11 KiB
11 KiB
type:: REVIEWS
tags:: #TEACHING/SE4AS
year:: 2024
venue:: STAF
full-title:: Model-Driven Development and Container Orchestration for Self-Adaptive Cloud Applications
date-start:: 04-01-2024 - 21:42
date-submitted:: 11-01-2024
external-links::
status:: DONE
deadline-submission:: 10-01-2024
file:: 
- ### [[Highlights]]
- #+BEGIN_IMPORTANT
((659d5a66-0cfc-477c-8d7c-4c69b11e6e92))
((659d5a82-edb3-472f-b006-31da1b1f2d7c))
((659f1d5e-fa08-4678-8add-af54e2c4f9b4))
#+END_IMPORTANT
- ((659fa362-09d1-437a-9ab7-79cc7faf2d56))
- What kind of MDE approach and toolchain are you referring to? Here it is too general!
- ((659fab7b-5931-473b-809c-8561ee2dcdfc))
- "This MDE approach"? It needs clarification
- ((659fab95-ffe9-4618-a000-408d7b733c77))
- The same comment applies also to RQ3
- ((659fb3de-ca58-45eb-b2ef-6f29422b49c7))
- Run-time changes are for changing the mode of operation (or mission types)? This needs to be clarified at this stage.
- It is mentioned here ((659fb608-6c2b-455c-b36f-110df91aee3c))
- It needs to be clarified earlier (Check if it has not been done already)
- ((659fb4c7-e987-485e-b735-7660754ae0cc))
- These are the parameters that are exposed to change the beaviour of UAVs
- altitude: {low,high}
- formation: {loose, tight}
- ECM: {on,off}
- It is not clear what are the mentioned electronic countermeasures that can be activated to decrease the probabiilty of being destroyed by threats.
- For instance:
- *Conservative strategy*: ((659fb550-7bc0-4fc0-ad34-c13964a198f0))
- *Aggressive strategy:* ((659fb56d-b298-48ac-b4bb-db52c53d7a8e))
- *Balanced strategy*: ((659fb5a5-1bc7-4ad9-9707-8b537acf2c58))
- Aftear reading **section 5** ( ((659fbb24-39c1-4c75-8f3b-deee7247db0a)) ), it contains a set of guidelines, but it is not clear how they are actually implemented.
- ((659fb685-5d3e-4c43-9d1d-dc07c05d11eb))
- By referring to the example presented in Sec 4, it is not clear why containerization is needed and how it is employed. Readers can do some gusses, but it's better to explicitely clarify these aspects.
- By referring to the relevant papers in the field of software engineering for adaptive systems I would recommend to characterize the proposed approach with respect to the different dimensions of self-adaptive systems as presented in (A survey on engineering approaches for self-adaptive systems - DOI: 10.1016/j.pmcj.2014.09.009). Moreover, I suggest also to explicitely mention what are the patterns that have been employed or that are supported by the proposed approach (see Home Software Engineering for Self-Adaptive Systems II Chapter
On Patterns for Decentralized Control in Self-Adaptive Systems - https://doi.org/10.1007/978-3-642-35813-5_4)
- ((659fb995-31d2-4c7e-b9a5-90f0458f1ef0))
- What are the modeling constructs that you had to add to enable / introduce adaptation capabilities of the system?
- ((659fb99b-6cc8-4b2b-aae4-daa32346dea7))
- Model transformation for what?
- ((659fba0d-705e-4489-a69d-b62805e20e51))
- This means that the system under development, to be managed by the adaptaption mechanisms of the proposed approach, needs to expose and Adaptation manager? It is necessary to present the software architecture that needs to be employed by the systems that can be managed by the approach. Only by looking at the sequence diagram shown in Fig. 10, it is possible to guess what are the needed components even though it is not clear what are those that are "application dependent" and those that instead are needed to enable the application of the proposed approach.
- ((659fbad5-74ba-4675-9f0d-80cdf67c61c8))
- How to enforce this? For instance, do you have sets of interfaces that need to be developed to meet some software architecture?
- ((659fbb7d-b34b-4787-b97e-4651b0079bdc))
- Where should this implementation be located / done?
- ((659fbdee-a82a-4cce-bebe-d60c5b46c8b5))
- What are the source and target metamodels underpinning such transformations?
- ((659fbe17-a647-4ff7-a149-4ae19ad3c718))
- It is easy to understand this step without having clarified first what is the architecture of the systems that the approach support. What are the software components that the system provides? What are the software components that the developer has to implement? What are the software constraints (including the protocols and communication mechanisms) that need to be satisfied?
- ((659fc08a-b07e-4aa2-9bc5-50b862229a1b))
- This is a too far reference!
