download0 view1,336
twitter facebook

공공누리This item is licensed Korea Open Government License

dc.contributor.author
여인성
dc.contributor.author
조기현
dc.contributor.author
최원국
dc.date.accessioned
2019-08-28T07:42:14Z
dc.date.available
2019-08-28T07:42:14Z
dc.date.issued
2018-05-01
dc.identifier.issn
0010-4655
dc.identifier.uri
https://repository.kisti.re.kr/handle/10580/14738
dc.identifier.uri
http://www.ndsl.kr/ndsl/search/detail/article/articleSearchResultDetail.do?cn=NART80843286
dc.description.abstract
This paper describes the low-energy physics profiling system in Geant4. A great diversity of physics applications and evolving
computing architecture are used to develop software, and we require profiling tools to evaluate these. For this, we need to draw on
community interests for a collateral effort. There are many efforts to develop software to save CPU time and memory. However,
there are not enough profiling systems. We have developed a low-energy physics profiling system for medical applications,
particularly brachytherapy code. As a starting point, we use the Fermilab National Accelerator Laboratory (FNAL) computing
performance profiling protocol and tools for high-energy physics for the Large Hadron Collider (LHC). The results show that
the newly developed profiling code for low-energy physics (less than MeV) is complementary to the current profiling code for
high-energy physics (greater than TeV).
dc.language
eng
dc.relation.ispartofseries
Computational Physics Communications
dc.title
Performance profiling for brachytherapy applications
dc.citation.endPage
186
dc.citation.startPage
180
dc.citation.volume
226
dc.subject.keyword
Parallel
dc.subject.keyword
GPU
dc.subject.keyword
Particle Physics
dc.subject.keyword
Medical Physics
dc.subject.keyword
Computational Physics
Appears in Collections:
7. KISTI 연구성과 > 학술지 발표논문
Files in This Item:
There are no files associated with this item.

Browse