At the BDEC plenary on 15 May, during the EuroHPC Summit Week 2019 in Poznan, Poland, Rosa M Badia from BSC talked about the synergies between the ELASTIC and the mF2C European projects.

In her presentation, titled “Workflows for continuum computing platforms (from edge to HP computing)”, Rosa explained how ELASTIC is taking benefit of the mF2C project with PyCOMPSs, which allows task-based applications to be built that are orchestrated along ELASTIC’s complex computing infrastructure.

“The audience was very receptive to the presentation, which was well appreciated. I received very positive feedback from the audience, including the European Commission,” said Rosa.

ELASTIC’s software architecture will be tested in the city of Florence, Italy, and equips tram vehicles with V2X communication and a variety of detectors aimed at improving traffic management, allowing predictive maintenance and enhancing the tramway operation.

The project uses a complex infrastructure with sensors and instruments in the edge that generate data and computation that will be done across the compute continuum, from edge to cloud. Executing applications on such distributed environments is complex, especially if real-time constraints need to be considered as well. The PyCOMPSs + dataClay technologies initially developed by mF2C will be enhanced to help to address these challenges.

On 26-28 March, ELASTIC researchers showcased the project’s technology at the DATE 2019 conference, which took place in Florence, Italy. ELASTIC had a joint booth with CLASS project, as part of DATE’s thought-provoking exhibition. The joint booth was kept busy as visitors from different backgrounds were eager to learn about the project’s aims and real-life applications.

The name of the booth was “A Software Architecture for Extreme-Scale Big-Data Analytics in Fog Computing Ecosystems” and the aim was to show ELASTIC’s novel software architecture to distribute resources from edge to cloud. The experts showed how that the software includes the development of a new concept of elasticity, ensuring smooth performance across the compute continuum in an innovative fog-computing environment.

Researcher Maria A. Serrano, based at the Barcelona Supercomputing Center (BSC), was one of the ELASTIC and CLASS researchers that were present at the event. “It was a pleasure to welcome so many visitors to our booth at DATE 2019 and to be able to showcase our technology with a live demonstration and videos. This was a great opportunity to interact with people in the field while presenting our research and receiving useful feedback”.

Visitors were informed about ELASTIC’s smart-mobility use case, which deploys a sensor network to collect the data from the tram network of Florence, Italy. The trams will feature advanced embedded architectures, V21 connectivity, heterogeneous sensors and access to cloud resources, improving public/private transport interaction and allowing the creation of advanced driving assistance applications.

About the event

The DATE conference combines the world’s favorite electronic systems design and test conference with an international exhibition for electronic design, automation and test, from system-level hardware and software implementation right down to integrated circuit design. Out of a total of 834 paper submissions received this year, a large share (38%) is coming from authors in Europe, 28% of submissions are from the Americas, 33% from Asia, and 1% from the rest of the world. This distribution demonstrates DATE’s international character, global reach and impact.

The ELASTIC project, recently funded by the European Union’s H2020 programme will develop a novel architecture for extreme-scale analytics. It will incorporate a novel elasticity concept to distribute computational resources across the compute continuum. The ELASTIC software architecture will form the technological basis for advanced mobility systems and autonomous transport networks.

The architecture will be used to deploy a smart mobility use-case in a real urban area within the City of Florence (Italy). The data generated by the tram network and trams vehicles themselves will be collected and processed using technologies such as embedded architectures, heterogeneous sensors, V2X connectivity and cloud resources.

Eduardo Quiñones, ELASTIC project coordinator at the Barcelona Supercomputing Center (BSC), said: “ELASTIC will allow the project to extend the concept of elasticity by developing a new type of distributed system capable of autonomously scaling computing and storage resources across the compute continuum, from edge to cloud”.  He went on to state: “This novel elasticity concept will not only increase the performance of distributed extreme-scale big-data analytics systems, but it will also achieve the fulfilment of other non-functional requirements such as real time, energy or security, which are crucial for the development of advanced and autonomous mobility systems of the future.”

As the project is set in the context of fog computing —an extended version of cloud computing to the network edge - it is suitable for internet of things (IoT) and applications that require real-time capabilities and the processing of Big Data.

In current architectures, when data analytics is moved to the cloud, the processing time and energy cost of computation is reduced, but the performance of the system (in terms of latency) is adversely affected, making it impossible to derive real-time guarantees. In addition, the level of security required increases to minimise potential attacks, which may end up affecting the safety assurance levels.

The ELASTIC technology will address these challenges while providing the technological background for the developing of new mobility services.

Image: ELASTIC launched its activities in December 2018 at Barcelona Supercomputing Center (BSC)