請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/19782完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 廖世偉 | |
| dc.contributor.author | Jian-Jang Shen | en |
| dc.contributor.author | 沈建志 | zh_TW |
| dc.date.accessioned | 2021-06-08T02:18:51Z | - |
| dc.date.copyright | 2015-09-17 | |
| dc.date.issued | 2015 | |
| dc.date.submitted | 2015-08-24 | |
| dc.identifier.citation | [1] Hybrid Map task scheduling onto CPUs and GPUs
[2] HadoopCL MapReduce on Distributed Heterogeneous Platforms Through Seamless Integration of Hadoop and OpenCL [3] “Aparapi,” http://code.google.com/p/aparapi/. [4] “Apache Hadoop,” http://hadoop.apache.org/. [5] J. Dean and S. Ghemawat, “MapReduce: simplified data processing on large clusters,” Communications of the ACM, vol. 51, no. 1, pp. 107–113, 2008 [6] “OpenCL: The open standard for parallel programming of heterogeneous systems,” https: //www.khronos.org/opencl/. | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/19782 | - |
| dc.description.abstract | Apache Hadoop近年來的蓬勃發展廣泛用在巨量資料的應用上, 但 Hadoop 系統框架在CPU上所遇到的效能瓶頸時常為人所詬病, 如果要將 Hadoop 應用程 式移植到圖形處理器上達到效能提升的目的, 程序員必須要花很多額外的心力。 在這篇理論當中我們將利用Aparapi 程式庫來做到將 Hadoop 應用程式移植到圖 形處理器上執行, 並且探討 Hadoop YARN 框架在異質性平台上的資源管理。 | zh_TW |
| dc.description.abstract | Apache Hadoop allows developers for the distributed processing of large data sets across clusters of computers using simple programming models. The booming of
Apache Hadoop solves many kinds of big data problems, and it is very suitable for parallel processing. But the poor performance of Hadoop applications due to the bottlenecks of computing is always reviled. Our research will proposed a framework which combines Haddop YARN and GPU, porting Aparapi libraries into YARN system for computing resources management in heterogeneous platforms. Extended the Application Master, which is a core component in YARN architecture to act as a role of resources request decision maker based on our scheduling algorithm. Besides, we adopt a preemptive, locality-aware task scheduling mechanism to fairly share CPU AND GPU resources. In the experiments, we show the overall speedup of an application, and analyze the effects to performance. | en |
| dc.description.provenance | Made available in DSpace on 2021-06-08T02:18:51Z (GMT). No. of bitstreams: 1 ntu-104-R02922093-1.pdf: 1059477 bytes, checksum: 2703777260995dec673c6f127edc8f01 (MD5) Previous issue date: 2015 | en |
| dc.description.tableofcontents | 口試委員會審定書.................................................................................................. # 中文摘要 .................................................................................................................... i
ABSTRACT ............................................................................................................... ii CONTENTS .............................................................................................................. iii LIST OF FIGURES ................................................................................................... iv Chapter 1 Introduction ............................................................................................ 1 Chapter 2 Background study................................................................................... 3 2.1 Hadoop YARN ............................................................................................... 3 2.2 Hadoop MapReduce ....................................................................................... 5 2.3 OpenCL and Aparapi ...................................................................................... 6 Chapter 3 Problem description and Related Works ............................................. 8 3.1 Problem description ........................................................................................ 8 3.2 Related Works ................................................................................................ 9 Chapter 4 Design and Implementation................................................................. 11 4.1 Heterogeneous Mapper ................................................................................. 11 4.2 Extended Application Master ....................................................................... 13 4.3 Preemptive Locality-aware scheduling......................................................... 16 Chapter 5 Experimental Results ........................................................................... 19 5.1 Experiment Environment.............................................................................. 19 5.2 Benchmarks .................................................................................................. 19 5.3 Execution Time of Jobs ................................................................................ 20 Chapter 6 Conclusion and Future Work.............................................................. 22 6.1 Conclusions................................................................................................... 22 6.2 Future Work.................................................................................................. 23 REFERENCE ........................................................................................................... 25 | |
| dc.language.iso | en | |
| dc.title | Hadoop 於異質平台之資源管理系統 | zh_TW |
| dc.title | Extending Resource Management System based on
Heterogeneous Hadoop Yarn Platform | en |
| dc.type | Thesis | |
| dc.date.schoolyear | 103-2 | |
| dc.description.degree | 碩士 | |
| dc.contributor.oralexamcommittee | 黃維中,涂嘉恆 | |
| dc.subject.keyword | 巨量資料,異質性分散式系統平台,圖形處理, | zh_TW |
| dc.subject.keyword | Hadoop,MapReduce,Yarn,OpenCL,Aparapi, | en |
| dc.relation.page | 25 | |
| dc.rights.note | 未授權 | |
| dc.date.accepted | 2015-08-24 | |
| dc.contributor.author-college | 電機資訊學院 | zh_TW |
| dc.contributor.author-dept | 資訊工程學研究所 | zh_TW |
| 顯示於系所單位： | 資訊工程學系 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-104-1.pdf 未授權公開取用 | 1.03 MB | Adobe PDF |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。