Wednesday, 22 June 2022 14:00 – 17:30

EVEREST: Design and Programming High-performance, distributed, reconfigurable
and heterogeneous platforms for extreme-scale analytics


Francesco Regazzoni, Univerisity of Amsterdam, The Netherlands and Università della Svizzera italiana, Switzerland
Jeronimo Castrillon, TU Dresden, Germany
Antonella Galizia, IMATI-CNR, italy and CIMA Foundation, Italy
Christian Pilato, Politecnico di Milano, Italy

Workshop Description:

The heterogeneous and distributed nature of the data involved in extreme-scale analytics, requires the development of novel computing systems seamlessly combining HPC, EDGE and IoT devices. Seamless integration is however not sufficient, since the whole power of these computing systems can be unleashed only when the problem is addressed holistically and the computation and communication of the system are jointly co-designed. This ultimately means to conceive novel design approaches, necessary data-driven, and to give designers the tools needed to simplify the programmability of such heterogeneous and distributed architectures.

The EVEREST Workshop aims at presenting relevant cases of study and the latest achievements in the field of design and co-design methods, domain specific languages, computational frameworks, and workflow organization in high performance platforms for extreme-scale analytics. Attendees will be exposed to the main challenges and opportunities in the domain, will gather information on the recent research results achieved by the most recent collaborative project, and will have the possibility to exchange ideas and discuss about open problems with academic and industrial stakeholders.

14:00 – 14:05 Welcome and Introduction

14:05 – 15:00 Keynote “Integrating HPC, AI, and data analytics with eFlows4HPC”

Speaker: Rosa M. Badia, Barcelona Supercomputing Center

Abstract: Distributed computing infrastructures are evolving from traditional models to environments that involve sensors, edge devices, instruments, etc, and, as well, high-end computing systems such as clouds and HPC clusters. A key aspect is how to describe and develop the applications to be executed in such platforms. What is more, the use of data analytics and artificial intelligence in general is on high demand in current HPC applications. However, the methodologies to develop such workflows that integrate HPC simulations and data analytics is not well integrated. eFlows4HPC project aims at providing workflow software stack and an additional set of services to enable the integration of HPC simulations and modelling with big data analytics and machine learning in scientific and industrial applications. The project is also developing the HPC Workflows as a Service (HPCWaaS) methodology that aims at providing tools to simplify the development, deployment, execution and reuse of workflows. The project will demonstrate its advances through three application Pillars with high industrial and social relevance: manufacturing, climate and urgent computing for natural hazards; these applications will help to prove how the realization of forthcoming efficient HPC and data-centric applications can be developed with new workflow technologies. The talk will present the motivation, challenges and project workplan.

15:00 – 15:30DAPHNE: An Open and Extensible System Infrastructure for Integrated Data Analysis Pipelines
Speaker: Matthias Boehm (TU Graz) 
Link to the Presentation: DAPHNE

14:15 – 14:45 Coffee Break

14:15 – 14:45 EVEREST Presentations
16:00 – 16:30 Domain specific languages and compilation flows in the EVEREST project
Speaker: Jeronimo Castrillon (TU Dresden)  Link to the Presentation: DSL
16:30 – 17:00 The LEXIS Platform: Easy access to heterogenous computational workflows execution
Speaker: Martin Golasowsk (IT4Innovations) Link to the Presentation: LEXIS
17:00 – 17:30 A journey of cloud FPGAs on federated data centers
Speaker: Dionysios Diamantopoulos (IBM)  Link to the Presentation: ESPER