Implementation and evaluation of remote booting and installation for linux cluster systems리눅스 클러스터 시스템의 원격 부팅 설치에 대한 구현 및 성능 측정

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 486
  • Download : 0
In this thesis, we describe the implementation of a diskless cluster system with remote booting and installation and present the performance evaluation of the diskless cluster system. Some of advantages of the diskless cluster systems are easy maintenance and upgrade, and high reliability. For the latter, we identified the disk drive is the most error prone device in each PC box, and it also generates more heat which in turn may increase the fault possibility. A diskless node can have less system faults from mechanical failures, and it is possible to install a system under harsh environment with noise and vibration. For the former issue of maintenance and upgrade, we can easily upgrade or replace each node``s software by changing one copy at the front-end server. Diskless nodes bootstrap over the network using PXE daemon of RedHat Linux and Beoboot based on PXE (Pre-boot Execution Environment) technology. We also install the whole Linux operation system automatically on equally configured nodes of cluster system using remote installation. Our performance study using MPBench and NAS Parallel Benchmarks shows that diskless cluster system performs on par with the traditional cluster systems with local disks.
Advisors
Han, Dong-Sooresearcher한동수researcher
Description
한국정보통신대학원대학교 : 공학부,
Publisher
한국정보통신대학원대학교
Issue Date
2000
Identifier
392000/225023 / 000983935
Language
eng
Description

학위논문(석사) - 한국정보통신대학원대학교 : 공학부, 2000, [ vi, 56 p. ]

Keywords

Cluster; Remote Boot; PXE; Diskless; 리눅스; 클러스터; 원격 부팅; Linux

URI
http://hdl.handle.net/10203/54680
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=392000&flag=dissertation
Appears in Collection
School of Engineering-Theses_Master(공학부 석사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0