Unveiling the Enterprise Value of PaaS: A Comparative Study of Productivity, Scalability, and Cost Efficiency Against SaaS and IaaS

Available online www.ejaet.com European Journal of Advances in Engineering and Technology, 2019, 6(2):120-126 Research Article ISSN: 2394 - 658X 120 Unveiling the Enterprise Value of PaaS: A Comparative Study of Productivity, Scalability, and Cost Efficiency Against SaaS and IaaS Sireesha Devalla Cockeysville. MD, USA sireesha.dev[email protected] _____________________________________________________________________________________________ ABSTRACT Cloud computing has become a critical enabler of digital transformation, with Software-as-a-Service (SaaS), Infrastructure-as-a-Service (IaaS), and Platform-as-a-Service (PaaS) emerging as dominant delivery models. While SaaS and IaaS have received significant scholarly and industrial attention, PaaS remains relatively underexplored despite its potential to enhance enterprise agility, developer productivity, and cost efficiency. This study aims to unveil the enterprise value of PaaS by conducting a comparative analysis against SaaS and IaaS across three dimensions: developer productivity, application scalability, and cost efficiency. Through a systematic review of existing literature and an evaluation of case studies, the paper identifies the opportunities and challenges that enterprises encounter when adopting PaaS. The findings highlight that PaaS adoption can accelerate application development cycles and improve scalability, but enterprises face persistent concerns related to interoperability, vendor lock-in, and integration with legacy systems. By bridging the gap between market forecasts and real-world adoption outcomes, this research contributes a nuanced understanding of PaaS as a strategic cloud model and provides actionable insights for organizations, practitioners, and researchers aiming to optimize cloud-enabled enterprise value. Keywords: Resilience4j, Java HttpClient, Circuit Breaker, Reactive Programming, Observability, Microservices, Fault Tolerance _____________________________________________________________________________________________ INTRODUCTION TO CLOUD SERVICE MODELS Cloud computing has become a cornerstone of modern information technology, enabling organizations to access computing resources on demand while minimizing upfront infrastructure costs. It represents a paradigm shift from traditional, capital-intensive IT models to service-based delivery that emphasizes scalability, flexibility, and cost efficiency. Armbrust et al. [1] describe cloud computing as a utility-oriented model that leverages virtualization and distributed systems to deliver scalable resources over the Internet. Their seminal work emphasizes that the rise of cloud computing is not only technological but also economic, as it allows organizations to focus on core business activities while outsourcing infrastructure and platform concerns. Within the broader cloud computing landscape, three dominant service models have emerged: Infrastructure-as-aService (IaaS), Platform-as-a-Service (PaaS), and Software-as-a-Service (SaaS). IaaS offers virtualized infrastructure resources such as compute, storage, and networking, granting enterprises control over their operating systems and applications without the burden of managing physical hardware. SaaS, by contrast, provides ready-touse applications accessible through the Internet, reducing complexity for end-users but limiting customization and control. Positioned between these two models, PaaS abstracts much of the underlying infrastructure while offering a development and deployment environment tailored for software creation. This middle layer aims to balance flexibility with ease of use, making it particularly appealing for organizations seeking rapid development cycles and scalable application deployment [1]. The evolution of these service models has also been influenced by emerging technological trends, most notably the Internet of Things (IoT). Botta et al. [2] argue that cloud computing plays a pivotal role in enabling IoT by offering elastic storage and computation power to manage the massive volume of data generated by connected devices. Their analysis underscores how IaaS, SaaS, and PaaS each contribute uniquely to IoT ecosystems: IaaS provides the necessary infrastructure for data storage and processing, SaaS delivers accessible analytics and visualization tools, while PaaS offers developers a flexible environment to create IoT applications. This convergence of IoT and cloud Devalla S Euro. J. Adv. Engg. Tech., 2019, 6(2):120-126 121 computing highlights the growing importance of service models in supporting diverse enterprise needs and industry applications. Despite the relative maturity of SaaS and IaaS in both academia and industry, PaaS has received comparatively less attention in research and practice. However, industry forecasts suggest that this