IPAC2015 - List of Authors (Song, K.)

Paper	Title	Page
MOPWI020	Development of Simple Tracking Libraries for ALS-U	1192
	H. Nishimura, D. Robin, K. Song, C. Steier, C. Sun, W. Wan LBNL, Berkeley, California, USA
	Funding: Work supported by the U.S. Department of Energy under Contract No. DE-AC02-05CH11231 The conceptual lattice design study of a new diffraction-limited light source has become much more computer intensive than that for the 3rd-generation rings. We are in a process of upgrading our existing accelerator modeling and simulation libraries, Goemon* in C++ and a new version Tracy#, to fulfil such new demand. The C++ version has been actively used on the ALS HPC cluster for multi-objective optimization (MOGA) to optimize the ALS lattice, and recently for ALS-U*. This time, based on the current version in C#, we extracted its subset and ported it to C and C++. The routines are made thread-safe to enable OpenMP locally, and CPU-time profiling was extensively used to remove redundancies. The new refitting method of quad settings brought smooth switching from 5-dim to 6-dim. The data structure itself is simplified for the use on GPU that is based on our previous effort of tracking particles in GPU**. Tracy# itself is also upgraded to cooperate with these C/C++ versions. Their use from Python will be also mentioned. H. Nishimura, PAC01, 3066-3068. H. Nishimura, ICAP09. * C. Sun, et. al.,PAC11, 793-795. ** H. Tarawneh, et. al.,J.Phys.493 012020, 2014. *** H. Nishimura, et. al.,PAC11, 1764-1766.
DOI •	reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2015-MOPWI020
Export •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

Paper

Title

Page

Development of Simple Tracking Libraries for ALS-U

1192

H. Nishimura, D. Robin, K. Song, C. Steier, C. Sun, W. Wan
LBNL, Berkeley, California, USA

Funding: Work supported by the U.S. Department of Energy under Contract No. DE-AC02-05CH11231
The conceptual lattice design study of a new diffraction-limited light source has become much more computer intensive than that for the 3rd-generation rings. We are in a process of upgrading our existing accelerator modeling and simulation libraries, Goemon* in C++ and a new version Tracy#**, to fulfil such new demand. The C++ version has been actively used on the ALS HPC cluster for multi-objective optimization (MOGA) to optimize the ALS lattice***, and recently for ALS-U****. This time, based on the current version in C#, we extracted its subset and ported it to C and C++. The routines are made thread-safe to enable OpenMP locally, and CPU-time profiling was extensively used to remove redundancies. The new refitting method of quad settings brought smooth switching from 5-dim to 6-dim. The data structure itself is simplified for the use on GPU that is based on our previous effort of tracking particles in GPU*****. Tracy# itself is also upgraded to cooperate with these C/C++ versions. Their use from Python will be also mentioned.
* H. Nishimura, PAC01, 3066-3068.
** H. Nishimura, ICAP09.
*** C. Sun, et. al.,PAC11, 793-795.
**** H. Tarawneh, et. al.,J.Phys.493 012020, 2014.
***** H. Nishimura, et. al.,PAC11, 1764-1766.

DOI •

reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2015-MOPWI020

Export •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)