Automated Quantum Hardware Selection for Quantum Workflows

Round 1

Reviewer 1 Report

This manuscript proposes a novel quantum hardware selection method.

It will help a lot when we consider the quantum algorithm, quantum computing, quantum communication. And the manuscript is well-written.

So I will accept the manuscript after some following concerns are resolved.

(1) I think the term Hardware is not exactly used. The term should be used here is Quantum hardware simulator or emulator. So author should correct with perfect term.

(2) The organization of manuscript is not good. The "Case study" and "Related work" should be merged to be one section, or will be the subpart of "System architecture..." part.

(3) Refer to Fig. 1, the icon of Quantum circuit loading task and Quantum circuit object are same. Kindly check them and correct if they are different.

Many congratulations to research team for a quality paper.

BRs,

mducng

Author Response

Thank you very much for the review. Please find a point-by-point list of how we adapted the manuscript regarding your suggestions attached.

Author Response File: Author Response.pdf

Reviewer 2 Report

This work proposes a quantum workflow, which works independently of the quantum computer performing the input task. The proposed solution allows to ease the problem of choosing a particular quantum computer for a concrete task, making this choice transparent for the user.

They enable automated quantum hardware selection by claiming the following contributions:
- Separate the quantum workflows and the quantum computers implementing them.
- Instead of deploying all services before workflow execution, they are deployed on-demand during execution.

With this aim, they build on two previous works of the authors: QuantME and NISQ Analyser. In particular, they extend QuantME with the Quantum Hardware Selection subprocess, and deal with the problem of its correct management during execution, being able to deploy the concrete required workflow fragments on-demand.

Finally, they showcase the proposed solution for the implementation of the Simon's algorithm, and the quantum computer is dynamically selected during workflow execution.

In my opinion, the work fits well this journal and the contribution is interesting. However, I think that the main novelty of their work is not clear when comparing to the state of the art. I also have some concerns and recommendations regarding the exposition of their results. I detail them below:

- The authors provide a very detailed description of related works, but they do not highlight which are the differences of their contribution with respect to them. So, as it is right now, it is just a description without comparison.

I would recommend to clearly state which are the authors' contributions while comparing with these works.

- I think it would be better to move this related work section to the beginning in order to showcase and better motivate the authors' contributions.

- Some parts of the manuscript describe a sequential order of steps in the workflow, but they are included in the text as consecutive sentences in the same paragraph (mainly in section 3). This sometimes makes hard its reading, so I recommend to better separate them, for example as different bullet points.

- Page 11, line 429: "However, due to space reasons, it is represented as a collapsed subprocess". I think there is no space limit in this journal, so the authors can include that description in another figure.

As a minor note:
- I would recommend to make a general pass through the document just to fix some typos and sentences which are confusing. For example: page 5, line 220, can not -> cannot; page 5, line 210: "independent of a quantum computer to use".

Author Response

Thank you very much for your review. Please find a point-by-point list attached of how we adapted the manuscript regarding your suggestions attached.

Author Response File: Author Response.pdf

Reviewer 3 Report

The manuscript is devoted to the valid problems related to one of the ``hot" contemporary topics – quantum computations. In particular, the Authors proposed a novel automated quantum hardware approach. They present an idea of the system that would be able to determine which kind of quantum computers and quantum circuits are the most suitable for solving a given problem. What is the most relevant, the proposed approach allows for dynamic adaptation of the workflows. Such adaptation should depend on the selection done at runtime and based on reusable workflow fragments. Finally, the Authors show how to implement their ideas in solving a particular problem that is related to the application of Simon’s algorithm. The Authors’ proposals seem to be innovative. They are relevant from the theoretical point of view and, additionally, can lead to novel practical realizations of the systems based on the quantum computing theory.

The article is well written in general. Besides the presentation of the proposed ideas, it gives a thorough introduction to the problems discussed in the manuscript giving a comprehensive list of references. Additionally, the Authors provide a proposal of future works that could be a continuation of that shown in the article. The paper is written in an instructive way and is accessible even for the Readers who are not necessarily experts in the field.

In conclusion, I can state that the manuscript fully deserves to be published and can be accepted in its present form.

Author Response

Thank you very much for your review. I don´t think there are any revisions required regarding this review.

Reviewer 4 Report

Research topics related to the development and use of quantum computing have attracted a lot of attention lately. The authors of the article touch upon a rather relevant topic of using the resources of quantum computers. In this case, it is not simulated, but real-life quantum computers provided by IBM and Righetti are meant. The work was prepared quite qualitatively and it uses the previous developments of the authors, for example, QuantME. The paper has an up-to-date and relevant list of sources. In order to increase interest in the article, I propose to clarify a few points.
1. It is better to explain how the approach proposed by the authors takes into account the real rather than model characteristics of quantum computers?
2. Does the proposed approach take into account the complexity of the algorithm that is supposed to be executed on a quantum computer? Are there limits on the complexity of algorithms that make quantum computing impractical?
3. What the database includes with Deployment Models in Fig. 5.?
4. How is the process of executing a task on a quantum computer organized, including the transfer of input parameters for executable algorithms?
These comments are not critical issues. In general, the article reflects qualitative scientific research in the field of information systems and quantum computing. 

Author Response

Thanks for your review. Please find a list with changes related to your suggestions/questions attached.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

I thank the authors for their answers to the questions from previous round. Actually, I think they have successfully addressed all my questions and recommendations:

- Clarifying the novelties when comparing to the state-of-the art.

- Moving the related work section to the beginning of the document.

- Adding bullet points in Section 3.2 and the description of Figure 7 is perfectly OK.

- Fixed mentioned typos.

In my opinion, the manuscript is ready for acceptance.