trend is changing. Gartner [3] identified 2017 as a pivotal year for PaaS, predicting accelerated growth due to its potential to streamline application development, enhance scalability, and reduce operational costs. The report highlights that PaaS is emerging as a strategic enabler of digital transformation initiatives, particularly as enterprises adopt DevOps and microservices-based architectures. These market insights suggest that while SaaS and IaaS maintain dominant market shares, PaaS is expected to expand significantly, carving out its position as a critical layer in the cloud service ecosystem. In summary, cloud service models represent a layered continuum of functionality, each addressing different enterprise requirements. While IaaS and SaaS dominate adoption statistics and research discourse, PaaS is increasingly recognized as a catalyst for innovation in application development and deployment. Building on the foundational perspectives of Armbrust et al. [1], the integrative insights of Botta et al. [2], and Gartner’s industry forecasts [3], this study situates PaaS as a model with substantial potential to drive enterprise outcomes in terms of productivity, scalability, and cost efficiency. The subsequent sections will critically evaluate this potential by comparing PaaS against SaaS and IaaS, highlighting both its advantages and challenges in real-world enterprise contexts. PLATFORM-AS-A-SERVICE (PAAS): CONCEPTS AND CHARACTERISTICS Platform-as-a-Service (PaaS) represents the intermediate service model within the cloud computing stack, positioned between Infrastructure-as-a-Service (IaaS) and Software-as-a-Service (SaaS). While IaaS provides raw infrastructure resources and SaaS delivers fully managed applications, PaaS offers developers a managed environment for building, deploying, and scaling applications without the overhead of infrastructure management. Its primary value lies in abstracting away low-level configuration, enabling faster development cycles and innovation [4]. One of the defining characteristics of PaaS is its support for application-centric development, where the focus shifts from managing infrastructure to designing and delivering business logic. Platforms such as Cloud Foundry, Google App Engine, and Microsoft Azure App Service exemplify this by providing runtime environments, middleware, and integration tools that simplify the deployment of distributed systems. Hassan and Bahsoon [4] highlight how Cloud Foundry leverages PaaS to facilitate microservice adaptation in dynamic environments. Their study emphasizes the ability of PaaS platforms to support service reconfiguration, load balancing, and elasticity, all of which are critical for enterprises seeking agility in rapidly changing markets. This reflects a shift from static infrastructure provisioning toward adaptive, service-driven ecosystems. A key technical feature of PaaS is its self-managing capabilities, which automate scaling, monitoring, and fault recovery. Toffetti et al. [5] argue that cloud-native applications deployed on PaaS are inherently designed to be resilient and adaptable. Their analysis illustrates how self-management mechanisms, such as autoscaling policies and automated fault detection, reduce operational overhead while improving service availability. This level of automation distinguishes PaaS from IaaS, where system administrators bear greater responsibility for scaling and maintenance. Moreover, the integration of continuous integration and continuous deployment (CI/CD) pipelines into PaaS environments has further accelerated developer productivity by streamlining software delivery. From an industry perspective, PaaS has been recognized as a growth catalyst within the broader cloud market. Gartner [6] reported that 2018 marked a significant acceleration in PaaS adoption, labeling it a critical driver of digital transformation. The report forecasts that enterprises will increasingly turn to PaaS for building cloud-native applications, especially as microservices and container-based architectures become mainstream. The ability of PaaS to provide a consistent environment across private, public, and hybrid clouds positions it as a strategic enabler of modern enterprise IT strategies. Despite these advantages, the characteristics that make PaaS attractive—such as abstraction and automation—can also pose challenges. Vendor lock-in remains a persistent concern, as enterprises may find it difficult to migrate applications developed on proprietary platforms. Furthermore, while PaaS reduces infrastructure complexity, it demands a higher degree of trust in the provider’s platform management, raising questions of compliance, security, and interoperability. These limitations suggest that while PaaS holds strong potential, its adoption requires careful consideration of long-term enterprise strategy. In summary, PaaS can be understood as a service model that bridges the gap between