Severo Ochoa

From AnnualReport

Jump to: navigation, search

BSC-CNS - Severo Ochoa Center of Excellence

Mateo Valero, Director BSC-CNS and Principal Investigator of the Severo Ochoa Project

BSC-CNS received the “Severo Ochoa Centers of Excellence" support and accreditation in the 2011 call, a programme of the Secretary of State for Research, Development and Innovation of the Ministry of Economy and Competitiveness that aims to promote excellence in scientific research in Spain. The award confirmed BSC-CNS’s reputation as one of the world’s leading centres applying and doing research on supercomputing technologies. The programme commenced in January 2012 with an ambitious research programme: to design novel hardware and software technologies to address the computational and Big Data requirements of three challenging applications in personalised medicine, biomechanics and high-resolution air quality climate modelling.

In addition to improving the capacity of world-class research centres to organise and carry out research, the programme also plans to consolidate best practices in support services, based on the recognition that efficient management of human resources, training and communication are key elements for promoting outstanding results in research.

The BSC-CNS Severo Ochoa Project

The BSC-CNS Severo Ochoa research programme takes as its starting point the fact that Exascale supercomputers will be crucial in tackling major societal problems in areas such as human health and the environment. The computational requirements of the numerical simulations in these scientific fields will go together with new requirements in terms of data management and storage, to so called Big Data. However, attaining Exascale and Big Data poses significant challenges, due to requirements on system hardware and software design, of which energy consumption is one of the main factors.

A picture is worth 1000 words: BSC-CNS's Severo Ochoa programme

The image on the left summarizes the BSC-CNS Severo Ochoa programme, which is made up of four main scientific sub-projects: the proposal and development of novel models in three application domains which rotate around novel components in the hardware and software stacks to achieve the Exascale and Big Data. In the following paragraphs the applications and achievements during 2014 are described.

2014 was a year of consolidation of various important developments of the programme, such as the deployment of improvement mechanisms in the management of Human Resources in Research, the empowerment of and networking between young researchers, conducting 12 sessions of Severo Ochoa Research Seminars (SORS), the celebration of the first BSC-CNS doctoral Symposium, the Amber, E2SCMS2 and PUMPS Summer Schools, as well as the mobility programme and MIRI scholarships for MSc in HPC at the UPC. Finally, the continued promotion of the BSC-CNS communication and outreach programmes aimed at engaging society with BSC-CNS activities.

Specific activities within the main scientific pillars of the Severo Ochoa programme included:

  • Personalised medicine (Life Sciences Department): The combination of genomics, proteomics and transcriptomics analysis with computer simulation is being explored to create models capable of predicting certain diseases or the effects of drugs on patients before a pathology starts.
  • Multi-scale air quality climate modelling (Earth Sciences Department): Further developments were made on a high-resolution modelling system for short- and long-term applications, as well as new methodologies to run high-resolution models on exascale machines and methodologies to handle big amounts of data produced by complex high-resolution models.
  • Computational biomechanics--Alya Red (Computer Applications in Science and Engineering Department): Implementation and validation of various advanced physiological models were carried out on a new Cardiac Computational Model (CCM) developed using the BSC-CNS Alya software platform.
  • Novel components in the hardware and software stacks to achieve the Exascale and Big Data (Computer Sciences Department): During 2014 the Computer Sciences Department continued the development of novel technologies to facilitate the implementation of the applications developed by the scientific departments, including: 1) the adaptation of the BSC-CNS's OmpSs programming mode for energy-efficiency and the development of PyCOMPSs workflow applications with BigData demands; 2) design of methodologies and technologies to support data organisation in key/value non-relational databases; 3) development of a new library for self-contained objects (dataClay), as as an alternative to the use of files and its integration with the COMPSs programming model and data organisation components; and 4) design of Domain Specific Languages (DSL) for accelerator-based clustered architectures.
Josep Casanovas, Executive Manager of the BSC-CNS Severo Ochoa Project
Eduard Ayguadé, Scientific Coordinator of the project and Associate Director of the Computer Sciences Department of BSC-CNS
Personal tools