Wicaksana, Narendra Hanif (2016) Implementasi Manajemen Transfer Rate Pada Proses Hadoop HDFS Berbasis SDN. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
5112100160-Undergraduate_Thesis.pdf - Accepted Version Download (1MB) | Preview |
Abstract
Hadoop dan Software Defined Networking merupakan dua teknologi yang sedang berkembang saat ini. Hadoop sebagai platform pengolahan big data dan Software Defined Networking arsitektur jaringan. Switch OpenFlow merupakan salah satu komponen utama pada arsitektur jaringan Software Defined Networking. Telah banyak dilakukan penelitian memanfaatkan Software Defined Networking dalam memanajemen jaringan, salah satunya adalah optimasi proses MapReduce pada klaster Hadoop. Infrastruktur arsitektur SDN juga merupakan salah satu permasalahan dikarenakan memerlukan switch yang mendukung OpenFlow. Saat ini terdapat metode untuk membuat switch OpenFlow dengan menggunakan komputer berbasis Linux. Pada Tugas Akhir akan dilakukan implementasi manajemen transfer rate pada proses penyimpanan data HDFS Hadoop. Manajemen transfer dilakukan dengan cara melakukan kategori lalu lintas jaringan pada klaster Hadoop dengan memanfaatkan fitur queue yang dimiliki oleh switch OpenFlow. Selain itu juga dibangun arsitektur Softwared Defined Networking menggunakan Raspberry Pi sebagai switch OpenFlow
=====================================================================================================
Hadoop and Software Defined Networking are emerging technologies at this time. Hadoop as a platform for big data processing and Software Defined Networking new network architecture. OpenFlow switch is one of the major components on the Software Defined Networking. A lot of research utilizing of the Software Defined Networking in management of the network, one of which is the optimization of the process of MapReduce in Hadoop clusters. SDN architecture infrastructure is also one of the problems due to the need to support OpenFlow switches. Currently there are methods to make OpenFlow switch to using Linux-based computers. In this the final project will be the implementing transfer rate management on Hadoop HDFS process. Management transfer is performed by categorized network traffic on a Hadoop cluster with queue. In this final project will build SDN infrastructure using Raspberry Pi as OpenFlow switches. The results of this implementation show decrease time on Hadoop HDFS with tranfer rate management.Data transfer for HDFS get more transfer rate than other traffic
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSIf 004.678 2 Wic i |
| Uncontrolled Keywords: | Software Defined Networking, OpenFlow, Hadoop, HDFS |
| Subjects: | Q Science > QA Mathematics > QA76.9.C55 Client/server computing |
| Divisions: | Faculty of Information Technology > Informatics Engineering > 55201-(S1) Undergraduate Thesis |
| Depositing User: | Yeni Anita Gonti |
| Date Deposited: | 29 Apr 2020 04:57 |
| Last Modified: | 29 Apr 2020 04:57 |
| URI: | http://repository.its.ac.id/id/eprint/75907 |
Actions (login required)
 |
View Item |