A bigraphical model for specifying cloud-based elastic systems and their behaviour

Abstract

a lIre laboratory, Constantine 2 university-abdelhamid mehri, Constantine, algeria; b lIuPPa laboratory, university of Pau, Pau, france ABSTRACTThe few recent years have witnessed the appearance of a new kind of selfadaptive systems called cloud based-elastic systems. These systems are particularly appealing for their ability to maintain a decent quality of service and reduce a system's operating cost at the same time. They achieve this by dynamically adjusting resources allocation in terms of elasticity. Meanwhile, complexity of structural and behavioural aspects related to cloud-based elastic systems increase the difficulty of designing and developing such systems. In this paper, we address this challenge by proposing a formal approach based on bigraphical reactive systems for modelling both structural and behavioural aspects of cloud-based elastic systems. In particular, we represent their behaviour in terms of client/application interactions and elasticity methods at different levels using bigraphical reaction rules. The feasibility of the proposed approach is illustrated through a motivating example running on the top of an Amazon Elastic Compute Cloud (EC2) infrastructure.Modelling cloud-based elastic systems using bigraphical reactive systems

Formalizing and simulating cross-layer elasticity strategies in Cloud systems

“…As for their executable capabilities, the few existing tools built around BRS as BigraphER [48] and BPL Tool [21] are limited and only suitable for some specific application domains. Furthermore, the BRS model-checker BigMC [40] that was used in [45], allows formal verification of safety properties. However, the possible verifications rely on very limited predefined predicates.…”

Formal modelling and verifying elasticity strategies in cloud systems

“…Bigraph CS is composed of two regions, noted 0 and 1 that, respectively, represent the hosting environment and the elasticity controller parts of the elastic cloud system. This configuration is obtained by the parallel composition of the hosting environment (back-end) and elasticity controller bigraphs as shown in [17]. The introduced sorting logic defines mapping rules and expresses all constraints and formation rules that CS satisfies to ensure proper and precise encoding of cloud semantics into BRS concepts.…”

“…In [17], authors introduced a formal approach based on BRS for modelling both structural and behavioural aspects of elastic cloud systems. Cloud elastic behaviours are represented in terms of client/application interaction.…”

Modeling and Evaluating Cross-layer Elasticity Strategies in Cloud Systems

“…As for our adopted formalism, BRS were proven useful in the specification of ubiquitous, context aware and distributed systems [24,17] and in other domains [6]. BRS were used in [31] to provide a generic model of elastic cloud systems. Authors modeled cloud structures with bigraphs in three parts: the frontend part, the back-end part and the elasticity controller.…”