Second Action WGs Workshop and Fourth MC meeting


When:
September 10, 2015 – September 11, 2015 all-day
2015-09-10T00:00:00+02:00
2015-09-12T00:00:00+02:00
Where:
Room 1.20. Dep. of Computer Science. AGH University of Science and technology.
D 17
Kawiory 21, 30-055 Kraków
Poland
Contact:
Roman Wyrzykowski

 

 

NESUS Second Workshop. Call For Papers

 

Proceedings (preliminary draft)

 

Proposed Agenda

 

NESUS – Second Action Workshop Program
Date Session Name Topic
10th 9:00-9:15 Jesus Carretero Workshop Opening
9:15-10:45 Invited speakers
 Chairs:Roman Wyrzykowski &Jesus Carretero Willem Deconinck
Accelerating extreme-scale Numerical Weather Prediction
European Centre for Medium-Range Weather Forecasts
Pawel Plaszczak Red Stack Tech
Joanna Kolozdiej
High-performance Modelling And Simulation For Big Data Applications (Chipset)
Action IC4106 Chair
10th 10:45–11:15 Coffe break Poster session
10th 11:15–13:00  Paper presentation.Chair: Helen Karatza
  • Pilar Gonzalez-Ferez and Angelos Bilas. NUMA impact on network storage protocols over high-speed raw Ethernet
  • Ricardo Morla, Pedro Gonçalves and Jorge Barbosa. A Scheduler for Cloud Bursting of Map-Intensive Traffic Analysis Jobs
  • Gabor Kecskemeti. Scheduler hierarchies for enabling peta-scale cloud simulations with DISSECT-CF
  • Robert Dietze, Michael Hofmann and Gudula Ruenger. Exploiting Heterogeneous Compute Resources for Optimizing Lightweight Structures
  • Francisco Rodrigo Duro, Fabrizio Marozzo, Javier Garcia Blas, Jesús Carretero, Domenico Talia and Paolo Trunfio. Evaluating data caching techniques in DMCF workflows using Hercules
10th 13:00-14:00 Lunch
10th 14:00-16:00  Paper presentationChair: Magne Haveraaen
  • Pablo Llopis Sanmillán, Manuel F. Dolz, Javier Garcia Blas, Florin Isaila, Jesús Carretero, Mohammad Reza Heidari and Michael Kuhn. Analyzing power consumption of I/O operations in HPC applications
  • Ilias Mavridis and Eleni Karatza. Log File Analysis in Cloud with Apache Hadoop and Apache Spark
  • Dana Petcu. On Autonomic HPC Clouds
  • Jing Gong, Stefano Markidis, Michael Schliephake, Erwin Laure, Luis Cebamanos, Alistair Hart and Paul Fischer. NekBone with Optimized OpenACC directives
  • Raimondas Ciegis. Distributed Parallel Computing for Visual Cryptography Algorithms
  • Neki Frasheri. PARALLEL PROCESSING FOR GRAVITY INVERSION
 16:00-16:30  Coffe break Poster session
10th 16:30–17:50  Paper presentation.Chair:  Trinh Anh Tuan
  • Beat Wolf, Pierre Kuonen and Thomas Dandekar. Multilevel parallelism in sequence alignment using a streaming approach
  • Anatoliy Melnyk, Viktor Melnyk and Lyubomyr Tsyhylyk. Chameleon© C2HDL Design Tool In Self-Configurable Ultrascale Computer Systems Based On Partially Reconfigurable FPGAs
  • Biljana Stamatovic. Labeling connected componets in binary images based on cellular automata
  • Atanas Hristov. Nature-Inspired Algorithm for Solving NP-Complete Problems
10th 17:50 Workshop Closing
10th 18:40 Excursion
20:00 Social Event
NESUS – Management Committee 
11th 9:00–11:00 MC meeting
11th 11:00-11:30 Coffe break Poster session
11th 11:30–13:30 WG meetings & Research roadmap
11th 13:30-14:00 Meeting close

 

