A Trusted Multi-Cloud Brokerage System for Validating Cloud Services Using Ranking Heuristics

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This work proposes an intelligent cloud broker for validating cloud services. In my opinion, the authors should treat:

1. Abstract. In the last sentence, please mention the models against which the comparisons are done, resulting in improvements. Also, the data used.
2. Section 3. The definitions for p3 and p6 need discussion. What if the customer needs feature1 and feature 2, and the provided services are feature1, feature 3, and feature4. Would p3 be 1.5?
3. Please explain what SLA from eq 9 is. 
4. Maybe each pj should have another index, for each cloud. Table 2 shows that there are 10 different p1 for example. Therefore, eq 14 should compute SRi
5. Section 4. Please describe the volume, number of instances, and number of features for the dataset used in section 4.1.
6. Table 3 needs statistical analysis. The values seem close, to me. I wonder if the data from column 3 (proposed method) significantly differ from the results from previously published papers (next two columns).

Author Response

Dear Reviewer,

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Recently, cloud computing has become a major platform for distributed computing. The cloud service broker is required to view the services offered by cloud service providers. The topic of the proposed research is interesting.

The relevance of this article would be better to argue more convincingly, since the same text and the experimental study appeared on March 23rd, 2023: (PDF) An Intelligent Cloud Broker with Service Ranking Algorithm for Validation and Verification of Cloud Services in Multi-cloud Environment. Available from: https://www.researchgate.net/publication/369496688_An_Intelligent_Cloud_Broker_with_Service_Ranking_Algorithm_for_Validation_and_Verification_of_Cloud_Services_in_Multi-cloud_Environment

Posted Date: March 23rd, 2023, DOI: https://doi.org/10.21203/rs.3.rs-2546082/v1

The two texts are almost identical; therefore, misunderstandings may arise when citing. There are very minor differences (mainly editorial) between the two texts, and only in lines 165-222 are the 5 literary sources with studies from 2023 and 2024 additionally described [23-27].

New research should be done with new data to reflect the aforementioned developments, such as “the growing number of fog infrastructure providers; multi-Cloud and multi-Fog scenarios; computational overhead of training numerous agents; a way to manage many clouds at a lower cost; optimised load balancing techniques”, etc.

The most valuable part of the paper may be the 13 service trust factors. However, various trust assessment models have been proposed to evaluate trust in cloud service providers. These models often take into account performance monitoring and service availability. The justification for the proposed factors is missing. Objective and subjective trust analysis may also be considered. Service Response Time (SRT) is mentioned only for IaaS, what about other types of services?

The following questions would be good to be developed and explained:

• How are the numerical values for SR determined in Algorithm 1, "Service ranking and cloud service cluster formation"?
• Which WEKA algorithms were used, with which parameters?

The simulation parameters for the experiments should be shown.

More details about the types of new services would be interesting for readers.

The experiments are performed by WEKA as an existing machine learning (ML) tool. However, the proposed model is planned to involve machine learning algorithms as FUTURE work. I recommend doing new experimental research.

Author Response

Dear Reviewer,

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors correctly treated the previous issues.