東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

GPU Computing in Statistics and R So...

Li, Yuan.

FindBook

Google Book

Amazon

博客來

GPU Computing in Statistics and R Solution.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	GPU Computing in Statistics and R Solution./
作者:	Li, Yuan.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2016,
面頁冊數:	99 p.
附註:	Source: Dissertation Abstracts International, Volume: 78-08(E), Section: B.
Contained By:	Dissertation Abstracts International78-08B(E).
標題:	Statistics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10583459
ISBN:	9781369621389

GPU Computing in Statistics and R Solution.
Li, Yuan.

GPU Computing in Statistics and R Solution. - Ann Arbor : ProQuest Dissertations & Theses, 2016 - 99 p.

Source: Dissertation Abstracts International, Volume: 78-08(E), Section: B.

Thesis (Ph.D.)--North Carolina State University, 2016.

Soon after the graphic processing unit (GPU) proving its potential in scientific research, the concept of general purpose GPU, commonly called GPGPU, has emerged and rapidly been adopted and integrated into fields that require intensive computation, such as digital image processing, bioinformatics, physics simulation, computational finance, et cetera. Statisticians, however, react slowly to this revolution even though there are a large number of statistical problems suitable for GPGPU application.

ISBN: 9781369621389Subjects--Topical Terms:

517247
Statistics.

GPU Computing in Statistics and R Solution.
LDR:03153nmm a2200301 4500 001 2128508
005 20180104132948.5
008 180830s2016 ||||||||||||||||| ||eng d
020 $a 9781369621389
035 $a (MiAaPQ)AAI10583459
035 $a AAI10583459
040 $a MiAaPQ $c MiAaPQ
100 1 $a Li, Yuan. $3 1682511
245 1 0 $a GPU Computing in Statistics and R Solution.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2016
300 $a 99 p.
500 $a Source: Dissertation Abstracts International, Volume: 78-08(E), Section: B.
500 $a Advisers: Hua Zhou; Brian Reich.
502 $a Thesis (Ph.D.)--North Carolina State University, 2016.
520 $a Soon after the graphic processing unit (GPU) proving its potential in scientific research, the concept of general purpose GPU, commonly called GPGPU, has emerged and rapidly been adopted and integrated into fields that require intensive computation, such as digital image processing, bioinformatics, physics simulation, computational finance, et cetera. Statisticians, however, react slowly to this revolution even though there are a large number of statistical problems suitable for GPGPU application.
520 $a In chapter one, we review the development of GPU, which has a distinctive design from the traditional central processing unit (CPU). Then we explore the evolution of GPGPU and its applications in scientific computation, especially in statistics. In chapter two, we summarize several classes of computationally challenging statistical methodologies and algorithms that naturally fit into the GPGPU framework. Furthermore, to help statisticians who are inexperienced in GPU programming to harness the power of GPGPU, we present a Compute Unified Device Architecture (CUDA) based GPU computing R package named RCUDA, which is a comprehensive computation environment for GPU accelerated linear algebra and random number generators. We show the technical details of the GPU implementation in R, which can serve as a tutorial for more research on this track. Finally, we provide some case studies by applying RCUDA to real statistical problems and compare the performances versus CPU.
520 $a Our contributions can be summarized as follow. First, we reviewed the evolution of GPGPU and its advances in statistical computing. Second, we identified several classes of statistical problems that match GPGPU framework. This motivates more statistical research that can benefit from GPU. Third, we evaluated existing GPU computing resources and provided an easy solution for statisticians: an R package that overcomes the shortcomings of currently available R GPU packages. Fourth, we presented a detailed outline of R GPU package building process, by following which statistician can readily write their own R GPU-accelerated packages. Fifth, we demonstrated the GPU computing advantage of performance over CPU by applying our solution to real statistical problems.
590 $a School code: 0155.
650 4 $a Statistics. $3 517247
690 $a 0463
710 2 $a North Carolina State University. $3 1018772
773 0 $t Dissertation Abstracts International $g 78-08B(E).
790 $a 0155
791 $a Ph.D.
792 $a 2016
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10583459