Towards Resource-aware Business Process Development in the Cloud

Tracking #: 765-1975

Authors: 
Emna Hachicha
Walid Gaaloul

Responsible editor: 
Guest Editors EKAW 2014 Schlobach Janowicz

Submission type: 
Conference Style
Abstract: 
In recent years, cloud environments are becoming more and more interesting and useful for the execution and the deployment of business processes. Indeed, it enables organizations to reduce their costs and optimize their processes. Many researches have been realized for providing support and enhancement to the resource perspective in business processes. Nevertheless, they have basically focused on human resources and have neglected other types of resources. This paper fills this gap by proposing an extension to the BPMN metamodel in order to optimally manage resources deployed in the cloud through resource constraints verifi cation. The purpose of our approach is to enable Business Process development to benefi t from economies of scale, faster provisioning times, decreased runtime costs, and reduced energy consumption. To do so, we aim at enriching Business Process Models with a semantic knowledge base about the consumed cloud resources that can be used to optimize resource management.
Full PDF Version: 
Tags: 
Reviewed

Decision/Status: 
[EKAW] reject

Solicited Reviews:
Click to Expand/Collapse
Review #1
Anonymous submitted on 25/Aug/2014
Suggestion:
[EKAW] reject
Review Comment:

Overall evaluation
Select your choice from the options below and write its number below.

== -1 weak reject

Reviewer's confidence
Select your choice from the options below and write its number below.

== 3 (medium)

Interest to the Knowledge Engineering and Knowledge Management Community
Select your choice from the options below and write its number below.

== 4 good

Novelty
Select your choice from the options below and write its number below.

== 2 poor

Technical quality
Select your choice from the options below and write its number below.

== 2 poor

Evaluation
Select your choice from the options below and write its number below.

== 3 fair

Clarity and presentation
Select your choice from the options below and write its number below.

== 3 fair

Review
Please provide your textual review here.

The paper provides an extension of the BPMN metamodel in order to manage resources deployed in the cloud through resource constraints verification. In particular, it proposes an extension to the BPMN notation to incorporate the notion of resource (section 6) and then an approach to encode properties to be verified in section 8, plus a signavio plugin for their enforcement.

The paper deals with an extremely interesting problem, that is the verification of resource constraint verification, pointing out that new challenges can arise due to the development of cloud computing. It is clearly written and organized in a logical fashion. Nonetheless, the paper fails to convince due to the following problems:

Major comments
--------------

- first of all the paper fails to provide any evidence on why resource allocation issues arising in a cloud environment should be treated differently from more "traditional" resource allocation challenges. I suggest the authors to point out at least a single example / issue / constraint *in practice* that could not be modelled and verified with current approaches for extending BPMN with the capability to verify resource allocation constraints.

- related to the comment above, the authors should make a better effort to relate to existing work in the field of verifying resource aware BPMN mopdels. In particular, the authors neglect an existing stream of work for the verification of BPMN models extended with resources. Examples are:

* Watahiki, K. Ishikawa, F. ; Hiraishi, K.
Formal verification of business processes with temporal and resource constraints
Systems, Man, and Cybernetics (SMC), 2011 IEEE International Conference on
URL: http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6083857

* Christian Wolter, Andreas Schaad. Modeling of Task-Based Authorization Constraints in BPMN
5th International Conference, BPM 2007, Brisbane, Australia, September 24-28, 2007. Proceedings
Lecture Notes in Computer Science Volume 4714, 2007, pp 64-79
URL: http://link.springer.com/chapter/10.1007/978-3-540-75183-0_5

- The embedding of the CouldPro Ontology and of the BPMN processes is not completely clear. As far as I know BPMO is inspired by BPMN entities but does not enable the representation of specific BPMN diagrams. Therefore it is not clear to me how the approach presented in this paper is able to check constraints on a specific BPMN model. related to this, the authors should provide a comparison with the approach of checking BPMN constraints (even if not resource specific constraints) presented in

Semantically-aided business process modeling (Chiara di Francescomarino, Chiara Ghidini, Marco Rospocher, Luciani Serafini, Paolo Tonella), In 8th International Semantic Web Conference (ISWC 2009), Springer Berlin / Heidelberg, volume 5823/2009, 2009.

and with its underlying BPMN ontology published in

An ontology for the Business Process Modelling Notation (Marco Rospocher, Chiara Ghidini, Luciano Serafini), In Proceedings of the 8th International Conference on Formal Ontology in Information Systems (FOIS 2014), 2014.

