Aitor Murguzur

Context-Aware Staged Configuration of Process Variants@Runtime

Process-based context-aware applications are increasingly becoming more complex and dynamic. Besides the large sets of process variants to be managed in such dynamic systems, process variants need to be context sensitive in order to accommodate new user requirements and intrinsic complexity. This paradigm shift forces us to defer decisions to runtime where process variants must be customized and executed based on a recognized context. However, there exists a lack of deferral of the entire process variant configuration and execution to perform an automated decision of subsequent variation points at runtime. In this paper, we present a holistic methodology to automatically resolve process variability at runtime. The proposed solution performs a staged configuration considering static and dynamic context data to accomplish effective decision making. We demonstrate our approach by exemplifying a storage operation process in a smart logistics scenario. Our evaluation demonstrates the performance and scalability results of our methodology.

Context variability modeling for runtime configuration of service-based dynamic software product lines

In emerging domains such as Cloud-based Industrial Control Systems (ICSs) and SCADA systems where data-intensive and high performance computing are needed, a higher degree of flexibility is being demanded to meet new stakeholder requirements, context changes and intrinsic complexity. In this light, Dynamic Software Product Lines (DSPLs) provide a way to build self-managing systems exploiting traditional product line engineering concepts at runtime. Although context-awareness is widely perceived to be a first-class concern in such runtime variability mechanisms, existing approaches do not provide the necessary level of formalization to model and enact context variability for DSPLs. This is crucial for operational analytics processes since variant configuration could differ from context to context depending on diverse data values linked to context features and cross-tree constraints in a feature model. In this paper, we propose a context variability modeling approach, demonstrate its applicability and usability via a wind farm use case, and present the fundamental building blocks of a framework for enabling context variability in service-based DSPLs which provide Workflow as a Service (WFaaS).

Process Flexibility in Service Orchestration: A Systematic Literature Review

In dynamic environments, changes are often unpredictable and complex. Process models cannot be fully specified up-front and process flexibility becomes a key issue. Enterprise applications and systems supporting such processes are increasingly being architected in a service-oriented style. In this light, our goal is to analyze service orchestration approaches from a process flexibility perspective. Through a systematic literature review, we evaluate 17 service orchestration approaches and analyze their support for: (i) variability, support for large collections of process variants, (ii) adaptation, need for instance changes during runtime, (iii) evolution, need for schema changes during runtime, and (iv) looseness, need for loosely-specified models. The review findings provide a clearer understanding of process flexibility requirements and service orchestration mechanisms that support them, helping us to understand the limitations and shed light on future research areas.

Model query translator: A model-level query approach for large-scale models

Persisting and querying models larger than a few tens of megabytes using XML introduces a significant time and memory footprint overhead to MDD workflows. In this paper, we present an approach that attempts to address this issue using an embedded relational database as an alternative persistence layer for EMF models, and runtime translation of OCL-like expressions for efficiently querying such models. We have performed an empirical study of the approach using a set of large-scale reverse engineered models and queries from the Grabats 2009 Reverse Engineering Contest. Main contribution of this paper is the Model Query Translator, an approach that translates (and executes) at runtime queries from model-level (EOL) to persistence-level (SQL).

On the support of multi-perspective process models variability for smart environments

Cloud service-based applications are to be adapted to serve multiple platforms and stakeholders. Atop of such services, Smart Green Buildings are fostering a plethora of processes within their sustainability life-cycle. This introduces a number of challenges, as how to support multiple perspectives of domain-specific variability and how to deal with large collections of related process variants. To tackle this, there is a need to handle multi-perspective variability for processes. This paper introduces an approach to manage multi-perspective process variability by means of a meta-model and a modeling methodology, representing separately people and things variability perspectives in smart environments. Initial experimental results are also described, which indicate encouraging results for managing highly complex variability models.

