Численное решение дифференциальных и интегральных уравнений
High performance computing powered by increased investments in software and hardware infrastructures is beginning to have a significant impact on accelerator design and analysis. The US Department of Energy (DOE) funded Accelerator Grand Challenge has laid the groundwork for developing a suite of electromagnetic codes that are based on unstructured grids and utilize parallel processing on supercomputers such as the Cray T3E and IBM SP2 as well as PC clusters. We will show how this new capability has enabled some of the most challenging problems in accelerator modeling to be solved in resolution, accuracy and turnaround time previously not possible. We will describe the technologies and resources required to support such large-scale simulations, and will discuss the benefits to present and future accelerator facilities from advancing simulation as the third tool of science. Further code development plans under the newly approved DOE SciDAC Initiative will be presented.
Примечание. Тезисы докладов публикуются в авторской редакции
Ваши комментарии Обратная связь |
[Головная страница] [Конференции] |
© 1996-2000, Институт вычислительных технологий СО РАН, Новосибирск
© 1996-2000, Сибирское отделение Российской академии наук, Новосибирск
Дата последней модификации: 06-Jul-2012 (11:45:20)