DATCloud: A Model-Driven Framework for Multi-Layered Data-Intensive Architectures

Moamin Abughazala 1 2 , Henry Muccini 1 · 2025

Read DATCloud: A Model-Driven Framework for Multi-Layered Data-Intensive Architectures by Moamin Abughazala 1 2 , Henry Muccini 1 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-18257.txt.

Introduction

Moamin Abughazala and Henry Muccini present a model-driven frame for designing data-heavy software systems, in a paper from two thousand twenty-five. Such systems span many layers and must stay flexible as needs shift. Their frame helps teams model, check, and refine layered designs before building.

The frame follows a global architecture standard and uses structure and behavior models with picture-based languages. A trial on a museum system in Florence cut modeling time by forty percent. Flexibility rose by about a third over hand methods. Code generation and simulation add-ons are planned, the authors say.

Catalog record

CREATORMoamin Abughazala 1 2 , Henry Muccini 1
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE20 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2501-18257

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

DATCloud: A Model-Driven Framework for Multi-Layered Data-Intensive Architectures Moamin Abughazala 1 2 , Henry Muccini 1 Abstract The complexity of multi-layered, data-intensive systems demands frameworks that ensure flexibility, scalability, and efficiency. DATCloud is a model-driven framework designed to facilitate the modeling, validation, and refinement of multi-layered architectures, addressing scalability, modularity, and real-world requirements. By adhering to ISO/IEC/IEEE 42010 standards, DATCloud leverages structural and behavioral meta-models and graphical domain-specific languages (DSLs) to enhance reusability and stakeholder communication. Initial validation through the VASARI system at the Uffizi Gallery demonstrates a 40% reduction in modeling time and a 32% improvement in flexibility compared to manual methods. While effective, DATCloud is a work in progress, with plans to integrate advanced code generation, simulation tools, and domain-specific extensions to further enhance its capabilities for applications in healthcare, smart cities, and other data-intensive domains. I Introduction The increasing complexity of data-intensive systems, driven by the rapid expansion of Internet of Things (IoT) devices, has introduced significant challenges for system modeling. Modern architectures must address scalability, modularity, and data management across multi-layered cloud, fog, and edge systems [ 1 ] . However, existing frameworks often lack robust modeling capabilities, resulting in inefficiencies when designing, adapting, and managing these architectures [ 2 ] . To address these challenges, this paper introduces DATCloud a model-driven framework designed to simplify and enhance the modeling of multi-layered, data-intensive systems. Unlike traditional methods DATCloud models and combines data systems across cloud, fog, and edge layers. DATCloud uses structural and behavioral meta-models to represent the logical architecture of systems, including components, data flows, and workflows. These meta-models provide architects with the tools needed to define, validate, and refine system designs efficiently. The framework further incorporates graphical domain-specific languages (DSLs), offering an intuitive interface for creating models that ensure consistency and compliance with architectural standards such as ISO/IEC/IEEE 42010 [ 3 ] . A key focus of DATCloud is improving the efficiency and flexibility of the modeling process. By supporting templates,and modular design, the framework reduces the effort required for iterative development and simplifies communication between stakeholders. These capabilities enable architects to handle complex architectures while minimizing manual overhead. Research, such as Monitor-IoT [ 4 ] , SimulatIoT [ 5 ] , Silva et al. [ 6 ] , and CHESSIoT [ 7 ] , has highlighted the potential of multilayer monitoring architectures for IoT systems. These studies cover software, hardware, physical, and deployment aspects, but reveal a notable gap in the Data View . The rest of this paper is structured as follows: Section II reviews related work, highlighting existing frameworks and their limitations in addressing the challenges of modeling multi-layered, data-intensive systems. Section III describes the design of DATCloud, detailing its structural and behavioral meta-models, core functionalities, and adherence to ISO/IEC/IEEE 42010 standards. Section IV presents the case study and discussion of initial results for DATCloud, focusing on metrics for modeling time and flexibility. It highlights stakeholder insights and demonstrates DATCloud’s practical effectiveness in managing visitor flow and architectural flexibility through its application in the VASARI system at the Uffizi Gallery. Finally, Section V outlines future directions for enhancing the framework, including advanced code generation and domain-specific extensions, and concludes with a summary of its contributions. II Related Work The rapid growth of IoT systems and data-intensive applications has prompted the development of various frameworks and methodologies for modeling multi-layered architectures. While significant progress has been made, some gaps still exist, particularly in the data view and in integrating data-intensive applications across the cloud, fog, and edge layers . This section reviews the key contributions in these areas and highlights the gaps that DATCloud aims to address. II-A Modeling IoT Multi-Layer Architectures Efficient architectures are essential for IoT systems to process distributed data across edge, fog, and cloud layers. Various frameworks have been proposed to model and manage these complex systems effectively. • Bauer et al. [ 2 ] introduced an IoT Reference Architecture emphasizing the roles of cloud, fog, and edge layers. While this model outlines task allocation, it lacks a concrete framework for modeling workflows and data interactions between these layers. • Taivalsaari and Mikkonen [ 8 ] focus on IoT systems’ lifecycle, emphasizing the interplay between edge devices and the cloud. However, their approach primarily addresses operational challenges rather than architectural modeling. • Ihirwe et al. [ 7 ] proposed CHESSIoT, a model-driven engineering (MDE) framework for IoT system design. This framework leverages domain-specific languages (DSLs) to model and verify IoT architectures. While effective for IoT-specific use cases, CHESSIoT does not generalize to multi-layered systems involving complex data workflows. While current frameworks recognize the hierarchical structure of IoT systems, they mainly concentrate on system behavior or task allocation. They do not adequately address the data view , which creates a gap in how data is modeled, stored, and processed across multi-layered architectures . Additionally, most frameworks are specific to certain domains and lack the flexibility needed for wider applications. II-B Modeling Data-Intensive Applications Data-intensive applications are defined by their need to manage, process, and analyze large volumes of data across distributed systems. Effective modeling of such systems requires frameworks that prioritize data workflows, scalability, and adaptability. • Kleppmann [ 9 ] explores the challenges of building scalable, distributed systems for handling data pipelines and real-time analytics. While this work provides a thorough understanding of data-intensive systems, it focuses on implementation rather than architectural modeling. • Abughazala et al. [ 10 ] [ 11 ] [ 12 ] proposed a Framework, one of the few works explicitly addressing modeling data-intensive workflows. The Framework introduces meta-models to describe how data flows between components. Its focus on industrial data workflows limits its relevance to general multi-layered architectures, including cloud, fog, and edge systems. • Raj [ 13 ] introduced a conceptual model for designing a data pipeline consisting of two primary components: nodes and connectors. The nodes serve as the core abstract data units, while the connectors facilitate data transmission and communication between these nodes. • Borelli [ 14 ] proposed a classification framework for key software components and their interrelationships to model software architectures tailored for specific IoT applications. These components are presented as abstract representations. • Erraissi [ 15 , 16 ] developed a meta-model encompassing data sources, ingestion layers, and a Big Data visualization layer to provide a structured approach to managing and visualizing Big Data workflows. Current frameworks for data-intensive applications lack a unified approach to modeling data workflows across cloud, fog, and edge layers . Instead of providing holistic integration, they often focus on isolated aspects. This gap highlights the need for a scalable and modular framework, such as DATCloud, to effectively tackle these complexities and ensure practical applicability in real-world scenarios. III Framework Design This section introduces DATCloud’s framework design, focusing on two key components: the structural meta-model , which defines the architecture of multi-layered systems, and the behavioral meta-model , which captures workflows and interactions. Together with graphical DSLs, these meta-models enable scalable, modular, and standards-compliant modeling for data-intensive systems. III-A Structural Meta-Model Using the structural meta-model, users can establish architecture across cloud, fog, and edge layers, model nodes and connections, choose data storage types like NoSQL or NewSQL, and select communication protocols such as

Read the complete text