Invited Speakers

Willem Deconinck, M.Sc.Eng. Accelerating extreme-scale Numerical Weather Prediction
Numerical Weather Prediction (NWP) and climate simulations have been intimately connected with progress in supercomputing since the first numerical forecast was made about 65 years ago. The biggest challenge to state-of-the-art computational NWP arises today from its own software productivity shortfall. The application software at the heart of all NWP services throughout Europe is ill-equipped to efficiently adapt to the rapidly evolving heterogeneous hardware provided by the supercomputing industry. If this challenge is not addressed it will have dramatic negative consequences for weather and climate prediction and associated services.
To address the challenge of extreme-scale, energy-efficient high-performance computing, the ECMWF is leading the ESCAPE project, a European funded project involving regional NWP centres, Universities, HPC centres and hardware vendors. The aim is to combine interdisciplinary expertise for defining and
co-designing the necessary steps towards affordable, Exascale high-performance simulations of weather and climate.willen-deconnick
Willem Deconinck is working at the European Centre for Medium Range Weather Forecasts (ECMWF) in the numerical aspects section, which is researching and maintaining the dynamical core of the Integrated Forecasting System (IFS). Willem is an aerospace engineer by education at the Free University of Brussels, with expertise in solving partial differential equations (PDE’s) using unstructured meshes with high-order discontinuous schemes. His work at ECMWF involves the development of Atlas, a flexible object-oriented parallel data structure framework to be used as foundation for new developments at ECMWF.
Pawel Plaszczak. Big Data Analytics – a commercial perspective

Abstract:
At Red Stack Tech, I work a lot with Big Data. I will explain why I never use this term, but instead I like to talk about Analytics. I will briefly cover the commercial terminology. I will explain why and where companies use Big Data today, where it is practical and useful, and where it is not possible or simply pointless. I will contradict some common myths: in fact, Big Data technologies do not need terabytes, technologies like Hadoop or paradigms like Cloud are not central to what we do, and mining data from social networks is… not interesting. Instead, I will use some examples from our existing data analytics projects, which we find more practical and more useful.

BIO:

Pawel Plaszczak is the Global Analytics Lead for the Red Stack Tech group. Before, Pawel was the Founder and CEO of Gridwise.pl. Pawel’s team delivered over 150 projects to Global 500 corporations, academic consortia and government agencies worldwide. When Red Stack Tech purchased Gridwise in December 2014, Pawel’s new role became the leader of the data analytics strategy worldwide. Pawel authored articles, presentations, and one book. He blogs at blog.gridwise.pl (Polish) or bigdatamatters.com (English).

Joanna Kolodziej
HIGH-PERFORMANCE MODELLING AND SIMULATION FOR BIG DATA APPLICATIONS (cHiPSet).The Big Data era poses a critically difficult challenge and striking development opportunities in High-Performance Computing (HPC): how to efficiently turn massively large data into valuable information and meaningful knowledge. Computationally effective HPC is required in a rapidly in-creasing number of data-intensive domains, such as Life and Physical Sciences, and Socioeconomic Systems. Modelling and Simulation (MS) offer suitable abstractions to manage the complexity of analysing Big Data in various scientific and engineering domains. Unfortunately, Big Da-ta problems are not always easily amenable to efficient MS over HPC. Also, MS communities may lack the detailed expertise required to exploit the full potential of HPC solutions, and HPC architects may not be fully aware of specific MS requirements.
cHiPSet action has been created to build cross-European coordination to facilitate interactions among data intensive MS and HPC experts, ensuring that the field, which is strategic and of long-standing interest in Europe, develops efficiently – from academic research to industrial practice. cHiPSet provides the integration to foster a novel, coordinated Big Data endeavour supported by HPC. It l strongly supports information exchange, synergy and coordination of activities among leading European research groups and top global partner institutions, and promotes European software industry competitiveness.
In this talk, cHiPSet Cost Action will be presented together with few examples of recent devel-opments in modeling and simulation in data intensive computing.
joanna-kolodziejJoanna Kołodziej (PhD, DsC) is Associate Professor at Cracow University of Technology (Cracow, Poland). She completed her Ph.D. (with honours) in Theoretical Computer Science at Jagiellonian University in Cracow in 2004, and habilitation (DsC) in Institute of Computer Science of Polish Academy of Sciences in Warsaw, in February 2013. The current topics of her research include artificial intelligence, multiagent systems, grid and cloud computing, energy effectiveness in large scale distributed systems, data intensive computing. Prof. Kołodziej has been awarded for the best MSc and PhD dissertations in theoretical mathematics and theoretical computer science by Polish Mathematical Society and Foundation for Polish Science. Prof. Kolodziej is a Principal Investigator and Chair of the big COST Action- CHIPSET –High Performance Modelling and Simula5on for Big Data Applications(Contract IC1406. She is the author of one monograph, and co-author of 3 other books, guest editor of over 20 special issues of major journals, and author and co-author of over 150 scientific papers in peer-reviewed journal, and conference proceedings. Her full research and expertise profile is available at www.joannakolodziej.org.

 

Hotels

 

The following hotels are near the conference location: