Andrea Vandin

A conceptual framework for adaptation

In this position paper we present a conceptual vision of adaptation, a key feature of autonomic systems. We put some stress on the role of control data and argue how some of the programming paradigms and models used for adaptive systems match with our conceptual framework.

Modelling and analyzing adaptive self-assembly strategies with Maude

Building adaptive systems with predictable emergent behavior is a challenging task and is becoming a critical need. The research community has accepted the challenge by proposing approaches of various nature: from software architectures, to programming paradigms, to analysis techniques. Our own contribution in this regard is a conceptual framework for adaptation centered around the stressed role of control data. The framework is naturally realized in a reflective logical language like Maude by using the Reflective Russian Dolls model, as we show in this paper. We exploit the recently released statistical model checker PVesta to analyze a prominent example of adaptive system: robot swarms equipped with obstacle-avoidance self-assembly strategies.

The SCEL Language: Design, Implementation, Verification

SCEL (Service Component Ensemble Language) is a new language specifically designed to rigorously model and program autonomic components and their interaction, while supporting formal reasoning on their behaviors. SCEL brings together various programming abstractions that allow one to directly represent aggregations, behaviors and knowledge according to specific policies. It also naturally supports programming interaction, self-awareness, context-awareness, and adaptation. The solid semantic grounds of the language is exploited for developing logics, tools and methodologies for formal reasoning on system behavior to establish qualitative and quantitative properties of both the individual components and the overall systems.

Reasoning (on) Service Component Ensembles in Rewriting Logic

Programming autonomic systems with massive number of heterogeneous components poses a number of challenges to language designers and software engineers and requires the integration of computational tools and reasoning tools. We present a general methodology to enrich SCEL, a recently introduced language for programming systems with massive numbers of components, with reasoning capabilities that are guaranteed by external reasoners. We show how the methodology can be instantiated by considering the Maude implementation of SCEL and a specific reasoner, Pirlo, implemented in Maude as well. Moreover we show how the actual integration can benefit from the existing analytical tools of the Maude framework. In particular, we demonstrate our approach by considering a simple scenario consisting of a group of robots moving in an arena aiming at minimising the number of collisions.

Symbolic computation of differential equivalences

Ordinary differential equations (ODEs) are widespread in many natural sciences including chemistry, ecology, and systems biology, and in disciplines such as control theory and electrical engineering. Building on the celebrated molecules-as-processes paradigm, they have become increasingly popular in computer science, with high-level languages and formal methods such as Petri nets, process algebra, and rule-based systems that are interpreted as ODEs. We consider the problem of comparing and minimizing ODEs automatically. Influenced by traditional approaches in the theory of programming, we propose differential equivalence relations. We study them for a basic intermediate language, for which we have decidability results, that can be targeted by a class of high-level specifications. An ODE implicitly represents an uncountable state space, hence reasoning techniques cannot be borrowed from established domains such as probabilistic programs with finite-state Markov chain semantics. We provide novel symbolic procedures to check an equivalence and compute the largest one via partition refinement algorithms that use satisfiability modulo theories. We illustrate the generality of our framework by showing that differential equivalences include (i) well-known notions for the minimization of continuous-time Markov chains (lumpability), (ii)~bisimulations for chemical reaction networks recently proposed by Cardelli et al., and (iii) behavioral relations for process algebra with ODE semantics. With a prototype implementation we are able to detect equivalences in biochemical models from the literature that cannot be reduced using competing automatic techniques.

Statistical analysis of probabilistic models of software product lines with quantitative constraints

We investigate the suitability of statistical model checking for the analysis of probabilistic models of software product lines with complex quantitative constraints and advanced feature installation options. Such models are specified in the feature-oriented language QFLan, a rich process algebra whose operational behaviour interacts with a store of constraints, neatly separating product configuration from product behaviour. The resulting probabilistic configurations and behaviour converge seamlessly in a semantics based on DTMCs, thus enabling quantitative analyses ranging from the likelihood of certain behaviour to the expected average cost of products. This is supported by a Maude implementation of QFLan, integrated with the SMT solver Z3 and the distributed statistical model checker MultiVeStA. Our approach is illustrated with a bikes product line case study.

Forward and Backward Bisimulations for Chemical Reaction Networks

We present two quantitative behavioral equivalences over species of a chemical reaction network (CRN) with semantics based on ordinary differential equations. Forward CRN bisimulation identifies a partition where each equivalence class represents the exact sum of the concentrations of the species belonging to that class. Backward CRN bisimulation relates species that have identical solutions at all time points when starting from the same initial conditions. Both notions can be checked using only CRN syntactical information, i.e., by inspection of the set of reactions. We provide a unified algorithm that computes the coarsest refinement up to our bisimulations in polynomial time. Further, we give algorithms to compute quotient CRNs induced by a bisimulation. As an application, we find significant reductions in a number of models of biological processes from the literature. In two cases we allow the analysis of benchmark models which would be otherwise intractable due to their memory requirements.

Modelling and analyzing adaptive self-assembly strategies with maude

Building adaptive systems with predictable emergent behavior is a challenging task and it is becoming a critical need. The research community has accepted the challenge by introducing approaches of various nature: from software architectures, to programming paradigms, to analysis techniques. We recently proposed a conceptual framework for adaptation centered around the role of control data. In this paper we show that it can be naturally realized in a reflective logical language like Maude by using the Reflective Russian Dolls model. Moreover, we exploit this model to specify and analyse a prominent example of adaptive system: robot swarms equipped with obstacle-avoidance self-assembly strategies. The analysis exploits the statistical model checker PVesta.

Adaptable Transition Systems

We present an essential model of adaptable transition systems inspired by white-box approaches to adaptation and based on foundational models of component based systems. The key feature of adaptable transition systems are control propositions, imposing a clear separation between ordinary, functional behaviours and adaptive ones. We instantiate our approach on interface automata yielding adaptable interface automata, but it may be instantiated on other foundational models of component-based systems as well. We discuss how control propositions can be exploited in the specification and analysis of adaptive systems, focusing on various notions proposed in the literature, like adaptability, control loops, and control synthesis.

Efficient Syntax-Driven Lumping of Differential Equations

We present an algorithm to compute exact aggregations of a class of systems of ordinary differential equations ODEs. Our approach consists in an extension of Paige and Tarjans seminal solution to the coarsest refinement problem by encoding an ODE system into a suitable discrete-state representation. In particular, we consider a simple extension of the syntax of elementary chemical reaction networks because i it can express ODEs with derivatives given by polynomials of degree at most two, which are relevant in many applications in natural sciences and engineering; and ii we can build on two recently introduced bisimulations, which yield two complementary notions of ODE lumping. Our algorithm computes the largest bisimulations in