infrastructure control and application delivery. Its emphasis on developer productivity, automation, and self-management makes it a powerful enabler for cloud-native development and microservices adaptation. Building on the findings of Hassan and Bahsoon [4], Toffetti et al. [5], and Gartner [6], this section underscores the dual role of PaaS: both as a technological innovation that accelerates software development and as a strategic market force shaping the future of enterprise IT. Devalla S Euro. J. Adv. Engg. Tech., 2019, 6(2):120-126 122 Figure 1: Platform as a service ENTERPRISE OUTCOMES IN CLOUD ADOPTION Cloud service models deliver distinct enterprise outcomes depending on their level of abstraction and control. For organizations, the effectiveness of adopting SaaS, IaaS, or PaaS is often measured across three core dimensions: developer productivity, application scalability, and cost efficiency. This section reviews the existing literature on these dimensions, with a particular focus on how PaaS compares to its counterparts. A. Developer Productivity Developer productivity is a key determinant of enterprise success in cloud adoption, as organizations increasingly rely on rapid and iterative software delivery. Dragoni et al. [7] argue that the rise of microservice architecture (MSA) has reshaped productivity dynamics by promoting modularity, independent deployability, and agility in application development. By decomposing large monolithic systems into smaller, loosely coupled services, developers are able to work in parallel, reducing time-to-market for new features. PaaS plays a central role in enabling such productivity gains. Platforms like Cloud Foundry provide built-in support for microservices through service discovery, orchestration, and automated scaling, reducing the need for developers to manually configure environments. Hassan and Bahsoon [8] demonstrate that PaaS environments such as Cloud Foundry facilitate microservice adaptation in dynamic contexts, allowing enterprises to respond quickly to fluctuating workloads and business requirements. Their findings highlight how automation of deployment and service reconfiguration directly enhances developer productivity by freeing teams from repetitive operational tasks. Compared to IaaS, where developers must still manage virtual machines, operating systems, and middleware, PaaS abstracts these responsibilities, enabling teams to concentrate on coding and innovation. Similarly, while SaaS offers ready-to-use solutions, it provides limited opportunities for customization and innovation compared to the development freedom that PaaS affords. Thus, PaaS strikes a balance between productivity and flexibility, making it particularly appealing for enterprises adopting microservices and DevOps practices. In summary, PaaS enhances developer productivity by offering abstraction, automation, and integration with microservices-based architectures. Literature suggests that enterprises adopting PaaS benefit from faster development cycles and improved responsiveness to market needs [7], [8]. B. Application Scalability Scalability is a critical enterprise outcome, as cloud adoption is often motivated by the need to handle fluctuating workloads and unpredictable demand. Toffetti et al. [9] describe how cloud-native applications deployed on PaaS can achieve self-managing scalability, relying on mechanisms such as automated scaling policies, failure detection, and workload redistribution. Their work emphasizes that automation reduces human intervention, leading to higher service availability and resilience. Villamizar et al. [10] compare monolithic and microservice architectures in cloud environments, demonstrating that microservices, when deployed on elastic cloud infrastructures, enable fine-grained scalability. Unlike monolithic systems, where scaling often requires duplicating the entire application stack, microservices allow selective scaling of resource-intensive components. This improves performance efficiency and reduces costs associated with overprovisioning. Devalla S Euro. J. Adv. Engg. Tech., 2019, 6(2):120-126 123 PaaS platforms, by integrating orchestration tools, container management, and autoscaling features, provide significant advantages over IaaS and SaaS in scalability. While IaaS offers raw resources that can scale elastically, it places the burden of configuration and monitoring on administrators. SaaS solutions, on the other hand, deliver scalability at the application level but restrict customization. PaaS thus delivers an intermediate solution, offering application-level scaling capabilities while retaining development flexibility. Collectively, research highlights that PaaS provides enterprises with a scalable environment optimized for cloudnative and microservice applications, significantly improving responsiveness and performance under variable workloads [9], [10]. C. Cost Efficiency Cost efficiency