Analysis of Complex Systems

Cloud computing is changing the traditional business model for Information Technology. It offers users with on-demand infrastructure and applications over the Internet with pay-as-you-go pricing. This changes how software suppliers build and deliver their programs. Cloud-based applications must be adaptable, modular, and quickly accessible. Furthermore, container technology allows applications can be run in communicating but self-contained environments that fully exploit the elasticity of cloud. These new possibilities and opportunities come with the cost of a more complex workflow. Development decisions are intertwined with the application deployment issues. Designing a complex system is a challenging: many possible settings and parameters can be tuned. Poor choices can result in system failure, high costs and displeased customers.

The ADAPT project started in 2018, with Lizeth Tapia Tarifa as leader and FRINATEK funding from the Research Council. This project aims to develop a new technique to improve data locality in parallel processing. The idea is systematically to coordinate task scheduling and data allocation using predicted patterns of data access to memory, i.e. reads and writes access to locations in memory. The approach will combine formal models of parallel systems with basic research in programming language theory to (1) capture abstractly the interaction of workflows with dynamically created tasks and memory locations on parallel computers, and (2) combine formal analysis with model-based simulations to explore how to systemically calibrate schedulers and allocators for specific applications. The result of this project will complement SIRIUS’ current toolbox on model-based prediction.

Our expertise lies in modelling complex parallel and distributed systems, including object-oriented and service-oriented systems, cloud computing and the Internet of Things. Analysis techniques for these models range from simulation, which analyses a single run of a system, to deductive verification, which analyses all possible runs of a system.  We analyse both functional and non-functional properties such as safety properties, timing properties, resource management and scaling strategies. Our work on resource-restricted parallel systems with timing constraints is currently being applied to planning and logistics in the context of SIRIUS.

We bring background from two, successfully completed, European projects: Envisage and HyVar. These have been described in detail in earlier reports. This background is unified through the ABS modelling language and analysis framework. Each of these projects provides important tools and experience that can be used in the SIRIUS experiments and pilots.  Today, we also bring solutions developed for planning and logistics in SIRIUS, as well as needs identified in this domain, back into the overall development of ABS.

It combines implementation-level specifications with verifiability, high-level design with executability, and formal semantics with practical usability. It is a concurrent, object-oriented, modelling language that features functional data-types. It is designed to develop executable models with a parallel, object-oriented program flow. ABS targets distributed and concurrent systems by means of concurrent object groups and asynchronous method calls. It supports model variability based on feature models and delta-oriented specifications. Deployment modelling can be based on high-level deployment models. The ABS system supports the modelling of resource-aware and resource-restricted systems and provides a range of techniques for model exploration and analysis, based on formal semantics.

Finally, ABS is an open source research project. You can find the project on GitHub at https://github.com/abstools/abstools.

To this end, the first international ABS workshop was held in Oslo, Norway (May 2017), the second international ABS workshop was held in Darmstadt, Germany (May 2018), and the third international ABS workshop will be held in Amsterdam, The Netherlands (May 2019). Researchers and research groups from, e.g., Oslo, Bergen, Darmstadt, Bologna, Madrid, Odense, Braunschweig, Torino, Paris, Djakarta, and Amsterdam are all stakeholders in the development of ABS and associated technology.