- Section 8 provides a high level definition of classes of properties to be verified, but nothing is said on the specific types of expressions that can be actually written, their computational properties and whether the RDF/RDFS language is expressive enough to capture an adequate number of concrete expressions (constraints) that have to be validated in the cloud setting. I suggest the authors to be more concrete in this, and provide an overview of the types of properties that the framework is able to express and verify.

- in section 8 the authors introduce simple and compplex rules. What is the reason for that instead of simply introducing (complex) rules which concern, as a particular case, only one resource? Does this distinction have concrete impacts or is there a specific reason to consider the difference?

Also, Figures 6 (a) and (b) should be increased in size.

Review #2
Anonymous submitted on 26/Aug/2014
Suggestion:
[EKAW] reject
Review Comment:

Overall evaluation
Select your choice from the options below and write its number below.

== 3 strong accept
== 2 accept
== 1 weak accept
== 0 borderline paper
== -1 weak reject
== -2 reject
== -3 strong reject

-2

Reviewer's confidence
Select your choice from the options below and write its number below.

== 5 (expert)
== 4 (high)
== 3 (medium)
== 2 (low)
== 1 (none)

4

Interest to the Knowledge Engineering and Knowledge Management Community
Select your choice from the options below and write its number below.

== 5 excellent
== 4 good
== 3 fair
== 2 poor
== 1 very poor

2

Novelty
Select your choice from the options below and write its number below.

== 5 excellent
== 4 good
== 3 fair
== 2 poor
== 1 very poor

3

Technical quality
Select your choice from the options below and write its number below.
== 5 excellent
== 4 good
== 3 fair
== 2 poor
== 1 very poor

3

Evaluation
Select your choice from the options below and write its number below.
== 5 excellent
== 4 good
== 3 fair
== 2 poor
== 1 not present

2

Clarity and presentation
Select your choice from the options below and write its number below.
== 5 excellent
== 4 good
== 3 fair
== 2 poor
== 1 very poor

2

Review
Please provide your textual review here.

This paper presents an approach for business process management (BPM) that integrates the "semantic" representation of cloud resources. In this sense, this paper does not really contribute to the semantic web and knowledge engineering areas, but mostly present an application of their technologies and principles. I have to admit that I am very far from familiar with current research in BPM, so I can't assess the claims of novelty in this paper related to this area. For what concerns the area of knowledge engineering however, as mentioned above, I don't think there is much that can be claimed as novel here. It is certainly an interesting application, but as far as I can see, the connection to the topics of EKAW are mostly that the authors created an ontology using RDF/RDFS and SWRL. The methodology to create this ontology is not really clear, the benefits of doing it this way (rather than using other approaches to the representation of cloud resources) is not very explicit, and there is not real evaluation of the knowledge engineering results (the ontologies and rules) in the proposed scenario and beyond. It would have in particular been good to discuss the possible reuse of other ontologies in similar/related domains, and how the one created could be reused, or whether the experience of building it could provide valuable insight to the EKAW audience.

Review #3
Anonymous submitted on 29/Aug/2014
Suggestion:
[EKAW] reject
Review Comment:

Overall evaluation
Select your choice from the options below and write its number below.

== 3 strong accept
== 2 accept
== 1 weak accept
== 0 borderline paper
== -1 weak reject
== -2 reject
== -3 strong reject

-1

Reviewer's confidence
Select your choice from the options below and write its number below.