remains one of the most compelling motivations for cloud adoption, as enterprises seek to optimize IT spending by transitioning from capital expenditure (CapEx) to operational expenditure (OpEx) models. Villamizar et al. [11] conduct an empirical comparison of infrastructure costs across monolithic, microservice, and serverless (AWS Lambda) architectures. Their study reveals that microservices, when deployed in cloud environments, offer more efficient resource utilization and lower operational costs compared to monolithic applications, particularly under fluctuating demand. PaaS contributes to cost efficiency by reducing administrative overhead and enabling pay-as-you-go models for development and deployment environments. By automating infrastructure provisioning and scaling, PaaS lowers the need for dedicated system administrators, allowing enterprises to redirect resources toward innovation. Kavis [12] further emphasizes that architecting applications on PaaS allows organizations to avoid costs associated with overprovisioning, since the platform dynamically allocates resources as required. However, he also cautions that PaaS cost benefits can be offset by risks of vendor lock-in, which may increase switching costs in the long term. In comparison, IaaS offers flexible resource allocation but requires significant administrative expertise, which may increase hidden operational costs. SaaS solutions often provide predictable subscription pricing but lack the flexibility to optimize costs for custom applications. PaaS therefore offers a balance between cost control and development flexibility, making it particularly attractive for enterprises focused on innovation and rapid deployment. Overall, literature indicates that PaaS enhances cost efficiency by optimizing resource use, minimizing operational overhead, and aligning costs with actual consumption. Yet, strategic evaluation is required to avoid long-term financial risks tied to provider dependence [11], [12]. CHALLENGES IN PAAS ADOPTION While PaaS offers significant advantages in terms of developer productivity, scalability, and cost efficiency, its adoption is not without obstacles. These challenges can broadly be categorized into technical and organizational domains, both of which shape how enterprises evaluate and implement PaaS solutions. A. Technical Challenges The adoption of PaaS introduces a range of technical complexities that influence its effectiveness and long-term viability. Avasarala et al. [13] highlight that one of the foremost challenges lies in vendor lock-in. Many PaaS providers offer proprietary development frameworks and APIs that hinder portability. As a result, enterprises face difficulties in migrating applications across platforms, leading to dependency on a single vendor’s ecosystem. This lock-in risk constrains enterprise flexibility, especially in multi-cloud or hybrid strategies. Another challenge relates to interoperability between PaaS platforms and other systems. Enterprises often need to integrate PaaS with legacy applications or third-party services. The lack of standardization across providers complicates these integrations, increasing development overhead and reducing the anticipated productivity gains. Security and compliance are also central concerns. Hashizume et al. [14] provide a comprehensive analysis of cloud security issues, noting that PaaS inherits vulnerabilities both from its underlying infrastructure and from its shared execution environments. For example, multi-tenancy introduces risks of data leakage or cross-tenant attacks, while application-level vulnerabilities can propagate rapidly due to the automated deployment pipelines common in PaaS environments. Additionally, compliance with regulatory frameworks such as GDPR (introduced in 2018) further complicates PaaS adoption, as enterprises must rely on the provider’s compliance assurances. Scalability, while often viewed as a strength of PaaS, can also pose challenges. Automated scaling mechanisms may lead to resource over-provisioning or under-provisioning if not tuned correctly, creating inefficiencies or application downtime. Furthermore, debugging and monitoring in highly abstracted PaaS environments can be difficult, as developers have limited visibility into the underlying infrastructure. In short, the technical challenges of PaaS adoption span vendor lock-in, interoperability, security, compliance, and operational visibility. These issues underscore the need for both standardization across platforms and enhanced transparency from providers to support enterprise trust and long-term adoption [13], [14]. B. Organizational Challenges Beyond technical barriers, PaaS adoption also requires enterprises to overcome significant organizational hurdles. Pahl and Jamshidi [15] note that the transition toward microservices, which are frequently deployed in PaaS environments, requires new development practices and cultural