Multi-perspective Process Variability: A Case for Smart Green Buildings (Short Paper)

The variability scale in large-scale Cyber-Physical Systems (CPSs) is high and complex due to the voluminousness, dynamicity and diversity of available computing resources (people, things and software services), domain-specific processes, domain-specific elements (stakeholders, assets and contracts), and their relationships. This requires us to go beyond current variability modeling and management techniques which neglect the complexity and the diversity of relevant stakeholders, data and assets, and thus cannot cope with intelligent business and analytics requirements in dynamic environments, such as smart city management. In this paper, we present a comprehensive analysis for understanding the multi-perspective variability in processes atop people, data and things in CPSs, particularly, for the sustainability governance of Smart Green Buildings (SGBs). We examine domain-specific processes and domain-specific elements and their relationships to derive a multiple-perspective variability management for SGBs. On the basis of this, we conceptualize a novel model for the multi-perspective process variability representation.

Process Variability through Automated Late Selection of Fragments

Process-aware information systems must encompass business process flexibility support due to business needs and factors coming from assorted sources, changing market conditions, customer needs, and regulations. However, flexibility may not be always achieved by pre-specified processes whereby, when context information is only available at runtime, decision making should be deferred to execution time. The late selection pattern defers the selection of placeholder activities’ implementations, binding applicable process fragments at runtime. This paper presents the foundations of a novel approach for an end-to-end variability management of process models through late selection of fragments by means of: (i) managing process fragments separately from the base model, (ii) resolving variation points automatically considering constraints and context data at runtime, and (iii) enabling process fragment recommendations based on experience logs.

Runtime Translation of Model-Level Queries to Persistence-Level

Different studies have proved that XMI (default persistence in Eclipse Modelling Framework) has some limitations when operating with large models. Recent approaches propose databases for persistence of models. Therefore, persistence level languages could be used to efficiently query models. While persistence level languages increase performance as they take advantages of underlying databases, they compromise usability for model engineers. Model engineers are familiar with model-level query languages (e.g. EOL or OCL). We present MQT (Model Query Translator), a runtime translation of model-level to persistence-level queries. Thus, we provide model engineers the usability of a model level language but also take advantage of performance optimization of databases. We have performed an empirical study of the approach using the GraBaTs 2009 case study (models from 8.8 MB to 646 MB) and results indicate that persisting models in a database, and combining it with runtime query translation provides a promising alternative for querying large models.

DRain: An Engine for Quality-of-Result Driven Process-Based Data Analytics

The analysis of massive amounts of diverse data provided by large cities, combined with the requirements from multiple domain experts and users, is becoming a challenging trend. Although current process-based solutions rise in data awareness, there is less coverage of approaches dealing with the Quality-of-Result (QoR) to assist data analytics in distributed data-intensive environments. In this paper, we present the fundamental building blocks of a framework for enabling process selection and configuration through user-defined QoR at runtime. These building blocks form the basis to support modeling, execution and configuration of data-aware process variants in order to perform analytics. They can be integrated with different underlying APIs, promoting abstraction, QoR-driven data interaction and configuration. Finally, we carry out a preliminary evaluation on the URBEM scenario, concluding that our framework spends little time on QoR-driven selection and configuration of data-aware processes.

Challenges in Enabling Quality of Analytics in the Cloud

Currently, domain scientists (DSs) face challenges in managing quality across multiple data analytics contexts (DACs). We identify and define quality of analytics (QoA) in dynamic and diverse environments, e.g., based on cloud computing resources for big data sources, as a composition of quality of data (data quality), performance, and cost, to name just the main factors. QoA is a complex matter and not just about quality of data or performance, which are typically considered separately when evaluating existing data analytics frameworks/algorithms. Frequently, the DS needs to utilize multiple frameworks to run different (sub)analytics, and, at the same time, the sub-analytics executed in these frameworks exchange inputs and outputs each other. In these cases, we observe different DACs, where a DAC refers to a particular situation in which the DS works with a specific framework to run a sub-analytics carried out by pipeline(s) or tasks in a pipeline. Each DAC has a set of interactions in the following categories: