Systematic Scalability Modeling of QoS-aware Dynamic Service Composition

Abstract

In Dynamic Service Composition(DSC), an application can be dynamically composed using web services to achieve its functional and Quality of Services (QoS) goals. DSC is a relatively mature area of research that crosscuts autonomous and services computing. Complex autonomous and self-adaptive computing paradigms (e.g. multi-tenant cloud services, mobile/smart services, services discovery and composition in intelligent environments such as smart cities) have been leveraging DSC to dynamically and adaptively maintain the desired QoS, cost and to stabilize long-lived software systems. While DSC is fundamentally known to be an NP-hard problem, systematic attempts to analyse its scalability have been limited, if not absent, though such analysis is of a paramount importance for their effective, efficient and stable operations. This paper reports on a new application of goal-modelling, providing a systematic technique that can support DSC designers and architects in identifying DSC relevant characteristics and metrics that can potentially affect the scalability goals of a system. The paper then applies the technique to two different approaches for QoS-aware dynamic services composition, where the paper describes two detailed exemplars that exemplify its application. The exemplars hope to provide researchers and practitioners with guidance and transferable knowledge, in situations where the scalability analysis may not be straightforward. The contributions provide architects and designers for QoS-aware dynamic service composition with the fundamentals for assessing the scalability of their own solutions, along with goal models and a list of application domain characteristics and metrics that might be relevant to other solutions. Our experience has shown that the technique was able to identify in both exemplars application domain characteristics and metrics that had been overlooked in previous scalability analyses of these DSC, some of which indeed limited their scalability. It has also shown that the experiences and knowledge can be transferable: the first exemplar was used as an example to inform and ease the work of applying the technique in the second one, reducing the time to create the model, even for a non-expert. Some limitations of the technique are also commented.

Synchronization and Standardization of Open Data Platforms: A Systematic Literature Review

“…While open data has numerous advantages, it also comes with its share of disadvantages and challenges [35]. This literature review explores the disadvantages of open data as identified in academic research and practical applications [36]. Privacy concerns, data quality issues, security risks, potential misuse, the digital divide, and resource constraints are among the key disadvantages associated with open data.…”

Scalable and Optimal QoS-Aware Manufacturing Service Composition via Business Process Decomposition

“…In fact, the QoS-aware service composition has been widely explored for Web services or cloud services in the service computing field [9][10][11][12][13][14]. By reviewing the literature, the techniques for QoS-aware service composition can be classified into four categories: local maximization approaches, linear optimization approaches, approximation approaches, and Pareto-optimization approaches [15].…”

QoS-Centric Diversified Web Service Recommendation Based on Personalized Determinantal Point Process

“…QoS-aware service recommendation is an important subproblem in the service computing field and can be regarded as a QoS-aware service selection optimization problem. The existing research work can be categorized as follows: utility-based methods [26][27][28], Skyline-based methods [12,13,29], collaborative-filtering-based methods [14,15], matrix-factorization-based methods [16,17,30], and factorization-machine-based methods [18,19,31]. Next, we introduce these methods and their application scenarios from the perspective of their advantages and disadvantages.…”