LEXIS project focuses on the convergence of large-scale HPC and cloud-run data analytics workflows. The emphasis of LEXIS will be on how HPC and cloud systems interact; how they can share data; and methods to compose workflows of tasks running on both cloud and HPC systems. LEXIS will develop infrastructure to enable these workflows and demonstrate its abilities through three large-scale socio-economic pilots, targeting aeronautics, weather climate, and catastrophe alert systems.
LEXIS OVERALL OBJECTIVES
Build a distributed HPC infrastructure for Big data and HPC convergence. The aim will on build an advanced architecture for big data analytics and HPC applications leveraging modern technologies from HPC, Big data and Cloud computing.
Validation from three large scale pilots. The aim will on improving performances in term of computational time and data management. The aim will on demonstrate benefits by creating an ecosystem for industrial applications context.
LEXIS infrastructure, platform ad services will be extended by Open call for external stakeholders mainly from existing pilots test-beds in E-science and industrial sectors: Aeronautics, Weather and Climate, and Earthquake and Tsunami.
LEXIS TEST-BED-SPECIFIC OBJECTIVES
The Aeronautics Large-scale Pilot lead by Avio Aero company in LEXIS is aimed to boost and promote a step change in the numerical investigations of complex fluid-dynamic behaviour of critical aeronautical engines’ components, allowing to improve their engineering design quality so as to optimize engine performance (specific fuel consumption and emissions) and to make lighter and greener aircraft.
Framed in this challenging context, the industrial applicability of LEXIS platform will be investigated both to reduce the running time of CPU-intensive and time-consuming engineering jobs and to enhance data handling and post-processing operations of the large dataset collected as jobs’ results.
O2.2) Weather and climate.
Within the context of Copernicus Services, the Weather & Climate pilot will increase the timeliness and quality of prediction and analyses.
Simplify the access to such services from the cloud, in order to expand the downstream markets: emergency management, sustainable food and energy production, air quality
O2.3) Earthquake and tsunami.
Earthquake and tsunami. Provide near real-time earthquake and tsunami damage/loss assessments and estimate of the tsunami inundation through simulations based on earthquake parameters, ensuring the delivery of expected consequences in time for fast response planning by emergency dispatchers.
LEXIS TECHNOLOGY-SPECIFIC OBJECTIVES
O3.1) Providing a ready-to-be-used HPC infrastructure that offers HPC-as-a-Service capabilities without incurring in performance/ efficiency slowdowns.
O3.2) Implement a heterogeneous data storage management system providing simplified access to huge amounts of data.
O3.3) Speed up CPU intensive and data/memory intensive algorithms also to support real time decision making.
O3.4) Optimize data management operations and analytics algorithmsthat exploit the underlying infrastructures at their best to eventually extract outputs from data that help stakeholders improve their businesses.
O3.5) Provide simple and secure HPDA service provisioning, through cloud technologies, for the pilot test-beds, accessible also for other users.
O3.6) Guarantee interoperability with external data sourcesand seamless integration with external systems.