CLASS · Project

Real-Time Big Data Processing Across Edge and Cloud for Smart Cities and Vehicles

digitalPilotedTRL 6

Imagine your city has thousands of sensors on roads, buildings, and cars — all generating massive amounts of data every second. Right now, systems either react fast but shallow (like a smoke alarm) or think deep but slow (like a monthly traffic report). CLASS built software that does both at the same time: quick reactions at the sensor level and deep analysis in the cloud, all working together seamlessly. They proved it works with real prototype cars and city sensor networks for traffic management and driver assistance.

By the numbers

consortium partners

countries involved (ES, IL, IT)

industry partners in the consortium

56%

industry ratio in consortium

total project deliverables

demo deliverables with working software releases

The business problem

What needed solving

Companies running large sensor networks — in cities, factories, or vehicle fleets — face an impossible choice: process data instantly at the edge for quick reactions, or send it all to the cloud for deeper analysis. Doing both at once typically requires two separate systems that don't talk to each other, doubling infrastructure costs and creating blind spots. The result is either slow insights or shallow ones, never both together.

The solution

What was built

The project delivered a complete edge-to-cloud software architecture with 21 deliverables, including: an integrated Spark/COMPSs analytics platform, cloud data analytics service management and scalability components, real-time analysis tools for edge devices, a multi-workload performance evaluation tool, and the full CLASS architecture demonstrated on a real smart-city deployment with prototype autonomous vehicles.

Audience

Who needs this

City traffic management departments running large urban sensor networksAutomotive OEMs and Tier-1 suppliers developing connected and autonomous vehicle platformsIndustrial IoT platform providers needing real-time plus batch analyticsSmart city solution integrators deploying citywide sensor infrastructureBig data analytics companies serving clients with mixed real-time and historical processing needs

Business applications

Who can put this to work

Smart City Infrastructure

enterprise

Target: City traffic management operators and municipal IoT platform providers

If you are a city traffic management operator dealing with sensor data overload — where thousands of road sensors and cameras generate more real-time data than your current systems can process — this project developed an integrated edge-to-cloud software architecture that processes data-in-motion and data-at-rest simultaneously. It was demonstrated on a real smart-city use case with heavy sensor infrastructure across a wide urban area, handling large heterogeneous data streams in real-time.

Autonomous Vehicles & ADAS

enterprise

Target: Automotive companies and Tier-1 suppliers developing connected vehicle systems

If you are an automotive supplier struggling to process heterogeneous sensor data from cameras, LiDAR, and V2I connectivity in real-time — this project built and tested a distributed computing platform using prototype cars equipped with heterogeneous sensors and actuators. The architecture distributes analytics between in-vehicle edge processing and cloud computing with real-time guarantees, directly applicable to advanced driving assistance and autonomous vehicle development.

Industrial IoT & Data Analytics

mid-size

Target: Companies offering big data analytics platforms for distributed sensor networks

If you are a data analytics provider whose customers need both instant alerts and deep batch analysis from the same data streams — this project developed a software architecture that transparently combines data-in-motion and data-at-rest analytics across the compute continuum. With 21 deliverables including performance evaluation tools for multi-workload, multi-tenant big data services, the platform handles the kind of mixed workloads that typically require separate systems.

Frequently asked

Quick answers

What would it cost to adopt this technology?

The project was funded as a Research and Innovation Action (RIA), meaning the software outputs are research-grade. Exact licensing or deployment costs are not specified in the project data. You would need to negotiate with the consortium partners — 5 of the 9 are industry players — for commercial terms.

Can this scale to industrial-level deployments?

The architecture was designed specifically for extreme-scale data processing, distributing workloads from edge devices to cloud infrastructure. It was demonstrated on a real smart-city use case with heavy sensor infrastructure across a wide urban area and prototype vehicles, which suggests it handles production-level data volumes. The final release includes scalability components for cloud data analytics services.

Who owns the intellectual property and how can I license it?

As an RIA project, IP is typically retained by the consortium partners who created it. The coordinator is Barcelona Supercomputing Center, and 5 industry partners from 3 countries (Spain, Israel, Italy) hold various components. Licensing terms would need to be discussed directly with the relevant partner holding the specific component you need.

How does this integrate with existing cloud and IoT infrastructure?

The platform integrates Apache Spark and COMPSs programming models within the CLASS architecture, as documented in their final release deliverables. It is designed to work across the full compute continuum from edge to cloud. The architecture includes service management and scalability components that suggest standard cloud deployment patterns.

What is the current status and can I use this today?

The project closed in June 2021 with all final releases completed, including the integrated analytics platform and real-time analysis tools. The software reached final release stage but would likely require adaptation for specific commercial deployments. Contact the consortium to discuss current availability and support.

Is this proven in real-world conditions or just lab testing?

It was demonstrated on a real smart-city use case featuring heavy sensor infrastructure collecting real-time data across a wide urban area, plus prototype cars with heterogeneous sensors, V2I connectivity, and cluster support. This goes beyond lab conditions into real-environment validation.

Consortium

Who built it

The CLASS consortium of 9 partners across Spain, Israel, and Italy is heavily industry-oriented at 56% industry ratio (5 industry, 2 research, 1 university, 1 other). It is led by Barcelona Supercomputing Center, one of Europe's top HPC facilities, which brings serious computational infrastructure credibility. The absence of SMEs and the concentration in just 3 countries suggests this is a focused, heavyweight partnership rather than a broad coalition. For a business looking to adopt this technology, the strong industry presence means the results are more likely to be practically applicable, though the relatively small geographic spread may limit immediate market reach beyond Southern Europe and Israel.

BARCELONA SUPERCOMPUTING CENTER CENTRO NACIONAL DE SUPERCOMPUTACIONCoordinator · ES
COMUNE DI MODENAparticipant · IT
CENTRO RICERCHE FIAT SCPAthirdparty · IT
MASERATI SPAparticipant · IT
ATOS SPAIN SAparticipant · ES
ATOS CONSULTING CANARIAS SA UNIPERSONALthirdparty · ES
ATOS IT SOLUTIONS AND SERVICES IBERIA SLthirdparty · ES
IBM ISRAEL - SCIENCE AND TECHNOLOGY LTDparticipant · IL
UNIVERSITA DEGLI STUDI DI MODENA E REGGIO EMILIAparticipant · IT

How to reach the team

Barcelona Supercomputing Center (BSC-CNS), Spain — a major European HPC center. Contact their technology transfer office for licensing discussions.

Order a report on this consortium See BARCELONA SUPERCOMPUTING CENTER CENTRO NACIONAL DE SUPERCOMPUTACION profile →

Next steps

Talk to the team behind this work.

Want an introduction to the CLASS team to discuss how their edge-to-cloud analytics platform could solve your real-time data processing challenges? SciTransfer can arrange a direct meeting with the right technical contact.

Order a report on this topic More in Digital & ICT