== 5 (expert)
== 4 (high)
== 3 (medium)
== 2 (low)
== 1 (none)

4

Interest to the Knowledge Engineering and Knowledge Management Community
Select your choice from the options below and write its number below.

== 5 excellent
== 4 good
== 3 fair
== 2 poor
== 1 very poor

3

Novelty
Select your choice from the options below and write its number below.

== 5 excellent
== 4 good
== 3 fair
== 2 poor
== 1 very poor

3

Technical quality
Select your choice from the options below and write its number below.
== 5 excellent
== 4 good
== 3 fair
== 2 poor
== 1 very poor

3

Evaluation
Select your choice from the options below and write its number below.
== 5 excellent
== 4 good
== 3 fair
== 2 poor
== 1 not present

2

Clarity and presentation
Select your choice from the options below and write its number below.
== 5 excellent
== 4 good
== 3 fair
== 2 poor
== 1 very poor

3

Review
Please provide your textual review here.

What methodology was used to develop the CloudPrO ontology? How did you gather requirements?
Why did you chose to include the classes and properties illustrated in Fig 5., only those and no others? There are no properties attached to the Storage, Compute and Network classes. Why?
For the properties defined in the ontology please specify domains, ranges, constraints if any.
In order to motivate and justify why the CloudPrO ontology has the shape proposed in the paper the previous questions need answers.

Is the CloudPrO ontology available on-line? If yes please specify where. Add a pointer to its RDF/RDFS representation.

The rule formalization in section 8.2 has resulted in all the three different types of rules (simple, complex and dependency-based) having the same format. I suggest defining them in more details and pointing the differences between them. Concrete examples of such rules could help the reader to better grasp the difference between them.

Section 9 needs more improvements. A tool plugin that uses the CloudPrO ontology in business process modelling is Ok as a mean to support the validation of the approach but more is needed. I would have expected a set of experiments using the plugin to be performed by business people and/or engineers involved in the modelling business processes and to validate the ontology in this process. How many rules, of which type and complexity
have been used in the validation process?

Several figures in the paper are not readable, for example Fig 3. and Fig 6.

Review #4
Anonymous submitted on 29/Aug/2014
Suggestion:
[EKAW] reject
Review Comment:

Overall evaluation == -1 weak reject
Reviewer's confidence == 3 (medium)
Interest to the Knowledge Engineering and Knowledge Management Community == 3 fair
Novelty == 3 fair
Technical quality == 2 poor
Evaluation == 1 not present
Clarity and presentation == 2 poor

Review
The paper describes the CloudPro ontology as an extension to the Business Process Modelling Ontology, which describes a subset of the Business Process Model and Notation. Additionally, the paper proposes a set of rules to constrain resource allocation/consumption in business process models. The paper is motivated from a need to model business processes that involve services that are hosted in a cloud. Therefore, the CloudPro ontology focusses on resources that are available via cloud services.
The paper aims to address a very interesting problem, the development of consistent business process models across distributed services and different ownerships. Semantic heterogeneities within such complex landscapes are identified as a problem.

However, while I can see that bringing together Business Process Modelling and Cloud Computing is of value, and that semantics and ontologies can play a central role in bringing these two together, the presentation of the contribution remains too shallow to be convincing.

The CloudPro ontology is formally only characterized on the level that already, for instance OWL2 provides. Besides, I have difficulties seeing how, ontologically, the concept ResourceExtension can be subclass of the concept Resource and how the concept HumanRessource is at the same time a subclass of the RessourceExtension concept. Also, it is not clear what a verification property is and how it differs from an object property. The explanation of design decisions and ontology constructs is missing.
The definitions given in Sect. 8 seem repetitive. The way the definitions are phrased, it reads as if they all give different definitions of the verification property. The restriction to individuals of concepts from a particular ontology (i.e. the Business Process Ontology) seems not very useful. I would suggest restricting the subject and objecting to a particular concept (or union of concepts) as done in Definitions 2-4.
The formal specifications of the different rules in Definitions 5-7 are all identical. Also the notation is very unconventional, what does the "iff" mean here? Why is it connected with a logical AND to a predicate? Why are the logical AND indexed? The SWRL rules in Table 3 use variables that occur in the consequent, but not in the antecedent, this violates the so-called safety condition in SWRL. Does that have any implication?

