{"id":11325,"date":"2023-05-23T11:02:37","date_gmt":"2023-05-23T09:02:37","guid":{"rendered":"https:\/\/www.sling.si\/event\/seminar-large-cluster-programming-and-chapel-language-with-example-of-evolutionary-algorithm-implementation\/"},"modified":"2023-05-23T11:02:37","modified_gmt":"2023-05-23T09:02:37","slug":"seminar-large-cluster-programming-and-chapel-language-with-example-of-evolutionary-algorithm-implementation","status":"publish","type":"event","link":"https:\/\/www.sling.si\/en\/event\/seminar-large-cluster-programming-and-chapel-language-with-example-of-evolutionary-algorithm-implementation\/","title":{"rendered":"Seminar: Large cluster programming and Chapel language with example of evolutionary algorithm implementation"},"content":{"rendered":"

English<\/strong>: <\/span>The application of evolutionary algorithms is frequently limited by available computational resources, especially in the case of so-called hybrid or memetic evolutionary algorithms in combination with large (or even Big) training data sets. HPC implementations not only of such algorithms are complicated. Unified programming environment of Chapel language opens way to implement algorithms working on large clusters much easier way than in the case of frequently used combinations of OpenMP+MPI or even OpenMP+MPI+Cuda technologies. This talk presents an alternative approach to programming heterogeneous systems in the Chapel language, demonstrates its basic approaches and presents their application on evolutionary algorithm parallel implementation.<\/span><\/span><\/span><\/span><\/span><\/p>\n

 <\/p>\n

SLO<\/strong>: Uporaba evolucijskih algoritmov je pogosto omejena z razpolo\u017eljivimi ra\u010dunalni\u0161kimi viri, zlasti v primeru tako imenovanih hibridnih ali memeti\u010dnih evolucijskih algoritmov v kombinaciji z velikimi ali celo<\/span>ogromnimi<\/span> mno\u017eicami<\/span> u\u010dnih <\/span> podatkov. HPC <\/span>i<\/span>zvedbe takih algoritmov so zapletene. Enotno programsko okolje jezika Chapel odpira pot za implementacijo algoritmov, ki delujejo na velikih grozdih, veliko la\u017eje kot v primeru pogosto uporabljenih kombinacij tehnologij OpenMP<\/span> +<\/span> MPI ali celo OpenMP<\/span> +<\/span> MPI<\/span> +<\/span> Cuda.<\/span> T<\/span>o<\/span> predavanje predstavlja alternativni pristop k programiranju heterogenih sistemov v jeziku Chapel, prikazuje njegove osnovne pristope in predstavlja njihovo uporabo pri vzporedn<\/span>i<\/span> implementaciji <\/span>evolucijskih algoritmov.<\/span><\/span><\/span><\/span>
\nThe seminar will be held in English.<\/span><\/span><\/span><\/span><\/p>\n

Difficulty: advanced<\/span><\/span><\/span><\/span><\/p>\n

\n

Organizer<\/strong><\/p>\n

  

<\/div><\/p>\n
\n

Lecturers<\/h3>\n\n\n\n\n\n
\n

Name:<\/p>\n<\/td>\n

\n

Dr. <\/span><\/span><\/span>Tomas Brandejsk<\/span><\/span><\/span><\/p>\n<\/td>\n<\/tr>\n

\n

Description:<\/p>\n<\/td>\n

\n

Doc. dr. Tomas Brandejsky is a member of the Faculty of Electrical Engineering and Informatics at University of Pardubice. He also works at Czech Technical University in Prague. His research interest is especially soft computing (genetic algorithms and fuzzy sets), parallel programming (in Pardubice) and safe software development. (in Prague).<\/span><\/span><\/span><\/p>\n<\/td>\n<\/tr>\n

\n

E-mail:<\/p>\n<\/td>\n

\n