Smart cities are challenging current big-data analytics software solutions. Smart cities require to process on the fly and at rest, huge amounts of heterogeneous data coming from geographically dispersed sources. Moreover, the future interaction between connected (or even autonomous) vehicles and smart cities for an enhancing safety mobility, will impose further non-functional requirements on big data systems such as real-time guarantees. CLASS, coordinated by Barcelona Supercomputing Center (BSC), and participated by the City of Modena, Maserati, University of Modena, IBM and Atos, is developing a novel software architecture that addresses all these challenges.
To do so, CLASS aims to converge and evolve high-performance, low-power embedded and big data analytics computing technologies into a unified software architecture capable of efficiently coordinating and distributing computation resources along the compute continuum (from edge to cloud computing resources), while providing real-time guarantees, as required by automotive systems.
The software architecture will integrate innovative software frameworks from multiple domains such as COMPSs, the software framework developed at BSC to design and execute high-performance applications in distributed environments, OpenWhisk, a serverless execution model developed by IBM, and Rotterdam, a Container as a Service (CaaS) components developed by Atos. Moreover, CLASS will integrate the most advanced parallel embedded computer architectures, featuring many-core and GPU acceleration technologies, to boost the computing capabilities at the edge side. These software components will integrate a unified software architecture to efficiently distributive big-data analytics workloads along the compute continuum while providing real-time guarantees to the overall execution.
This technology will be tested in a real urban area of one square kilometre in the city of Modena, using three prototype connected vehicles provided by Maserati and equipped with sensors and connectivity.
The project will pave the way towards better big-data systems for the smart city domain that will improve sustainability, services and safe mobility. In addition, the project will prepare the technological background for the advent of trustworthy autonomous vehicles. Furthermore, the technology behind the project may serve as a reference for applying big-data analytics in additional application domains combining field/edge and clouds (fog), such as production floors, logistics operations and more.
‘CLASS is a very challenging project whose objective is to develop a novel software framework for a new generation of highly distributed computing systems with big data analytics and real-time requirements, capable of coordinating computing resources along the compute continuum. BDVA members are at the heart of this innovative project, whose technology will help bring about the smart cities of tomorrow’ says Eduardo Quiñones, CLASS project coordinator and BSC Computer Sciences researcher.
NAME: CLASS: Edge and Cloud Computation: a Highly Distributed Software for Big Data Analytics
START/END DATE: 01/01/2018 – 31/12/2020
KEY THEMES: Smart cities, connected vehicles, data analytics, big data
PARTNERS: Spain: Barcelona Supercomputing Center (BSC), Atos; Italy: Università
BUDGET: €3.9 million
WEBSITE: class-project.eu
CLASS has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement no. 780622