I think a background section that introduces the basic notions and concepts that are not common knowledge in the knowledge engineering community would be helpful. I would count Business Process, BPMN and BPEL and even Cloud computing among the notions that should be explained. The preliminaries that come in Sect. 4, start at a very detailed level of a particular perspective within Business Processes, without having made clear what Business Processes are, and what perspectives are in Business Processes. From this point, the paper would better fit an enterprise computing or business modelling conference.

Personally, I would have liked to see a stronger evaluation. The validation was very difficult for me to understand. Mostly, because I failed to properly understand what exactly you did to evaluate your solution and under which conditions. The two screenshots are very small and I can only tell that some annotations appeared, but I cannot see how and why and what role the ontology played in this. Also you use two tools, Signavio and Protege+Pellet for the validation. How do the two tools integrate? How does the Business Process Model from Signavio end up in Protege to be checked?
However, there is a second evaluation in Sect. 2, where the approach is compared to existing approaches. But neither the criteria, nor the way the different approaches where judged are transparent to the reader.

The paper contains many grammatical and orthographic errors, especially mixing singular and plural, missing articles, and missing commas after connectives, such as:

Abstract

- “In recent years […] are becoming”
- “researches have been realized”

Introduction
- Furhtermore -> Furthermore (a spell-checker would help with such typos)
- such field (?)
- “Regarding syntactic models, formal semantic models came to struggle its limitations”: I do not understand the sentence: Who is struggling and with what?
- “[…] incorporating the semantic notion through resource knowledge base.” I do not understand this sentence either: What is a “semantic notion”, and what is “resource knowledge base”?
- aiming to verification -> aiming for verification (?)
- “validation beneath Signavio”: What does beneath mean here? and where does the validation take place?

Related work
- There exits […] works -> There exists […] work
- The abbreviations BPMN and BPEL are not explained
- On the contrary our approach integrates cloud aspect […] -> On the contrary, our approach integrates cloud aspects […]
- additioning -> adding
- Nevertheless they have […] -> Nevertheless, they have […]
- [9],[10],[8],[7] -> [7,8,9,10] or [7-10]
- What does well-defined mean for cloud resources and resource dependencies?
- All approaches have considered resource perspective -> All approaches have considered the resource perspective

Motivating Example
- company -> company
- Check Order -> Cancel Order
- What is “semantically interdependent”? The semantics of one depends on the other and vice versa?
- business processes descriptions -> business process description

Preliminaries
- informations -> information
- What is “joining resources with a concept”?
- assigning resources with sets -> assigning resources to sets
- description of resource perspective -> description of the resource perspective
- Actually there is different ways -> Actually, there are different ways

Approach Overview
- using rules SWRL -> using SWRL rules
- The last paragraph is redundant, it repeats the last paragraph of Sect. 1

BPMN Extension
- does not provide explicit definition -> does not provide an explicit definition OR does not provide explicit definitions
- which called ResourceExtension -> which is called ResourceExtension
- in order to define cloud resources structure -> in order to define cloud resources OR in order to define a cloud resource structure OR in order to define a structure of cloud resources
- occi -> OCCI
- firstly -> remove, there is no secondly
- “core of its tag”: What is that?

Semantic Model for resource management in business processes
-> Caps in the heading!
- As discussed above […] -> redundant
- “adding semantic enhancement on resource concepts” -> What does that mean?
- resources defines -> resources define
- What are similar capacities? What is “obviously similar”, it is not obvious to me.

Privacy, security and optimisation properties
- “or even as a predicate” ?
- sub concepts -> subconcepts
- The second set […] establish -> The second set […] establishes
- operations including in -> operations included in
- It exists three types -> There are there types

Validation
- What are the units Go and Mo?
- Cancel Order activity have -> Cancel Order activity has