BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Speeding up Scientific Codes in HPC Architectures by Code Moderniz
 ation: Lessons Learned
DTSTART;VALUE=DATE-TIME:20190605T090000Z
DTEND;VALUE=DATE-TIME:20190605T100000Z
DTSTAMP;VALUE=DATE-TIME:20190320T122524Z
UID:indico-event-777114@indico.cern.ch
DESCRIPTION:The computational resources required in scientific research fo
 r key areas\, such as medicine\, physics\, bioinformatics or climate model
 ling\, are continuously increasing every year. To meet this demand\, high-
 performance computing (HPC) systems keep growing in scale and complexity. 
 However\, the race for higher processor frequency is (temporally) over. Th
 e most profitable way to use the available silicon resources is offered by
  new-generation parallel multi/many core architectures\, such as NVIDIA GP
 Us\, or Intel Xeon Scalable or Intel Xeon Phi. These processors put forwar
 d new opportunities to enhance scientific computations\, also increasing t
 he performance per watt\, but shifting to a different programming model to
  exploit the parallelism (task\, data and thread-level). However\, both de
 velopers and users are usually reluctant to modify their working codes to 
 adapt to new systems. While the benefits of migrating codes to new systems
  are clear\, it is important to evaluate how to do so in a simple and gene
 ral way. ‘‘Code modernization’’ is a new paradigm that aims to pro
 vide both code and performance portability. In this talk\, we identify the
  key issues that determine performance in modernized code\, such as: (a) t
 he ability to scale as with core count\, (b) ensure a proper usage of the 
 vectorization capabilities of the system and (c) the exploitation of data 
 locality. We show three use cases from key applications used in computatio
 nal science: 1) 3-D Stencil-based codes as they are the basis for solving 
 partial differential equations (PDEs)\, which are widely used as a mathema
 tical model in many applications from a wide variety of fields of science 
 and engineering\, 2) A population based metaheuristic for solving NP-hard 
 optimization problems\, such as the Traveling Salesman Problem\, called An
 t Colony Optimization (ACO)\, which is a bio-inspired method\, based on an
 t’s foraging process and\, 3) A Semantic Web Integration Tool (SWIT) tha
 t transforms and integrates heterogeneous biomedical data for generating o
 pen semantic repositories defining mapping rules between an input schema a
 nd an OWL ontology. The talk will shed light on modernizing these use case
 s to new Intel architectures. \n\nBIOGRAPHY OF JOSÉ M. GARCÍA\n\nJosé 
 M. García is a full professor of Computer Architecture at the Department 
 of Computer Engineering at the University of Murcia (Spain). He served as 
 Director of the Computer Engineering Department from 1998 till 2004\, and 
 later\, served as the Dean of the School of Computer Science for seven yea
 rs (2006-2012). He has developed several courses on Computer Structure\, C
 omputer Architecture\, Parallel Computer Architecture\, Peripheral Devices
 \, and Multicomputer Design. He was involved in the “EA-Grid: Euro-Asia 
 United Establishment of Double Degree Master Programme in Grid Computing
 ”\, which was an Education and Research Network in Grid Computing betwee
 n the EU and Asia funded by the European Commission. Prof. García is the 
 Head of the Research Group on Parallel Computer Architecture. His current 
 research interests lie in the design of power-efficient heterogeneous syst
 ems\, the code development and modernization of data-intensive application
 s for those systems (especially bioinspired evolutionary algorithms\, and 
 bioinformatics apps for new drug discovery)\, and HPC architectures for De
 ep Learning algorithms. He has supervised seventeen doctoral Theses and ha
 s published more than 170 refereed papers in different journals and confer
 ences in these fields. Prof. García is member of HiPEAC\, the European Ne
 twork of Excellence on High Performance and Embedded Architecture and Comp
 ilation\, and also member of several international associations such as th
 e IEEE and ACM.\n\nhttps://indico.cern.ch/event/777114/
LOCATION:CERN 31-3-004 - IT Amphitheatre
URL:https://indico.cern.ch/event/777114/
END:VEVENT
END:VCALENDAR