Cloud-based services provide an alternative to the in-house implementation of various types of functionality. Organizations rely on such services to minimize the need for long-term commitments and enhance scalability and ubiquitous access to the services. However, achieving complex tasks that require a combination of services is not well studied, despite the potential added value. This paper investigates the requirements encountered when integrating cloud-based services in the modern organization. The paper proposes a modeldriven solution for capturing the requirements for integrating cloud-based services. The model is to be used within the larger context of the organizational design; modeling components used to describe requirements are related to other views of the organization. A prototype tool and an example business case are presented to illustrate how the requirements model can be elicited and designed. The models are capable of being transformed into an integration solution.
Cloud services are becoming an increasingly appealing solution for organizations that
want to outsource general functionality. Cloud computing relies on several
technologies, and its definition is still being debated [
This paper proposes extending the unifying meta-model for enterprise modeling
[
The paper is structured as follows. Section 2 is an overview of cloud computing and services. Section 3 explores the requirements of cloud-based services in the literature. Section 4 presents the proposed requirements model for integrating cloud services and demonstrates its use with an example case study. Finally, section 5 includes concluding remarks and future work. 2
Cloud computing is gaining popularity as an emerging technology that resides at the
intersection of concurrent advancements in diverse fields of technology [
The details of what constitutes ―the cloud‖ are still debated among researchers and
practitioners [
This definition highlights the technical perspective of cloud computing. Business
aspects, especially the influence of adopting cloud services on value configurations,
are often neglected [
The flexibility in provisioning and pricing is facilitated through a service-oriented
3-layered architecture [
Despite the foreseen advantages, organizations hesitate to fully adopt cloud
computing—a young domain with many unresolved challenges and associated risks and
uncertainties [
Variations in legal jurisdiction and the continuously evolving privacy laws
complicate the adoption of cloud services. Service Level Agreements (SLAs) are used as
tools to govern the relationship between cloud provider and cloud consumer
according to organizational policies and applicable rules and regulations [
Most literature describing the requirements for cloud services focuses on the
design of individual services. However, efforts are made to facilitate dynamic provider
switching, such as the proposals in [
To describe requirements for integrated, complex cloud services, we propose a modelbased approach that captures the design of the integrated services on a higher level of abstraction, i.e. the enterprise level. Doing so: a) addresses the business aspects of cloud services and exposes the value added by the integrated services to the overall organizational design, b) improves the alignment between the business goals and the eventual implementation of the integrated services, and c) facilitates the automated switching between candidate service providers, hence taking a step closer towards realizing the vision of global cloud marketplaces.
Information systems requirements must describe the underlying motivation for
developing a system in addition to the desired functionality of the system [
Enterprise Modeling (EM) provides means for describing the elements of business,
information, and technology components [
The RequirementsTable modeling component is a list of stakeholder-friendly requirements. A Requirement in the table contains the textual statement of the requirement (in the traditional sense). It is annotated by a priority value to create an order in which requirements should be implemented. Each requirement is associated with a RequirementView, a representational modeling component that groups together the components comprising a single requirement.
Concept from the CV represents the entities affected by the requirement. Process and Role from the BPV represent the functionality expressed by and the organizational structures affecting the requirement, respectively. The consume and produce IMRs capture how the concepts are processed. The attributes frequencyOfExecution and concurrentInstances describe non-functional performance aspects of a requirement.
BusinessGoal from the GV and the motivate IMR account for the motivation of a requirement. The motivate relationships enables motivating the requirement directly (a requirement being a model component itself) or indirectly through the requirement’s parts (e.g. concepts, processes, etc.). Problem (either internal, i.e. Weakness, or external, i.e. Threat) and Constraint capture the limitations affecting a requirement via the underlying goal of the requirement. Describing the Cause of a problem helps in designing suitable solutions to mitigate it. Opportunity describes growth potential in the organization and is used to model future desired requirements. Finally, Key Performance Indicators (KPIs), which are part of the GV, provide means for evaluating the fulfilment of a requirement in the implemented solution.
System from the SV represents the systems affected by the requirement. The type attribute captures the specific type of the modelled system. System maybe specialized to reflect possible implementation platforms, including cloud-based services. The type and specialization of a system affect the selection of the appropriate transformation rules that are applied to generate the integration solution. The use of the proposed RV for modeling cloud integration requirements is illustrated in this section using the example business case of the imaginary organization Aid&More. The requirements model is developed using a prototype tool that implements the unifying-meta model for enterprise modeling. The tool supports modeling enterprise- and system-level models as well as transforming the designed models to a platform-specific implementation.
Overview of the Example Business Case. Aid&More is a Non-Governmental Organization (NGO) active in the humanitarian aid domain. It offers support to people in distress situations as a result of natural disasters. Each aid project includes several types of aid and targets a specific distress situation described by: the type of natural disaster, number of affected people, target geographical location, etc. Types of aid in a project prescribe the necessary resources to deliver the aid—personnel, vehicles, warehouses, offices, and aid items. Creating a recognizable and popular brand is essential for the growth of Aid&More’s work; the organization relies mainly on public donations to fund its operations. Using information collected from social media, the NGO aims to understand and analyse its reputation and identify improvement potential, both in terms of project execution and public relations.
Aid&More is looking for a solution that automatically retrieves mentions of the NGO in social media. The solution should then, with the help of online analyses services, generate summary reports of how Aid&More is perceived. The desired architecture is illustrated in Fig. 3. The solution has access to project details, managed using a central Project Management System (called internally as PROM). The system has a public interface on the Web (i.e. website) to advertise statistics and results from projects. Google Analytics is employed to monitor and provide feedback on the impact of the Website. AnalyzeThis is an online service used for analysing data collected from social networks.
scribed using a goal model (i.e. instance of the GV), a part of which is shown in Fig. 4. Only the goals that motivate the cloud integration requirements are shown.
A section of the requirements table describing the desired integration solution is shown in Fig. 5. Each of the requirements has an associated requirement view (i.e. instance of the RV) that includes the design details of the requirement.
The requirement view for Requirement A3774OOJ is illustrated in Fig. 6. To fulfill the requirement (i.e. retrieve Aid&More mentions), the integration solution performs two activities: ―send search query‖ and ―retrieve search results.‖ The search query is affected by two factors: the aid project and the specific social media outlet being queried. The details of the aid project are used to formulate the content of the query. The interface details of the social media outlet describe the parameters used to establish a connection for sending the query and retrieving the results.
The concept ―SocialMediaMention‖ represents occurrences in social networks, including tags, referrals, shares, likes, re-tweets, etc. The model enables switching the RESTful services that require (i.e. receive) the search query. The switching is then reflected in the implementation of the integration solution by applying a different transformation rule that is suitable for the new service. A new transformation rule needs to be developed for every new service to be added, but this is done only once and the transformation rule can be reused in this or other projects. Existing research concerning requirements for cloud-based service focuses on individual services and tends to address the more technical aspects of cloud computing. Business-related aspects are overlooked. We propose in this paper modelling requirements for integrating cloud services as part of the organizational design. The unifying meta-model for enterprise modeling is extended with the Requirements View (RV), a composite view that covers components involved in describing requirements. The unifying meta-model enables relating requirements to other parts of the organizational design. The modelled systems can describe cloud-based services and thereby offer a mechanism for automated service provisioning at the enterprise-design level and contributing to achieving the envisioned global cloud marketplaces.
An example business case illustrated the use of the proposed RV with the help of a prototype tool. The requirements models can be transformed into a platform-specific integration solution that supports integrating different cloud services. The modularity of the unifying-meta model allows the RV to be extended with new modeling components that capture additional details about the requirements.