 4.Figures 6 and 9 are blurred, and the text is not visible. Please enhance the figure quality, including pixels and size, for better clarity.

Thank you very much for the reviewer's good suggestions. We have enhanced the quality of Figures 6 and Figures 9.

Thanks again! 

 

 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

The case study must contain the process from request input to output.

Author Response

Response to Reviewer Comments

We sincerely thank the Editor and the anonymous reviewers for their detailed, insightful and valuable comments. We have carefully considered the reviewers’ comments and revised the paper(computers-3166593-v4) accordingly. The revised contents are indicated in blue font in the modified version, and the main corrections in the paper are explained with red boxes. In the following, please find our responses to each reviewer’s comment and summary of revisions. For the ease of reading, we include the original review comments in italicized blue font in a box and our response follows.

Responses to Reviewer #1

1.The case study must contain the process from request input to output.

Thank you very much for pointing out the shortcomings of the article and for continuous improvement of the article.

We have strengthened the description process of high-performance node cases realization from request input to output according to requirements.

Please refer to the page 9-13.

Eg.:5.2. HA Implementation for Proxy Nodes

The implementation steps are as follows:

Update configuration /etc/nova/nova.conf；/etc/neutron/plugins/ml2/ml2_conf.ini;

Open vSwitch，configuring OVS，ovs-vsctl add-br br-eth1；ovs-vsctl add-port br-eth1 eth1; Restart service, systemctl restart nova-compute; systemctl restart neutron-openvswitch-agent; Verification configuration, systemctl status nova-compute; systemctl status neutron-openvswitch-agent.

5.3. HA Implementation for Control Nodes

In OpenStack, the HAProxy program provides load balancing for OpenStack Service APIs and MariaDB Galera services, with its own high availability implemented in an Active-Passive mode by Pacemaker+Corosync. Pacemaker is responsible for providing a Virtual IP (VIP) to HAProxy, ensuring that at any given time, only one HAProxy instance is in the Active state and providing services. The specific implementation steps are as follows:

Pcs resource create rip ocf: heartbeat: IPaddr2 params ip=”Server-VIP” cidr_netmask=”48” op monitor imerval=”30s”, Pcs resource create lb-haproxy systemd: haproxy-clone, Pcs constraint order start vip then lb-haproxy-clone kind=Optional, Pcs constraint colocation add vip with lb-haproxyclone, the Memcached component inherently supports an Active-Active configuration, requiring only the configuration of all its node names within OpenStack, for example, memcached_servers = controller01:11211, controller02:11211. In the event that controller01: 11211 fails, the OpenStack service components will automatically switch to using controller02:11211.

5.4. HA Implementation for Data Nodes

The implementation steps are as follows:

Install Cinder and Ceph on the data node，apt install cinder-volume ceph ceph-common; Configure Cinder to use Ceph as the backend，/etc/cinder/cinder.conf, volume_driver = cinder.volume.drivers.rbd.RBDVolumeDriver, volume_backend_name = ceph, rbd_pool = ZLC_volumes, rbd_user = ZLC_cinder, rbd_secret_uuid = <secret_uuid>, ceph_conf = /etc/ceph/ceph.conf, systemctl restart cinder-volume, journalctl -u cinder-volume.

5.5. HA Implementation for Compute Nodes

To achieve live migration of virtual machine instances using Ceph distributed storage, we leverage Ceph as the distributed file system to ensure high availability across compute nodes. With Ceph Monitor installed and configured on the Controller node, and Ceph-Client installed and configured on Compute nodes, we can build a HA implementation effect of the compute node in Figure 9. The specific configuration statements would be outlined as follows:

Live_migration_flag = VIR_MIGRATE_UNDEFINE_SOURCE, VIR_MIGRATE_

PEER2PEER, VIR_MIGRATE_LIVE, VIR_MIGRATE_TUNNELLED.

5.6. HA Implementation for Storage Nodes

The implementation steps are as follows:

Configure Cinder to use Ceph as a high-performance block storage backend, /etc/cinder/cinder.conf, enabled_backends = ceph, volume_driver = cinder.volume.drivers.rbd.RBDVolumeDriver, volume_backend_name = ceph, rbd_pool = volumes, rbd_user = cinder, rbd_secret_uuid = <secret_uuid>  # 从ceph auth get-key client.cinder, ceph_conf = /etc/ceph/ceph.conf, rbd_flatten_volume_from_snapshot = false, rbd_max_clone_depth = 5, rbd_store_chunk_size = 4. The HA implementation effect of the storage node is shown in Figure 10.