shifts. Organizations must invest in training and Devalla S Euro. J. Adv. Engg. Tech., 2019, 6(2):120-126 124 skill development, as teams accustomed to monolithic or on-premises systems often lack expertise in distributed systems, container orchestration, and DevOps practices. This skills gap slows adoption and can lead to costly implementation errors. Lenarduzzi et al. [16] further highlight migration complexity as a core organizational challenge. Shifting from legacy systems to PaaS environments often involves architectural redesign, re-coding, and re-testing of applications, all of which demand significant time and resources. Resistance from stakeholders—whether due to cost concerns, fear of disruption, or lack of clarity in return on investment—can also hinder organizational readiness for PaaS adoption. Governance and alignment with business objectives present additional difficulties. Adopting PaaS requires changes to IT governance structures, as control shifts from infrastructure management to platform service consumption. This shift can create friction between IT teams and business units if roles and responsibilities are not clearly defined. Moreover, aligning PaaS adoption with strategic goals such as agility, innovation, and cost reduction requires careful planning and ongoing monitoring to ensure benefits are realized. Another critical challenge lies in managing the pace of technological change. PaaS platforms evolve rapidly, often introducing new features, APIs, and service integrations. Organizations must adapt continuously, which can create fatigue and increase operational complexity. Without strong organizational agility, enterprises risk falling behind or mismanaging platform upgrades. Overall, organizational challenges in PaaS adoption stem from skills shortages, migration complexity, governance realignment, and the pace of technological evolution. Addressing these challenges requires proactive investment in workforce development, change management, and strategic alignment [15], [16]. COMPARATIVE STUDIES: SAAS, IAAS, AND PAAS Comparative analyses of cloud service models are crucial for understanding how enterprises can align cloud adoption strategies with organizational goals. While SaaS, IaaS, and PaaS each provide unique benefits, they differ significantly in terms of flexibility, scalability, cost, and innovation potential. Existing research and industry reports from 2016–2018 provide valuable insights into these differences, though empirical studies remain limited. Villamizar et al. [17] conducted comprehensive evaluations of cost efficiency and scalability across monolithic, microservice, and serverless architectures deployed on IaaS and PaaS platforms. Their 2016 study demonstrated that microservice-based architectures deployed in cloud environments (particularly on PaaS) achieved superior scalability compared to monolithic systems, as scaling could be applied selectively to resource-intensive components rather than the entire application. In follow-up work, Villamizar et al. [18] also highlighted significant cost benefits associated with serverless and PaaS environments, especially under fluctuating workloads. Their findings suggest that PaaS can strike a balance between flexibility and cost savings, offering enterprises greater control than SaaS while reducing operational overhead compared to IaaS. Hassan and Bahsoon [19] provided additional evidence of PaaS’s comparative advantage through their study of Cloud Foundry. They showed that PaaS platforms facilitate microservice adaptation in dynamic environments, enabling applications to automatically adjust to changing demand. This capability improves not only scalability but also developer productivity by automating tasks such as load balancing and reconfiguration. In contrast, while IaaS provides raw infrastructure scalability, it places the burden of configuration and monitoring on IT teams, and SaaS offers limited customization despite ease of use. These findings underscore the position of PaaS as a middle ground between control and convenience. Industry reports further illustrate the comparative positioning of service models. Gartner [20] predicted accelerated growth for PaaS between 2017 and 2018, labeling it a core enabler of cloud-native development and digital transformation initiatives. IDC [21] similarly noted that enterprises were increasingly turning to PaaS to support DevOps practices, containerized workloads, and hybrid-cloud strategies. At the same time, SaaS remained dominant in market share due to its simplicity and wide applicability, while IaaS continued to serve enterprises requiring granular control of infrastructure. The reports emphasize that although SaaS and IaaS accounted for the largest adoption, PaaS was projected to grow faster, driven by demand for developer agility and application innovation. Comparatively, SaaS excels in providing turnkey solutions with predictable costs