View Item 
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Mathematics and Natural Sciences
      • UT - Computer Science
      • View Item
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Mathematics and Natural Sciences
      • UT - Computer Science
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Pembangunan Laboratorium Praktikum Komputasi Paralel Berbiaya Rendah dengan Open MPI dan NFS.

      Thumbnail
      View/Open
      Cover (582.6Kb)
      Fulltext (827.1Kb)
      Date
      2023-08
      Author
      Thoriq, Muhammad Alpha
      Rahmawan, Hendra
      Akbar, Auriza Rahmad
      Metadata
      Show full item record
      Abstract
      High performance computing (HPC) merupakan salah satu bidang penelitian paling aktif di bidang ilmu komputer dan rekayasa (engineering). HPC dikembangkan untuk memecahkan sejumlah masalah yang kompleks dalam ilmu komputasi, seperti permodelan, analisis data dan desain produk. Saat ini pembelajaran kuliah dan praktikum Mata Kuliah Komputasi Paralel (KOM 613) masih menggunakan komputer tunggal untuk mempelajari pemrograman HPC tersebut, Padahal untuk pemrograman HPC yang bersifat distributed memory idealnya menggunakan komputer klaster. Penelitian ini bertujuan untuk membuat Laboratorium Komputer yang dapat mendukung pembelajaran pemrograman HPC. Laboratorium Komputer menggunakan beberapa komputer (multicomputer). Setiap komputer terbagi dalam beberapa peran, yaitu head node, file server dan compute node. Head node sebagai kontrol utama, compute node menjalankan program yang dibagikan, dan file server menjadi penyimpan file. Seluruh komputer disambungkan secara fisik melalui local area network. Metode penelitian yang dilakukan meliputi analisis, perancangan, implementasi dan evaluasi. Pengukuran kinerja komputasi dan komunikasi dilakukan terhadap komputer klaster dengan menggunakan perangkat lunak benchmark High Performance Linpack dan mpiBench, dan menghasilkan kinerja sebesar 21 GFLOPS dan 473 ms. Berdasarkan hasil benchmark, kemampuan komputer klaster dapat ditingkatkan dengan menggunakan komputer dan perangkat keras jaringan yang lebih baik dan sebaiknya juga digunakan job scheduler untuk memudahkan pengguna dalam menggunakan komputer klaster.
       
      High performance computing (HPC) is one of the most active research areas in computer science and engineering. HPC was developed to solve a number of complex problems in computational science, such as modeling, data analysis and product design. Currently lectures and practicum courses for Parallel Computing Course (KOM 613) still use a single computer to learn the HPC programming, whereas for HPC programming which is distributed memory in nature it is ideal to use a cluster computer. This study aims to create a Computer Laboratory that can support HPC programming learning. Computer Laboratory uses several computers (multicomputer). Each computer is divided into several roles, namely head node, file server and compute node. The head node is the main control, the compute node is running the shared program, and the file server is the file repository. All computers are physically connected via a local area network. The research method used includes analysis, design, implementation and evaluation. Computing and communication performance measurements were performed on cluster computers using the High Performance Linpack and mpiBench benchmark software, and resulted in a performance of 21 GFLOPS and 473 ms. Based on the benchmark results, the ability of cluster computers can be improved by using better computers and network hardware and a job scheduler should also be used to make it easier for users to use cluster computers.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/123097
      Collections
      • UT - Computer Science [2482]

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository
        

       

      Browse

      All of IPB RepositoryCollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

      My Account

      Login

      Application

      google store

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository