Á¤º¸°úÇÐȸ ³í¹®Áö C : ÄÄÇ»ÆÃÀÇ ½ÇÁ¦
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
GPU »ó¿¡¼ ´ÙÁßÇà·Ä󸮸¦ ÀÌ¿ëÇÑ Hessenberg Ãà¾à ¾Ë°í¸®ÁòÀÇ °í¼Ó ¼öÇà ±â¹ý |
¿µ¹®Á¦¸ñ(English Title) |
A Fast Solution of Hessenberg Reduction Algorithm Using Multi-Matrix Processing on GPU |
ÀúÀÚ(Author) |
Àª ¸® ¾ö
ÀÌ ¿µ ÁÖ
¹Ú ÀÎ ±Ô
Williem
Youngjoo Lee
In Kyu Park
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 18 NO. 07 PP. 0509 ~ 0514 (2012. 07) |
Çѱ۳»¿ë (Korean Abstract) |
º» ³í¹®¿¡¼´Â GPU º´·Äó¸® ȯ°æ¿¡¼ ´ÙÁßÇà·Ä󸮸¦ ÅëÇÑ Hessenberg Reduction ¹®Á¦ÀÇ °í¼Ó ó¸® ±â¹ýÀ» Á¦¾ÈÇÑ´Ù. Hessenberg Reduction ¹®Á¦´Â °íÀ¯Ä¡ ¹®Á¦¸¦ ÇØ°áÇϱâ À§ÇÑ ÇʼöÀûÀÎ ´Ü°è ·Î¼, À̸¦ ÇØ°áÇϱâ À§ÇØ ¸¹Àº °è»ê·®ÀÌ ÇÊ¿äÇÏ´Ù. ±âÁ¸ÀÇ GPU ±â¹ÝÀÇ Ã³¸® ±â¹ýÀº ÀÛÀº Å©±âÀÇ Çà·Ä󸮿¡¼ CPU¿¡¼ÀÇ Ã³¸®¿¡ ºñÇØ È¿À²ÀûÀÌÁö ¸øÇÏ´Ù. º» ³í¹®¿¡¼´Â GPUÀÇ ÀÚ¿øÀÌ Çã¿ëÇÏ´Â ÃÖ´ëÀÇ Çà·ÄÀ» µ¿½Ã¿¡ ó¸®ÇÔÀ¸·Î½á CPU¿Í GPU°£ÀÇ ¸Þ¸ð¸® Àü¼Û·üÀ» ³ôÀÌ°í ¶ÇÇÑ GPUÀÇ ´ë¿ë·® º´·Äó¸® ±¸Á¶¿¡ ºÎÇÕÇÏ¿© °í¼Ó ¼öÇàÀÌ °¡´ÉÇϵµ·Ï ÇÑ´Ù. ½ÇÇè °á°ú ±âÁ¸ÀÇ ÃÖÀûÈµÈ »ó¿ë ¶óÀ̺귯¸®¿Í ºñ±³ÇÏ¿© Á¦¾ÈÇÏ´Â ¾Ë°í¸®ÁòÀÇ ¼º´ÉÀÌ ¿ì¼öÇÔÀ» º¸ÀδÙ. |
¿µ¹®³»¿ë (English Abstract) |
In this letter, we propose a multi-matrix programming model in GPU computation to deal with Hessenberg reduction problem. Hessenberg reduction problem is the most important step in eigenvalue problem, which is computationally expensive. Conventional method using GPU is inefficient in GPU resource usage when it deals with small matrix. The proposed method computes as many matrices as possible which maximally utilizes the GPU resources. Therefore, the proposed method achieves higher memory transfer rate between CPU and GPU, and massive parallel computation that is well fit for GPU computation. Experimental results show that the proposed outperforms the optimized commercial packages. |
Å°¿öµå(Keyword) |
GPU º´·Äó¸®
´ÙÁßÇà·Äó¸®
¼±Çü´ë¼ö
°íÀ¯Ä¡ ¹®Á¦
Hessenberg Reduction
GPU computation
multi-matrix programming
linear algebra
eigenvalue problem
Hessenberg Reduction
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|