Practitioner's perspective on High Performance Computing services for innovative research and education
2 likes754 views
The document discusses the role of high performance computing (HPC) in innovative research and education, emphasizing the significance of digital transformation in various sectors. It highlights the services provided by the Interdisciplinary Centre for Mathematical and Computational Modelling at the University of Warsaw, including HPC infrastructure and educational programs. Additionally, it presents several case studies showcasing applications of big data analytics in collaboration with Huawei, including notable projects like Copernicus Complexio and OSIRIS.
Practitioner's perspective on High Performance Computing services for innovative research and education
- 1. Practitioner's perspective on High Performance Computing services for innovative research and education Bartosz Borucki Interdisciplinary Centre for Mathematical and Computational Modelling University of Warsaw
- 2. 2 The four paradigms of science • EMPIRICAL – describing natural phenomena (last millennia) • THEORETICAL – building models and generalisations (last centuries) • COMPUTATIONAL – simulating complex phenomena (last decades) • DATA EXPLORATION – „data-intensive” scientific discovery (last years) Jim Gray
- 3. 3 Innovation in research and education KNOWLEDGE DATA COMPUTATIONS INNOVATION DISCOVERY NEW KNOWLEDGE
- 4. 4 Digital transformation (DX) • Digital technology is and will keep on changing the landscape of business, education, entertainment and government • DX is coupling digital technologies with organizational, operational and business models Source: IDC Predictions 2016 By 2017 over 50% of ICT expenses will cover 3rd platform technologies By 2018 DX implementing companies will double IT development resources (programmers) By 2020 60% of ICT expenses will address the cloud
- 6. 6 Challenges & solutions High Performance Computing centres • Centralised infrastructure and e- infrastructure • Simplified management • Optimized cost effectiveness • Adequate answers to crucial challenges • Internalized implementation Externalized services • Infrastructure (IaaS) • Platform (PaaS) • Software (SaaS) • HPC • Knowledge repositories • Data repositories
- 7. 7 ICM @ University of Warsaw University of Warsaw • Largest university in Poland • Leading education and research facility • Celebrating 200 years (founded 1816) Interdisciplinary Centre for Mathematical and Computational Modelling (ICM) • Leading HPC and R&D centre in Poland • Providing research infrastructure within ICT and HPC • Interdisciplinary research • Applied mathematics, computational sciences and data science • Ranked 3rd in Europe and 13th in the World in 2016 Webometrics ranking of research centres (http://research.webometrics.info/)
- 8. 8 ICM services for research and education Over 20 years of experience Concept of Open National Campus • HPC & ICT infrastructure • Cloud services • Software licences • Open knowledge repositories • Open data repositories • Online academic TV • Internet services Virtual Library of Science European Digital Math Library Wireless Internet access for Education Polish Optical Internet Science Services Platform High Performance Computing Polish Grid Infrastructure (NGI)
- 9. 9 Education in HPC @ ICM Computational Engineering • New degree course at ICM • Starting academic year 2016/2017 • Practical profile • Cooperation with business and industry Topics • Applied scientific computations • Large scale computational systems • HPC programming paradigms • Data science • Data processing and visualization • Domain applications • Practical R&D projects
- 10. 10 Huawei-ICM Cooperation • Creation of the Data Science Innovation Centre (HIDSIC) • June 20th, 2016, 4th Poland-China Regional Forum • R&D collaboration • Promotion of innovative solutions in data processing and analytics as well as in cloud computing, high performance computing and large scale modelling
- 11. 11 Huawei supercomputers @ ICM TOPOLA • 240 servers Huawei FusionServer CH121V3 • x86_64 Intel Xeon CPU E5-2697v3 • Over 6000 CPU cores • Over 19 TB memory • Rpeak performance: ~270 TFLOP/s • Generic computing cluster • Part of Polish Grid Infrastructure (PL-Grid)
- 12. 12 Huawei supercomputers @ ICM ENIGMA 谜 • 360 servers Huawei RH1288 V3 • x86_64 Intel Xeon CPU E5-2680v3 • Over 8000 CPU cores • Over 45 TB memory • Over 8PB storage • Rpeak performance: ~340 TFLOP/s • Largest European supercomputer dedicated to Big Data analytics • Hadoop Ecosystem
- 13. 13 Case study – OpenAIRE Bridge between all science stakeholders and the world of scholarly publication • Support for open platforms of science • Open access repositories, archives, journals • Scholarly communication analysis Big Data analytics • >15 millions of research publications in open repositories • Automated content analysis • Social network analysis • Building scholarly communication maps • Mapping collaborations and impacts • Using ENIGMA supercomputer with Hadoop and HDFS Observatory of Europe's ...research output ...OA infrastructures ...policies
- 14. 14 Case study – Copernicus Complexio COCO: Copernicus Complexio • Named after famous Polish astronomer Nicolaus Copernicus • High-resolution view of the small-scale universe • Largescale Cosmological N-body simulation – 13 billions particles • Targeted at dark matter analysis and long term universe evolution Big Data analytics • >200TB of raw simulation data (>600TB with replication) • SQL-like query engine • Post processing • Deep analysis • Using ENIGMA supercomputer • Apache SPARK • Cloudera / Apache Impala MPP database
- 15. 15 Case study – OSIRIS OSIRIS: Open Science and Innovation e-Infrastructure • Targeted at data analytics • Bring analytics to data • ICT/HPC platform integration • Data repositories • Cloud services Big Data analytics e-infrastructure • Providing HPC services • Business • Government • Research • Education • Collaboration with HUAWEI • Implementing FusionInsight - Big Data analytics platform
- 16. THANK YOU Copyright©2016 Huawei Technologies Co., Ltd. All Rights Reserved. The information in this document may contain predictive statements including, without limitation, statements regarding the future financial and operating results, future product portfolio, new technology, etc. There are a number of factors that could cause actual results and developments to differ materially from those expressed or implied in the predictive statements. Therefore, such information is provided for reference purpose only and constitutes neither an offer nor an acceptance. Huawei may change the information at any time without notice.