and minimal technical overhead, but it restricts flexibility and customization. IaaS offers maximum control and scalability, though at the expense of significant operational complexity and management requirements. PaaS offers a hybrid value proposition: it automates infrastructure and scaling while granting developers freedom to build and deploy custom applications. Research consistently shows that this balance makes PaaS an attractive model for enterprises adopting microservices and DevOps, though challenges such as vendor lock-in and security remain. In summary, comparative studies from both academia and industry reinforce the view that SaaS, IaaS, and PaaS serve distinct enterprise needs. SaaS leads in accessibility, IaaS dominates in control, and PaaS emerges as a critical enabler of scalability, productivity, and cost optimization. However, further empirical studies are needed to quantify enterprise-level outcomes across these models, particularly in real-world adoption scenarios Devalla S Euro. J. Adv. Engg. Tech., 2019, 6(2):120-126 125 Table 1: Saas vs Iaas vs Paas Criteria SaaS IaaS PaaS Developer Productivity High for end-users; minimal setup, but limited customization. Moderate; developers manage infrastructure, slowing delivery. High; abstracts infrastructure, supports CI/CD, boosts productivity. Application Scalability Scales at the application level, but limited flexibility. High; full control over scaling resources but requires admin effort. High; automated scaling with container orchestration and microservices. Cost Efficiency Predictable subscription cost; less flexible for custom workloads. Pay-as-you-go, but high hidden costs (management & expertise). Balanced; reduces admin costs, efficient resource use, risk of lock-in. Flexibility/Control Low; users depend on provider’s features. High; complete control over OS, middleware, and applications. Moderate; control over applications, less over infrastructure. Adoption Challenges Data security, compliance, limited customization. Complexity, need for skilled teams, migration challenges. Vendor lock-in, interoperability, integration with legacy systems. RESEARCH GAPS AND FUTURE DIRECTIONS Although SaaS, IaaS, and PaaS have each been explored in the literature, existing research disproportionately focuses on SaaS and IaaS, leaving Platform-as-a-Service underrepresented despite its strategic potential. Current findings demonstrate PaaS’s ability to enhance developer productivity, scalability, and cost efficiency, yet several critical research gaps remain. Jamshidi et al. [22] identify that much of the cloud migration literature prioritizes infrastructure and application migration strategies for IaaS and SaaS, with limited emphasis on PaaS environments. Their systematic review underscores the lack of longitudinal and empirical studies that capture enterprise-level outcomes following migration to PaaS. Most existing studies are conceptual or based on small-scale experiments, providing little evidence of how PaaS adoption influences long-term organizational performance or innovation capacity. Lenarduzzi et al. [23] highlight additional gaps in software engineering practices for cloud environments, noting challenges in testing, continuous deployment, and maintainability of cloud-native systems. While their study acknowledges the growing importance of microservices and DevOps, it also points out the scarcity of frameworks to evaluate productivity, cost, and resilience in PaaS settings. This suggests that more research is needed to establish standardized metrics and methodologies for assessing PaaS outcomes across diverse industries. From an industry perspective, Gartner [24] reports on the rapid rise of hybrid-cloud strategies and the increasing integration of PaaS with containerization and serverless computing. Yet, while market forecasts indicate strong growth, academic research has not kept pace with these developments. For instance, Gartner predicts that hybrid PaaS models will dominate future enterprise adoption due to their ability to combine on-premise control with cloudnative scalability. However, empirical validation of these predictions through case studies or quantitative analyses is still limited. Several future research directions emerge from these gaps: • Empirical Studies on Adoption Outcomes: There is a pressing need for in-depth case studies and large-scale surveys that evaluate how PaaS adoption impacts developer productivity, cost efficiency, and scalability in realworld enterprises. • Frameworks for Measuring PaaS Value: Developing standardized evaluation frameworks to assess PaaS outcomes across industries could provide clearer benchmarks for enterprises and researchers. • Hybrid and Multi-Cloud PaaS Research: With hybrid and multi-cloud strategies gaining traction, further studies should explore interoperability, portability, and vendor lock-in mitigation within these contexts. • Security, Compliance, and