Events

Tutorials 2019

“Deep and Machine Learning methods for document clustering and classification”

In this tutorial, we consider a complete workflow of a typical Data Science project dealing with text documents. We define a problem, generate data, analyze data, explore relevant features – discuss several ways how to extract and describe semantic information, and show how to incorporate/augment it by an additional non-semantic one (which might help to improve the results). Next, we consider, construct and apply several standard Machine Learning (ML) models to describe our data:  we cast it to a classification and regression problems.  Then, we analyze an efficiency of the ML methods as well as a role, impact and relevance of our semantic and non-sematic features. Next, we show how to apply Deep Learning methods to attack the same problem – we consider simple DNN (Deep Neural Network) and CNN (Convolutional Neural Network) models. At the end we contrast our ML and DL results, discuss their pluses and minuses: efficiencies, required computational resources, possible way to improve them…

Tutorial supports an active and passive participations. I will use an alive Jupiter Notebook presentation to describe, discuss and execute each end every block of the Python-code requited for the above program/workflow. The corresponding blocks will be shared/available on a dedicated Slack channel (HybriLIT subscription required: https://web-stc.jinr.ru). If you have a valid account on the HybriLIT cluster you will be able to copy/paste them from the Slack channel and re-execute it in on-line mode in your own Notebook via GITLab (https://jhub.jinr.ru/) service. No extra work on your side to install, tune, support the required python packages: JHub – already did it for you.

Registration is available at: https://indico-hlit.jinr.ru/event/146/
SAP Leonardo
Priv.-Doz. Dr Alexej I. Streltsov Senior Data Scientist, ML Deep Learning COE WDF SAP SE, Dietmar-Hopp-Allee 16, 69190 Walldorf, Germany

Tutorials 2018

HybriLIT Team from Laboratory of Information Technologies of the Join Institute for Nuclear Research with colleagues from the Institute of the Experimental and Applied Physics, Faculty of Electrical Engineering University of West Bohemia and Department of Mathematics and Theoretical Informatics, Faculty of Electrical Engineering and Informatics, Technical University of Kosice will hold two-day training course on Parallel programming technologies and high-performance computing within the HybriLIT computing platform at Pilsen and Prague.
Sample DescriptionThe training course by Intel will be held on September 26, 2018 (Wednesday) at the Laboratory of Information Technologies, room 407. Training will be held on the heterogeneous computing platform HybriLIT. To participate in the event, please fill the registration form via link. In case you have any particular questions to be answered by Intel and HybriLIT teams during the training course, please send them to strel@jinr.ru before September 24. Also, if you have any program codes to be optimized using special tools that will be presented by Intel, please upload it to HybriLIT's nfs resource: /nfs/store2.jinr.ru/scratch before September 24.
Intel HybriLit
 

Presentation 2018

Dear colleagues!

The Directorate of the Laboratory of Information Technologies JINR and the companies Intel, RSC Technologies, NVIDIA, IBS Platformix invite you to take part in the presentation of the N.N. Govorun Supercomputer, that will be held on 27 March 2018 at 4:00 PM in the LIT Conference Hall.

 

Agenda

 
  • 4:00 PM Seminar-presentation (LIT Conference Hall)
    • Supercomputer is a promising project of development of the JINR basic facilities, V.V. Korenkov (LIT)
    • Hоt theoretical physics topics for HPC, D.I. Kazakov (BLTP)
    • The NICA mega science project, V.D. Kekelidze (VBLHEP)
    • New Intel architectures and technologies for high-performance computations, N.S. Mester (Intel)
    • Superdense and energy efficient decisions of RSC, A.A. Moskovsky (RSC Technologies)
    • IBS Platformix – Infrastructure decisions, V. Kochetov (IBS Platformix)
  • Excursion to the Multifunctional Information and Computing Complex of JINR (LIT, 2nd floor)
  • 6:00 PM Buffet

Tutorials 2017

During these 3 days you will get an idea of OpenCL, learn how to launch OpenCL programs on the HybriLIT heterogeneous cluster and develop your own applications using OpenCL. In the course of the tutorial, participants will be provided with interesting examples and tasks.
Event program:
Day 1: Lecture on "Introduction to OpenCL for computations using GPU", compilation and launching of OpenCL programs on the heterogeneous cluster.
Day 2: Practice with OpenCL, memory hierarchy, routine algorithms of parallel reduction for binary operations.
Day 3: Writing your own program.
More information at Indico
Advanced course on C/C++ from 16 January 2017 to 6 February 2017
This course is aimed at knowing the C and C ++ language syntax, but wish to continue improving their knowledge. During the training course a lot of attention will be paid to the work of memory, as these skills are the most important for understanding the program is written in C and C ++. Also, will be discus the object model of the C ++ language.
Program:
1) Using the computer memory in program. The difference between references and pointers. Frames Valgrind for finding errors when using the computer's memory.
2) Structures in C and C ++. The difference between the structure and the class. Inheritance and Polymorphism.
3) Operator Overloading. Exceptions.
4) Creating and loading dynamic libraries. Using utility GNU Make.

More information at Indico
The study and use of MPI technology opportunities in the writing parallel programming applications.

MPI - one of the core technologies of parallel programming, which enables communication between parallel processes to address common challenges.

The course will be divided into two parts:
* The first part will be devoted to the basic level and will include minimally necessary set of functions for code parallelization;
* The second part will examine the combined functions to optimize the already written parallel code, as well as profiling and tracing of MPI-applications using Intel Parallel Studio tools.

More information at Indico