BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Data analisys in high energy phisics using CUDA and OpenCL
DTSTART;VALUE=DATE-TIME:20121018T134500Z
DTEND;VALUE=DATE-TIME:20121018T150000Z
DTSTAMP;VALUE=DATE-TIME:20261011T001605Z
UID:indico-contribution-201@indico.ipb.ac.rs
DESCRIPTION:Speakers: Bogdan Alexandru Dumitru (Institute of Space Science
 )\, Ciprian Mitu (Institute of Space Science)\nWe developed an aplication 
 that allows both the analysis of data and also the real-time visualisation
  on the same time. This is possible with the help of the GPU. To use the p
 ower of GPU is need of an specific framework (we use CUDA and OpenCL).\n  
   The aplication use data generated by UrQMD (Ultra relativistic Quantum M
 olecular Dynamics) to run a variety of algorithms and 3D visualization. Ho
 w it works:\n    As a front-end to the user\, the application uses Qt libr
 aries to create the GUI (graphical user interface library) and user intera
 ction. The input data\, obtained from a SQLite3 database file\, is fetched
  and loaded on the graphic device. After this\, the data is processed on G
 PU using CUDA (Compute Unified Device Architecture) or OpenCL (Open Comput
 ing Language)\, depends on the GPU use. The processed data is output in a 
 OpenGL(Open Graphics Library) viewport and in ROOT's graphs or histograms.
  Using widgets\, one can select/filter events\, frames or particles that a
 re going to be processed. Also\, render all frames in an event continually
 \, such that one sees the evolution of particles as a fluid animation. Jus
 t like movie player\, with buttons to play/pause or go forward\, backward.
 \n        When one wants fast analysis and real-time animation\, one choos
 es parallel processing. Having this in mind\, we put move all processing o
 n the best and cheapest parallel unit one can afford: GPU parallel process
 ing. The framework developed by NVIDIA named CUDA and the framework develo
 ped by Khronos Group named OpenCL  allows one to harness the power of the 
 GPU device using parallel processing. The programming execution model on G
 PU in CUDA is SIMT (Single Instruction Multiple Thread) and in OpenCL is S
 IMD(Simple Instruction Multiple Data). This means\, only one function (run
 ning\nin multiple threads or instances) can run on the GPU at one time pro
 cessing the data. In CUDA and OpenCL terminology\, this function is named 
 a kernel.\n       After querying database for the interested data\, and lo
 ading it in GPU\, depending on the one's action\, a CUDA or OpenCL kernel 
 (depends on the GPU used) is launched accordingly. The kernel\nabove\, tak
 es in particles information (x\,y\,z\,px\,py\,pz\,E\,...) and outputs the 
 computed values of: transverse momentum\, direct and elliptic flow paramet
 ers\, pseudo-rapidity and OpenGL's necessary data to render the particles.
  The representation of the processed data can be seen in OpenGL viewport (
 particles as spheres) and parameters in ROOTs graphs and histograms.\n\nht
 tps://events.saifa.rs/event/291/contributions/201/
LOCATION:National Library of Serbia
URL:https://events.saifa.rs/event/291/contributions/201/
END:VEVENT
END:VCALENDAR