- Concerning the evaluation
- ((659fc282-decf-4f3c-914a-9dd0b55c772c))
- Research question 1 described in Sec 7.1 is different than that listed in Section 3. I like more the one in Sec 7.1 even though it is not clear what are the metrics that are used to judged if the obtained results are good or not. THe authors present 4 different examples, giving more emphasis to the UAV example, even though it is not clear how to evaluate them with respect to the research question definitions previously given.
- Research question 2 (ie., the failure capabilities of the systems developed with the proposed approach) is affected by problems similar of those of RQ1. What are the contributions of the authors in addition of those already provided by existing container management platforms?
- Concerning research question 3, it is not possible to appreciate the extendibility mechanism of the proposed approach, because it has not been presented earlier in the paper. As previously mentioned, Fig. 10 is the first figure that shows some components that might think of those that are part of the proposed framework. However, it is not clear if some of those components are application dependent.
- To summarize, the paper is about an interesting and relevent problem. However, the paper is affected by severe presentation issues that affect the overal paper quality. I suggest the authors to restructure the paper with the aim of clarifying the least the following aspects:
- What are the ingredients of the proposed approach
- What are the peculiar modeling constructs that you defined / used for enabling the specification and management of self-adaptive systems?
- What are the peculiar software components of the proposed approach that enable the management of self-adaptive capabilities of the supported systems
- Unfortunately, as you can see from the comments above, the current organization of the work is confusing by making the understanding of the strenghts and limitations of the proposed approach difficult.
- ### [[Comments]]
- The paper introduces an MDE approach and toolchain designed for adaptive containerized applications, focusing on leveraging the failure recovery capabilities of container management platforms.
-
- The paper addresses an interesting and relevant problem. However, it suffers from significant presentation issues that impact the overall quality. I would recommend restructuring the paper to address, at the very least, the following aspects:
- Clarify the ingredients of the proposed approach.
- Specify the modeling constructs used for enabling the specification and management of self-adaptive systems.
- Explicitly outline the peculiar software components enabling the management of self-adaptive capabilities in supported systems.
- Unfortunately, the current organization of the work is confusing, making it challenging to understand the strengths and limitations of the proposed approach.
- Sections 1, 2, and 4 are informative and well-written. However, other sections require substantial revision. For instance, Section 5 (Approach and Prototype Toolchain) offers guidelines on developing self-adaptive systems, but it needs clarification on how the provided toolchain supports these guidelines. Including a clear description of the software components comprising the toolchain would be beneficial.
- I recommend referencing relevant papers in the software engineering field for adaptive systems. Characterizing the proposed approach in alignment with the different dimensions of self-adaptive systems, as presented in "A survey on engineering approaches for self-adaptive systems - DOI: 10.1016/j.pmcj.2014.09.009," would enhance the paper. Additionally, explicitly mentioning supported patterns, as seen in "Software Engineering for Self-Adaptive Systems II Chapter On Patterns for Decentralized Control in Self-Adaptive Systems - [https://doi.org/10.1007/978-3-642-35813-5_4](https://doi.org/10.1007/978-3-642-35813-5_4)," is advisable.
- The recommendation to use the MAPE-K reference architecture or suitable variants lacks clarification on how such architectures are managed in the proposed approach. Specifically, it's unclear whether interfaces need to be developed to reflect these reference architectures.
- The need for developing suitable monitors is mentioned in several parts of the paper. However, understanding this step is challenging without first clarifying the architecture of the systems the approach supports. Information on the provided software components, components to be developed by the developer, and the necessary constraints and communication mechanisms would be valuable.
- Section 3, detailing research questions, requires revision. Research Question 1, as described in Sec 7.1, differs from the one listed in Section 3. While the Sec 7.1 version is preferable, it needs clarification on the metrics used to judge obtained results. Research Question 2, focusing on failure capabilities, faces similar problems and requires clarification on the authors' contributions beyond container management platforms. Concerning Research Question 3, it's impossible to appreciate the proposed approach's extendibility mechanism without an earlier presentation in the paper. Figure 10 provides some insight into components that might be part of the proposed framework, but it's unclear if they are application-dependent or part of the approach implementation.
-
- ### [[REVIEWS/Notes]]
- #TEACHING/SE4AS Invite Jurgen Dingel to deliver a seminar during the next edition of the SE4AS course.
- ((659d5ea0-d513-4c93-a9ce-5e7f84d7c74c)) #ideas #TEACHING/SE4AS Develop an example of adaptation by making use of [[kubernetes]]
- ((659f1d3a-c22e-44ce-93df-d46a50a86408))
- ((659fa2f7-8d95-4089-a572-be19cda8034c))
- That's interesting
- ### ❓️Questions
- {{query (and [[question]] [[STAF_2024_paper_8]] )[[question]]}}
query-table:: true
query-properties:: [:block]
- https://app.todoist.com/showTask?id=7492394553