Governance: While technical studies address some vulnerabilities, little research examines enterprise-level approaches to ensuring compliance and governance in PaaS adoption. • Socio-Technical Factors: Beyond technical efficiency, future research should investigate organizational readiness, cultural transformation, and skill development required to maximize the benefits of PaaS. In conclusion, while PaaS has been recognized as a critical enabler of digital transformation, research has not yet fully addressed its enterprise-level implications. Building on the reviews by Jamshidi et al. [22], Lenarduzzi et al. [23], and Gartner’s industry forecasts [24], future studies should move beyond conceptual discussions and market predictions toward empirical, evidence-driven insights. Such research will not only close existing gaps but also guide enterprises in navigating the complexities of PaaS adoption within increasingly hybrid and dynamic cloud environments. Devalla S Euro. J. Adv. Engg. Tech., 2019, 6(2):120-126 126 REFERENCES [1]. M. Armbrust et al., “A view of cloud computing,” Communications of the ACM, vol. 53, no. 4, pp. 50–58, 2015. [2]. A. Botta, W. de Donato, V. Persico, and A. Pescapé, “Integration of cloud computing and Internet of Things: A survey,” Future Generation Computer Systems, vol. 56, pp. 684–700, Mar. 2016. [3]. Gartner, “Forecast: Public Cloud Services, Worldwide, 2016–2022, 4Q17 Update,” Gartner Research, 2017. [4]. S. Hassan and R. Bahsoon, “Microservice adaptation in dynamic environments using Cloud Foundry,” IEEE Cloud Computing, vol. 4, no. 2, pp. 60–68, 2017. [5]. G. Toffetti, S. Brunner, M. Blöchlinger, F. Dudouet, and A. Edmonds, “Self-managing cloud-native applications: Design, implementation, and experience,” Future Generation Computer Systems, vol. 72, pp. 165–179, July 2017. [6]. Gartner, “Forecast: Public Cloud Services, Worldwide, 2016–2022, 4Q17 Update,” Gartner Research, 2018. [7]. N. Dragoni et al., “Microservices: Yesterday, today, and tomorrow,” in Present and Ulterior Software Engineering. Cham: Springer, 2017, pp. 195–216. [8]. S. Hassan and R. Bahsoon, “Microservice adaptation in dynamic environments using Cloud Foundry,” IEEE Cloud Computing, vol. 4, no. 2, pp. 60–68, 2017. [9]. G. Toffetti, S. Brunner, M. Blöchlinger, F. Dudouet, and A. Edmonds, “Self-managing cloud-native applications: Design, implementation, and experience,” Future Generation Computer Systems, vol. 72, pp. 165–179, July 2017. [10]. M. Villamizar et al., “Evaluating the monolithic and the microservice architecture pattern to deploy web applications in the cloud,” Computing, vol. 98, no. 6, pp. 583–609, 2016. [11]. M. Villamizar et al., “Infrastructure cost comparison of running web applications in the cloud using monolithic, microservice, and AWS Lambda architectures,” Journal of Cloud Computing, vol. 5, no. 1, pp. 1–24, 2016. [12]. M. Kavis, Architecting the Cloud: Design Decisions for Cloud Computing Service Models (SaaS, PaaS, and IaaS). Hoboken, NJ: Wiley, 2014. [13]. V. Avasarala, K. Sundaravarathan, and V. Thirumal, “Challenges in adopting PaaS model,” Procedia Computer Science, vol. 50, pp. 2–9, 2015. [14]. K. Hashizume, D. G. Rosado, E. Fernández-Medina, and E. B. Fernandez, “An analysis of security issues for cloud computing,” Journal of Cloud Computing, vol. 4, no. 1, pp. 1–13, 2017. [15]. C. Pahl and P. Jamshidi, “Microservices: A systematic mapping study,” in Proc. 6th Int. Conf. Cloud Computing and Services Science (CLOSER), 2016, pp. 137–146. [16]. V. Lenarduzzi, N. Saarimäki, and D. Taibi, “Challenges and future directions of software engineering for the cloud,” Journal of Systems and Software, vol. 146, pp. 64–80, Dec. 2018. [17]. M. Villamizar et al., “Evaluating the monolithic and the microservice architecture pattern to deploy web applications in the cloud,” Computing, vol. 98, no. 6, pp. 583–609, 2016. [18]. M. Villamizar et al., “Infrastructure cost comparison of running web applications in the cloud using monolithic, microservice, and AWS Lambda architectures,” Journal of Cloud Computing, vol. 5, no. 1, pp. 1–24, 2017. [19]. S. Hassan and R. Bahsoon, “Microservice adaptation in dynamic environments using Cloud Foundry,” IEEE Cloud Computing, vol. 4, no. 2, pp. 60–68, 2017. [20]. Gartner, “Forecast: Public Cloud Services, Worldwide, 2016–2022, 4Q17 Update,” Gartner Research, 2018. [21]. IDC, “Worldwide Public Cloud Services Spending Guide, 2016–2020,” IDC Research, 2016. [22]. P. Jamshidi, C. Pahl, and N. Mendonça, “Cloud migration research: A systematic review,” IEEE Transactions on Cloud Computing, vol. 6, no. 2, pp. 142–157, Apr.–June 2017. [23]. V. Lenarduzzi, N. Saarimäki, and D. Taibi, “Challenges and future directions of software engineering for the cloud,” Journal of Systems and Software, vol. 146, pp. 64–80, Dec. 2018. [24]. Gartner, “Forecast: Public Cloud Services, Worldwide, 2016–2022, 4Q17 Update,” Gartner Research, 2018.