5.7. HA Implementation for Network Nodes

There are three ways to achieve Neutron L3 high availability: 1) Utilize Pacemaker+corosync to implement Active-Passive high availability; 2) For compatibility with high availability and scalability, employ VRRP-based Active-Active high availability; 3) Implement Neutron L3 high availability using Distributed Virtual Routing (DVR) introduced in the Juno release. With DVR, this solution deploys DHCP and SNAT only on the network control nodes, while deploying NAT and L3 Agent to the compute nodes where VMs reside. This simultaneously resolves the high availability issues of L3 Agent and Metadata Agent, with specific configurations outlined in Table 2 and the HA implementation effect of the network node is shown in Figure 11.

The nodes implementation are summarized in Table 2.

Thanks again!   

Author Response File: Author Response.pdf

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

It is the development of a private cloud platform using OpenStack and Ceph technologies to optimize the utilization and management of educational resources. On a broader level It aims to reduces the construction costs of investment in universities and improves the efficiency of teaching resource utilization. Here are some suggestions for improvements.

 

1.     The claims made in the introduction should be cited. For example, the line “OpenStack and Ceph technologies provide a revolutionary solution…” needs to be supported. How did you measure its revolutionary impact? If someone performed benchmarking and evaluations, they should be cited. 

2.     Similarly, the claim “the current open-source public education cloud resource databases lack accuracy and completeness, with weak privacy protection and data security…” should be validated and cited. Which open-source databases exactly lack these qualities? Please mention their names and explain how they lack accuracy and completeness. Adding more details in this context would be logical. 

3.     The introduction section should be extended by adding a problem statement and motivation. Additionally, more references addressing the same underlying problems should be included. Currently, it is relatively short and less effective in communicating the key problem and the need for the research. 

4.     There is a need to conduct a thorough literature review on this topic. Recent articles from the past 5 to 7 years should be evaluated to add credibility and justify the need for introducing the proposed solution.

5.     A comparative analysis of the proposed technique with existing similar solutions in the market should be added. A cross-comparison of the benefits and limitations should be conducted for a fair comparison.

6.     Finally, the authors should acknowledge the limitations of the proposed system and consider adding possible future research directions for the article.

 

Comments on the Quality of English Language

Minor editing of English language required.

Author Response

Response to Reviewer Comments

We sincerely thank the Editor and the anonymous reviewers for their detailed, insightful and valuable comments. We have carefully considered the reviewers’ comments and revised the paper(computers-3166593) accordingly. The revised contents are indicated in blue font in the modified version, and the main corrections in the paper are explained with red boxes. In the following, please find our responses to each reviewer’s comment and summary of revisions. For the ease of reading, we include the original review comments in italicized blue font in a box and our response follows.

Responses to Reviewer #1

1.the line “OpenStack and Ceph technologies provide a revolutionary solution…” needs to be supported.

Thanks for pointing this out. 

We have changed the description following your suggestion, have added the benchmarking and evaluations of Abbasi et al., and applied their results.

Eg: OpenStack and Ceph technologies provide present an open-source alternative for building virtual local or cloud setups that support petabytes of data, unlimited scale, and configurable networking[4]. The tool’s features make it an ideal solution for large scale virtualization, reducing maintenance costs and optimizing hardware resource utilization, particularly in education and government sectors.

Please refer to the page 1.

Thanks again!   

 

2.Similarly, the claim “the current open-source public education cloud resource databases lack accuracy and completeness, with weak privacy protection and data security…”  

Thanks for pointing this out. 

We have changed the introduction following your suggestion, verified and cited the introduction statements, submitted the database and briefly explained the reasons.

Eg:However, the open educational resources (OER) make it difficult to monitor the learning process, teachers might not be willing to apply OER because they might be afraid of losing control over the learning process[1]. For instance, Open source educational resources published by educator A on Repository X are difficult to guarantee their accuracy and completeness[2]. They also lack highly customizable, directly controllable resource allocation, resulting in low efficiency in the utilization of teaching resources[1-3].

Please refer to the page 1.

Thanks again!   

 

 

 

3. 3.Take a careful look at your bibliography and how you cite papers listed in it. Make sure it is current and cites recent work. Please cite a variety of different sources of literature. Please do not make excessive citation to arXiv papers, or papers from a single conference series. Do not cite large groups of papers without individually commenting on them. So we discourage "In prior work [1,2,3,4,5,6] …".Your bibliography should only exceptionally exceed about 40 items.

We are very grateful to the constructive comments.

We have carefully revised the introduction section according to the requirements. The introduction section has been expanded by adding a problem statement and motivation. We cite a total of 17 more references that address the same underlying problems.

Please refer to the page 1~2.

Thanks again! 

   

4. 4.There is a need to conduct a thorough literature review on this topic. Recent articles from the past 5 to 7 years should be evaluated to add credibility and justify the need for introducing the proposed solution.

Thanks for pointing this out. 

We have added a literature review on the topic as suggested, evaluated 40 recent articles from the past 5 to 7 years, and illustrated the need for a solution.

Eg: The related work content in Chapter 2 and the OpenStack and Ceph content in Chapter 3.

Please refer to the page 2~4.

Thanks again! 

 

5. 5.A comparative analysis of the proposed technique with existing similar solutions in the market should be added. A cross-comparison of the benefits and limitations should be conducted for a fair comparison.

Thanks for pointing this out. 

We have added the content of fair comparison according to your suggestion, and analyzed and compared the advantages and disadvantages of the proposed method.

Eg: The detailed content of fair comparison is in the test and analysis of the private cloud platform for educational resources in Chapter 6.

Please refer to the page 13~14.

Thanks again! 

 

 

 

 

 

 

 

6. 6.the authors should acknowledge the limitations of the proposed system and consider adding possible future research directions for the article

Thank you very much for giving such a good and constructive opinion.

We have revised it according to your request. The detailed contents are discussed in Chapter 6 and the conclusion in Chapter 7.

Please refer to the page 13~14.

Thanks again! 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This study seeks to develop a private cloud management platform specifically for educational purposes and to implement it in the management of teaching resources, thereby improving both the quality and efficiency of resource management. This paper outlines the development plan for a campus-specific teaching private cloud platform and details the construction process. It offers a solution for effectively managing and fully utilizing teaching resources. Unfortunately, this study is currently only a proposal and has not yet been implemented or tested for performance. However, this article does not meet the publication standards for a scientific journal for the following reasons:

-        Much of the content in this work has already been covered by existing research.

-        The proposed architecture lacks consistency across its different layers, particularly in terms of how communications occur and the software used for these interactions.

-        The authors claim that "response time for user requests can be reduced from 10 seconds to 2 seconds," but there is no evidence or simulation data provided to support this conclusion.

-        Overall, most of the recommendations and conclusions are based solely on the authors' estimates, lacking a solid scientific foundation.

Author Response

Response to Reviewer Comments

We sincerely thank the Editor and the anonymous reviewers for their detailed, insightful and valuable comments. We have carefully considered the reviewers’ comments and revised the paper(computers-3166593) accordingly. The revised contents are indicated in blue font in the modified version, and the main corrections in the paper are explained with red boxes. In the following, please find our responses to each reviewer’s comment and summary of revisions. For the ease of reading, we include the original review comments in italicized blue font in a box and our response follows.

Responses to Reviewer #2

1.Much of the content in this work has already been covered by existing research

Thanks for pointing this out. 

You are absolutely right that many of the contents in this work have been covered by existing research.

However, there are relatively few innovative application studies on the integration of OpenStack and Ceph, and it is a very worthy research topic. Therefore, based on the suggestions of reviewer #1, the revised manuscript comprehensively expounds the necessity of the solution in this paper.

Please refer to the page 1-2.  

Thanks again!

 

2. The proposed architecture lacks consistency across its different layers, particularly in terms of how communications occur and the software used for these interactions

Thanks for pointing this out.

We have improved the architectural design and strengthened the consistent design between the different layers based on your comments. The original scheme of this paper does miss the design and implementation of network nodes, and lacks the introduction of communication technology and interaction, which has been strengthened in the revised draft.

Eg: the network nodes content of the deployment architecture design in section 4.2 and the high-performance implementation for network nodes in section 5.4.

Please refer to the page 7-9.  

Thanks again!

 

3.The authors claim that "response time for user requests can be reduced from 10 seconds to 2 seconds," but there is no evidence or simulation data provided to support this conclusion 

Many thanks to the reviewer for pointing out this problem. 

The original draft of this paper really lacks relevant content introduction and comparison, and the revised draft adds test environment and test comparison introduction as suggested. The performance improvement in response time is mainly the result of comparing with OpenStack+Swift schemes.

Eg: Test server configuration table 1 is in deployment architecture design in Chapter 3.2, and test comparison results are in  Chapter 6 and Chapter 7.

Please refer to the page 6, 15-16.  

Thanks again!

  

4. 4. Overall, most of the recommendations and conclusions are based solely on the authors' estimates, lacking a solid scientific foundation.

Thank you very much for the problems pointed out by the reviewer.

The level of experts is very high, and it is very helpful to improve the quality of our manuscripts.

In the initial draft manuscript, we do not elaborate scientific experiments and related foundations, and lack the realization of renderings. Therefore, in the revised manuscript, we have strengthened the scientific basis and implementation effect presentation in accordance with the recommendations. 

Please refer to the page 9-13, Figure 6 -Figure 11.

Thanks again!

Author Response File: Author Response.pdf