DocumentCode
344525
Title
Eigenmodes of superconducting cavities calculated on an APE-100 supercomputer (SIMD)
Author
Neugebauer, F. ; van Rienen, Ursula
Author_Institution
DESY Zeuthen
Volume
4
fYear
1999
fDate
1999
Firstpage
2764
Abstract
The construction of modern accelerators is usually supported by the numerical determination of eigenmodes in the accelerating cavities. Often the rotational symmetry of the cavity is used to simplify the numerical simulation. However, in cases where the cavity lacks rotational symmetry resp. where attached components like couplers have to be taken into account, a fully three dimensional treatment of the Maxwell equations is necessary. This requires more computer power than is available on a normal high end workstation. Therefore, in the present approach a parallel SIMD super computer (APE-100) is used to compute the eigenmodes of accelerating cavities. As an example parts of the superconducting TESLA structure are investigated. The geometry input is parsed by MAFIA which transfers the resulting system matrix, incorporating geometry and material information, to the APE-100. The result of the diagonalization procedure is then read back to the MAFIA host where further data analysis and visualization can be done
Keywords
Maxwell equations; accelerator cavities; eigenvalues and eigenfunctions; numerical analysis; parallel algorithms; physics computing; superconducting cavity resonators; APE-100 supercomputer; MAFIA; Maxwell equations; accelerating cavities; attached components; eigenmodes; parallel SIMD super computer; rotational symmetry; superconducting TESLA structure; superconducting cavities; Acceleration; Concurrent computing; Couplers; Data analysis; Data visualization; Information geometry; Maxwell equations; Numerical simulation; Superconducting materials; Workstations;
fLanguage
English
Publisher
ieee
Conference_Titel
Particle Accelerator Conference, 1999. Proceedings of the 1999
Conference_Location
New York, NY
Print_ISBN
0-7803-5573-3
Type
conf
DOI
10.1109/PAC.1999.792930
Filename
792930
Link To Document