Proceedings of the 2012 IEEE International Symposium on Antennas and Propagation 2012
DOI: 10.1109/aps.2012.6348419
|View full text |Cite
|
Sign up to set email alerts
|

Efficient method of moment simulation based on higher order bases and CPU/GPU parallelization

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4

Citation Types

0
7
0

Year Published

2014
2014
2014
2014

Publication Types

Select...
1

Relationship

0
1

Authors

Journals

citations
Cited by 1 publication
(7 citation statements)
references
References 1 publication
0
7
0
Order By: Relevance
“…The authors of [25], [26] used a thread to deal with those integrations caused by one pair of patches, which is ideologically similar to the method in [27]. Note that the method in [27] is for the RWG basis functions rather than the higher order basis functions.…”
Section: Strategy Comparisonmentioning
confidence: 99%
See 4 more Smart Citations
“…The authors of [25], [26] used a thread to deal with those integrations caused by one pair of patches, which is ideologically similar to the method in [27]. Note that the method in [27] is for the RWG basis functions rather than the higher order basis functions.…”
Section: Strategy Comparisonmentioning
confidence: 99%
“…For example, when when . If all the integrations caused by a pair of patches were assigned to one thread as in [25], [26], then at least single-precision or double-precision units in the on-chip memory of this thread should be enabled for the high efficiency, which is not consistent with GPU-oriented programming. In that case, the NVCC complier will automatically enable the local memory of this thread as a supplement [21], leading to the significant access latency because the local memory is on-board like the GPU global memory, which is harmful to the performance of the whole program.…”
Section: Strategy Comparisonmentioning
confidence: 99%
See 3 more Smart Citations